Oxford Studies in Political Philosophy Volume 4 [4] 9780198813972, 9780198813989

Citation preview

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Oxford Studies In Political Philosophy

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Oxford Studies in Political Philosophy Volume 4

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Edited by D av i d S o b e l, Pe t e r Va l l e n t y n e, a n d S t ev e n Wa l l

1

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

3 Great Clarendon Street, Oxford, ox2 6dp, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © the several contributors 2018 Te moral rights of the authors have been asserted First Edition published in 2018 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Library of Congress Cataloging in Publication Data Data available ISBN  978–0–19–881397–2 (Hbk.) ISBN  978–0–19–881398–9 (Pbk.) Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Preface

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Te chapters collected here were presented at the fourth annual Workshop for Oxford Studies in Political Philosophy in Barcelona, Spain. Local logistics were superbly organized by Serena Olsaretti, Paula Casal, and Andrew Williams at ICREA–Universitat Pompeu Fabra in Barcelona. We thank them, and the graduate students who helped them, for laying the foundation for a successful workshop. We would like to thank all the participants at the conference for making this a stimulating and fun event. Members of our editorial board refereed the chapters in this volume. As always, our great thanks go to them for quietly doing vital work needed to make OSPP successful. And of course thanks also to the authors of the papers collected here for bringing fascinating ideas to the table for us all to consider. Congratulations to Alex Zakaras, whose paper won the Sanders Prize in Political Philosophy and to Tomas Sinclair, whose paper was runner-up for the Prize. Travel by editor Steven Wall to the workshop at which the papers were presented was made possible through the support of a grant from the John Templeton Foundation. Te opinions expressed in this publication are those of the authors and do not necessarily refect the views of the John Templeton Foundation.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Contents List of Contributors

ix

Part I:  Legitimacy 1. Dethroning Democratic Legitimacy Zofa Stemplowska and Adam Swift 2. Te Power of Public Positions: Ofcial Roles in Kantian Legitimacy Tomas Sinclair 3. Institutional Legitimacy Allen Buchanan

3

28 53

Part II:  Egalitarianism

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4. ‘(Luck and Relational) Egalitarians of the World, Unite!’ Kasper Lippert-Rasmussen 5. Varieties of Relational Egalitarianism Zoltan Miklosi

81 110

Part III:  Liberty and Coercion 6. Moralizing Liberty Ralf M. Bader 7. May a Government Mandate More Comprehensive Health Insurance than Citizens Want for Temselves? Alex Voorhoeve 8. Complicity and Coercion: Toward an Ethics of Political Participation Alex Zakaras Index

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

141

167

192

219

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

List of Contributors

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Ralf M. Bader, University of Oxford Allen Buchanan, Duke University Kasper Lippert-Rasmussen, Aarhus University Zoltan Miklosi, Central European University Tomas Sinclair, University of Oxford Zofa Stemplowska, University of Oxford Adam Swift, University of Warwick Alex Voorhoeve, London School of Economics Alex Zakaras, University of Vermont

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

PA RT I

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

LEGITIMACY

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

1

Dethroning Democratic Legitimacy Zofa Stemplowska and Adam Swift

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. INTRODUCTION Democracies adopt many questionable policies. Some allow parents to opt out from vaccination for their children without medical consultation, even though mass vaccination is needed for herd immunity. Some devote more tax revenue to protecting afuent neighbourhoods from crime than they allocate to poor ones. Some make no serious legislative eforts to tackle pay gaps between citizens who difer only in gender or ethnicity. Some make no efort to tackle anti-immigrant hatred or demonization of single mothers in the press. We consider these to be cases of social injustice. Of course, no real state has democratic procedures that are entirely free of corruption and other democracy-undermining distortions. But suppose such a state existed, and suppose it decided to adopt any of these policies. Would the coercive enforcement of such policies be permissible? Or could those at the receiving end claim they were wronged? Suppose, further, that you had the power not only to resist the coercion but coercively to impose an alternative policy on others. Perhaps you have some control over police budgets and can secretly divert resources from some neighbourhoods to others. Perhaps you are the clerk to a legislature and can amend legislation without anybody noticing (Overland and Barry, 2011). Could it be permissible for you to do so? Te conventional view in the philosophical literature holds both that it would be impermissible for you to do so and that it would be permissible for the state to coerce you. In what follows we focus mainly on the latter: the permissibility of state coercion. Te conventional view is that democratic decisions are permissibly enforceable unless they are gravely unjust—perhaps when they blatantly violate basic human rights—but we assume that none of the cases we have outlined meets that standard. In cases of less serious social injustice, it is generally thought that their democratic provenance

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4

Zofa Stemplowska and Adam Swift

sufces to render the decisions permissibly enforceable.1 One may, of course, seek to change the decision by democratic methods. But the fact that a democratically approved policy is unjust does not undermine the permissibility of the state coercion. We question this conventional view. We agree that whether it is permissible to enforce a decision depends on both its provenance—the procedure by which it was made—and its content—the extent to which the decision delivers social justice. But we hold that social injustice does not need to be grave in order to render impermissible the enforcement of the decision with the democratic provenance. Rather, whether it is the democratic provenance or the just content of the decision that determines the permissibility of its  enforcement depends on whether, on a case by case basis, we have a greater reason in any given case to advance (i.e., to respect or promote) justice or legitimacy. Let us fx some terminology. When we say that a decision is or is not just we mean to evaluate its content. Decisions are just in this sense when people get what they are due (which, depending on the understanding of justice one favours, might mean various things: that they get what they are historically entitled to, what they need, what they deserve, what they would be entitled to in a society regulated by principles that would have been agreed to under certain idealized conditions, etc.). In addition to their content, decisions also have some provenance. We  assume that democratic procedures should be set up in polities to reach at least some coercive decisions. We refer to coercive decisions that, in appropriate contexts, have such democratic provenance as legitimate. Te outcomes of legitimate procedures are therefore legitimate only in the sense that they are the outcomes of the right procedures for making those decisions, which in the case we present here amounts to their being democratic. Our use of “legitimate” difers from that in some of the philosophical literature. Sometimes the question of whether a decision is legitimate is simply that of whether it is permissible to enforce the decision. (Sometimes the question of legitimacy is substituted for, or taken together with, the 1   Something is permissibly enforceable, let us say, when enforcing it (in line with any further requirements of proportionality and status of the enforcer) does not wrong anyone. A decision may be permissibly enforceable without there being any reason to enforce it: the fact that the decision was made in a given way may make it permissibly enforceable without there being a good reason to enforce it all the same. Tat said, in what follows we assume that there is a reason to enforce the permissibly enforceable decisions in question. We are grateful to David Estlund and Andrew Williams for discussion.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

5

further question of whether the decision has authority; i.e., whether those subject to coercion have a duty to comply with the coercion, or even the duty to obey.) Te issue of permissible enforceability is generally thought to depend on both the content and the provenance of the decision; the conventional view holds that democratic decisions are not permissibly enforceable in cases where their content is gravely unjust. We, by contrast, restrict the term “legitimate” to the issue of provenance, to the procedural aspect. Whether a coercive decision is legitimate, for us, depends entirely on how it was made; it is a further question whether it is in fact permissibly enforceable. One can thus accept that a procedure—and the decision it has produced—are perfectly legitimate, while insisting that the injustice of its content would make it impermissible to enforce that decision. Te dethroning of democratic legitimacy we aim to achieve in this chapter is a weakening of the contribution that legitimacy in our sense makes to legitimacy in the other sense. We have said that the justice of a decision is a matter of its content rather than its provenance. But of course ways of making decisions can themselves be evaluated as just or unjust, and so we could have framed our argument in terms of the familiar contrast between “procedural” and “substantive” justice. If citizens have a right to an equal say in the procedures by which their laws are made, then justice itself—procedural justice—requires that the  laws be made that way. Indeed, we might have followed Pettit (2015) in framing the “content/provenance” distinction as being between “social justice” and “political justice”: the latter, for him, coincides “with what is often described as the legitimacy as distinct from the justice—I would say, the social justice—of the structure” (11). For Pettit, then, what we are calling “legitimacy” is labelled, in the political context, “political justice”. We have no objection to analyses of legitimacy that invoke procedural (or political) justice. It may well be the justice of a procedure that confers legitimacy on its decisions. Tis still leaves conceptual space between the justice of a decision-making procedure and the justice of the content of its outcomes. And it raises the question of how the two kinds of justice relate to each another when it comes to assessing the normative status of particular decisions. Our dethroning of democratic legitimacy will unfold as follows. We shall frst—in §2—set out in greater detail the conventional view that we dispute. In §3, we outline our alternative, which we call the balancing view. Section 4 sets out our positive case for that view and our negative case against the conventional view. Section 5 considers further objections to the balancing view. We conclude with §6, which broadens the picture to explore the implications of our analysis for non-ideal circumstances.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

6

Zofa Stemplowska and Adam Swift 2.  THE CONVENTIONAL VIEW

According to the conventional view, a law or policy is permissibly enforceable if it was decided by a proper democratic procedure, even if the law or policy is somewhat unjust.2 When it is gravely unjust, however, it is not permissibly enforceable, even if it was decided by that same procedure. Te view, then, has two elements. First, it holds that democratic provenance can bestow permissible enforceability on a decision. Tis claim is widely accepted and has gained recent support from David Estlund (2008), Niko Kolodny (2014a and b) and Daniel Viehof (2014). Second, the conventional view holds that democratic provenance of the right type fails to ground permissible enforceability only if the content of the decision is gravely unjust.3 Here is Rawls (1993: 428), setting out what we take to be the conventional view:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

A legitimate procedure gives rise to legitimate laws and policies made in accordance with it; and legitimate procedures may be customary, long established, and accepted as such. Neither the procedures nor the laws need be just by a strict standard of justice, even if, what is also true, they cannot be too gravely unjust.4 At some point, the injustice of the outcomes of a legitimate democratic procedure corrupts its legitimacy, and so will the injustice of the political constitution itself. But before this point is reached, the outcomes of a legitimate procedure are legitimate whatever they are. Tis gives us purely procedural democratic legitimacy and distinguishes it from justice, even granting that justice is not specifed procedurally. Legitimacy allows an undetermined range of injustice that justice might not permit.

Tis passage is somewhat obscure. When he says “At some point, the injustice of the outcomes of a legitimate democratic procedure corrupts its 2   Te conventional view is also often claimed to establish authority, or is ambiguous between the two claims; we put this aside. 3  We put aside some complications, such as that in circumstances of emergency permissible enforceability could be grounded in other ways. In addition to Rawls, this view is endorsed by Tomas Christiano (2004,  2008), Philip Pettit (2015), Jonathan Quong (2010), Tomas Sinclair (chapter 2 of this book [and private correspondence]), Laura Valentini (2012), and, as we suggest in §4.5, possibly Daniel Viehof (2014). More generally, we think that most Rawlsians are supporters of the view and possibly also most Kantians, though the latter may deny that justice and democracy can come apart. Estlund (2008: 111) is carefully non-committal on the limits of democracy’s ability to permit enforcement. 4   Notice that Rawls talks about the case when procedures or laws are “too gravely unjust” (our italics). For him, apparently, there can be grave injustices that are not “too grave” and which do not thereby corrupt the legitimacy of the procedure. We will ignore this complication, henceforward taking him to hold that “grave” injustice is grave enough to “corrupt” legitimacy. An alternative reading of the phrase is also possible, according to which Rawls holds an implicit balancing view.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

7

legitimacy,” he seems to suggest that the gravity of injustice in an outcome deprives not only that outcome but also the procedure itself of legitimacy. Perhaps, though, Rawls means that the legitimacy that is “corrupted” is only that of the gravely unjust outcome: i.e., in our terminology, the outcome is  not permissibly enforceable. (Tis would be the natural reading if he had  written “At some point, the injustice of the outcome of a legitimate democratic procedure corrupts its legitimacy” (our emphasis).) On this interpretation, a legitimate procedure may yield an outcome the legitimacy of which is corrupted by its gravely unjust content. In that case the decision lacks legitimacy in the sense that it is not permissibly enforceable, even though it issued from a legitimate procedure. We interpret Rawls’s position here as one according to which avoidance of grave injustice5 is necessary and sufcient for permissible enforceability. However, if a decision is not gravely unjust, then the legitimate provenance of the decision is sufcient for permissible enforceability.6 Before presenting and defending our alternative view, let us acknowledge— in order to put aside—that there will be many points of overlap between decisions that are democratically legitimate and those that are socially just. We are happy to accept that what is democratically legitimate might be a complex matter: a procedure might qualify because of its purely procedural properties or because, say, of its tendency to produce given—perhaps just— outcomes (Estlund 2008). We do not set out the conditions for democratic legitimacy but we accept that it can make distributive demands and that for a procedure to count as democratic it may have explicitly to rule out, or in other ways make impossible, certain outcomes.7 We are also happy to accept that what social justice requires may itself include procedural elements. Political rights may be requirements of social justice and, further, perhaps certain ways in which citizens meet one another’s justice claims qualify as “social justice” only if they are put, and kept, in place by democratic procedures (Pettit 2015). All this can be granted while retaining the crucial point that the demands of social justice, fully and properly understood, may difer from, and confict with, the demands that result even from perfectly legitimate democratic procedures.   Assuming the absence of less gravely unjust alternatives.   Whether or not it is necessary is left unclear, given Rawls’s views on global justice, which we put aside. We dispute both the sufciency and the necessity claims. 7   Tis can be done by stipulating the conditions for a genuinely democratic process (e.g. that all participants see their co-citizens as their moral equals). Such a move might also ensure that we could always locate some value in the democratic process. We are sympathetic to such a stipulation but insist that, while it may rule out some substantively unjust outcomes from being decided by the demos, it cannot plausibly rule out all of them. 5 6

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

8

Zofa Stemplowska and Adam Swift

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.  THE BALANCING VIEW Our alternative proposal is that democratic legitimacy and social justice both ground pro tanto claims to a decision’s permissible enforceability; these claims have to be weighed or otherwise taken into account on a case-by-case basis. We say “weighed or otherwise taken into account,” since we are not committed to a simple view of weighing (though we refer to “weighing” throughout, for simplicity). Perhaps in some contexts one consideration can pre-empt or otherwise eliminate the need to consider the other: what matters is whether, taking into account what can be achieved, there is greater reason to advance legitimacy or justice in any given case.8 Te two ideals at stake invoke diferent normative considerations; they embody and advance (that is, respect or promote) diferent values. Even with grave injustice of the table, we see no reason to regard the considerations invoked by the former as invariably weightier or more forceful than those invoked by the latter. In terms of our proposed analysis, the conventional view can be regarded as a claim about a particular case: it holds that only injustices of a particular level of severity—those deemed “grave”—are unjust enough to outweigh or otherwise block the pro tanto permissible enforceability conferred on them in virtue of their having been produced by procedures that realize or promote legitimacy values to a certain degree. Tat degree need not be 100%—recall Rawls: “A legitimate procedure gives rise to legitimate laws and policies made in accordance with it; and legitimate procedures may be customary, long established, and accepted as such. Neither the procedures nor the laws need be just by a strict standard of justice, even if, what is also true, they cannot be too gravely unjust” (our emphasis). For us, casting the conventional view in this light—bringing out the scalarity in both components of the judgement about proper enforceability—reveals an arbitrariness in judging only “grave” injustices to lack permissible enforceability. In our view, less-than-grave injustice may also lack that property even if it enjoys democratic provenance. Tus not only can there be no general presumption that only grave injustice prevents democratic provenance from conferring permissible enforceability on a decision, it is also the case that the need to avoid even less-than-grave injustice sometimes grounds the permissible enforceability of outcomes rejected by the democratic process. We think 8   It might be doubted that democratic procedures always give even a pro tanto reason for its being permissible to enforce a decision. Unlike our balancing view, the conventional view can deny that the fact that a gravely unjust outcome was the outcome of an otherwise ideal procedure does anything to support the view that that outcome is permissibly enforceable. We fnd this position implausible: if the value of the procedure does not depend solely on the outcome it delivers, it is unclear why the mere fact that an outcome is unjust, even if gravely so, deprives the procedure of all value.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Dethroning Democratic Legitimacy

9

that this relationship holds whether we are interested in what is right and wrong relative to the facts or relative to the evidence available.9 More schematically, the balancing view holds that the avoidance of (more) grave injustice is necessary and sufcient for permissible enforceability: it is permissible to enforce the outcome that avoids (more) grave injustice than the alternative, and it is impermissible to enforce any alternative. So the avoidance of (more) grave illegitimacy is neither necessary nor sufcient for permissible enforceability. But we also deny that full justice is necessary for  permissible enforceability: sometimes the imperfectly just legitimate decision will be permissibly enforceable.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4.  DEFENDING THE BALANCING VIEW Our negative case for our balancing view is to point out the problems with the conventional view, which is either vague or implausible. It is vague if it amounts to the view that anything that outweighs legitimacy should count as grave injustice. If so, then the conventional view may even collapse into the balancing view on the quiet. If, on the other hand, the view invokes some independent account of what counts as grave injustice then it is implausible. Tere is no good reason to think that democratic legitimacy should generally outweigh social justice when the two values confict, even with grave injustice of the table. Te conventional view gains credibility when the value of legitimacy— democratic procedures—is not properly analysed. Once it is (and our analysis is mostly ofered in §§4.3–4.5) we see that the credence-giving value of democratic procedures is not unique to them, while the remaining value or values—variously articulated as non-subordination or treating people as equals, or with equal respect—cannot ground the conventional view. But let us begin with our more positive case. We have already pointed out that there are values on both sides of the question—democratic legitimacy and social justice. While we frequently hear why democracy might be valuable, the value of justice is often taken for granted, making it less clear what is at stake in decisions’ being just or unjust. In what follows we attempt to say a bit more about the latter.

4.1  Justice Judgements can Go beyond Mere Recommendations What is at stake in judgements of social justice? One way to make sense of the conventional view is to see judgements of social justice (except those 9   On “fact-relativity,” and distinctions between it and “belief-relativity” and “evidencerelativity,” see Parft (2011: 150–62).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

10

Zofa Stemplowska and Adam Swift

of  grave injustice) as necessarily conditional: “Tis is what social justice demands but I do not regard that fact as making a claim about its permissible enforceability unless further conditions are met. Te correctness of my judgement about social justice is not in question. But it matters also what others think.” Judgements about justice are rather, as Pettit (2015: 16) nicely puts it, “recommendations”; their claim to permissible enforceability is provisional on further conditions—conditions of legitimate decision-making (or political justice)—being in place (16). Similarly Valentini (2012: 600–1), who agrees that in ideal circumstances the permissible enforceability of judgements of justice is “conditional on democratic approval,” sees them as having “fragile normative status.” All this is compatible with regarding judgements of social justice as correct or not; true or false. A legitimate decision may thus be mistaken about what arrangements people are owed as a matter of justice, and acting on that mistaken decision means failing to enforce the arrangements that citizens would, in fact, be right to demand as a matter of justice. But the fact that a judgement is correct about what it would be right to demand does not change the fact that the permissible enforceability of the demand is conditional on its being democratically decided.10 Tis is a coherent view but it comes at a cost. To see the cost consider the following three cases, in which citizens vote on (a) whether to paint the public lamp posts purple or magenta, (b) whether to enact legislation to protect an area of outstanding natural beauty, and (c) whether to devote some resources to tackling a less than gravely unjust gender pay gap. When, as part of what I accept to be a legitimate decision procedure, I vote for magenta because it’s a colour that I like, there is no difculty in my accepting that what is permissibly enforceable should depend on, for example, what is the majority view. In this case, let us suppose, the outcome derives its permissible enforceability entirely from its being the outcome of the procedure, because there is no procedure-independent reason to enforce one colour over the other. If so, my vote cannot even represent an independent judgement of what colour it would be permissible to enforce. When I vote on whether to protect an area of outstanding natural beauty I may believe that there is a procedure-independent reason to enact such coercive legislation—this area of natural beauty ought to be protected no matter what everyone thinks. But I may also think that the decision is not 10   Tat said, the view that judgements of justice are mere recommendations could not hold that a vote for X means “X ought to be enforced if and only if it is democratically decided.” Tat view is satisfed even if nothing is democratically decided, since then the conditional has not been violated. A vote surely means more than “don’t enforce this unless it is democratically passed.” We are grateful to David Estlund for this point as well as for making us clarify our argument in this section.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

11

one of justice and in such cases only the procedure can confer permissible enforceability on the decision. But we can also imagine cases where in voting for a given outcome, I  am making an independent judgement about what just outcome it would be permissible to enforce. Put diferently, we can imagine cases where I am making a judgement about what justice demands about what should be decided and enforced. In such cases if I lose, and the fact that I lose does not lead me to change my mind, I believe that we have made a mistake somewhere along the line when we enforce the outcome I voted against. At this point we can consider two possibilities. First, we might hold that the mistake occurred at the point of decision-making but, once the wrong decision was made, it became permissible to enforce the wrong decision: only the procedure—and not the correctness of the judgment—confers permissible enforceability on the decision. Tis is compatible with seeing judgements of justice as provisional recommendations. Second, we might hold that the mistake occurred both at the point of decision-making and at the point of enforcement of the (wrong) decision. We could accept that a democratic body should decide X and enforce what it decided, without being committed to believing that the body should enforce what is decided if it did not decide X. We could further hold that X is so important that it should be enforced even if X was not decided by the democratic body. If so, when voting, we could be inputting our judgement about X precisely as a judgement about what permissibly enforceable claims people have, and the judgement about what those claims are can be made on grounds independent of the collective procedure to which they are inputting their judgement. On this view, the question others are getting the wrong answer to when they do not vote for X is precisely what it would be permissible to enforce on grounds independent of the procedure for collectively deciding that question, rather than the question of what should be provisionally recommended for enforcement. Te “recommendation” view rules out the very possibility that individuals’ inputs to democratic procedures could be understood by them (and others) as judgements about citizens’ claims against one another that are permissibly enforceable on procedure-independent grounds. Our claim is not that we must, on pain of conceptual incoherence, accept that some judgements of social justice take this unconditional form. Rather, we hold that seeing some claims of justice this way confers special importance on them, or rather, in our view, captures the importance they already have. Tey are claims that deny others the normative power to block their enforcement. Tese claims are so important, in other words, that others lose the standing objection to coercion they enjoy in other cases.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

12

Zofa Stemplowska and Adam Swift

Proponents of the “recommendation” view could insist—and we will say more to develop this argument in §4.3—that their view captures the fact of  disagreement about what these claims are, the need for a method for resolving it, and the way in which that method can bestow normative force on its outcomes. But our “balancing” view does not deny any of that: it readily acknowledges that the values promoted or respected by disagreementresolving procedures can confer considerable weight on the outcomes they produce. We simply add to the picture in emphasizing the fact that the object of disagreement is citizens’ claims of justice against one another, and we insist that getting that right must be weighed against whatever value attaches to outcomes in virtue of their issuing from legitimate procedures. Te proponents of the “recommendation” view could also claim that, when voting, we should always ask only what would be permissibly enforceable if it were decided by a legitimate procedure for making such a decision— after all why have a vote at all? But there could be many reasons to see the  democratic procedures as a desirable way of making decisions about enforcement. For example, we could care about the stability that having democratic procedures is more likely to bring about. But, fnally, note that the conventional view itself cannot hold that judgements of justice are always merely recommendations if it is to allow that it may be permissible to go against the democratic decision when grave injustice is at stake. Tis brings into focus the puzzling nature of the conventional view: it allows that considerations of grave injustice can block the ability of legitimate procedures to confer permissible enforceability on democratic decisions but denies this power to all other justice considerations. But unless all grave injustice, and only grave injustice, undermines the democratic nature of democratic procedures—which is implausible—it is unclear how this trick can be pulled of. We will argue that it cannot.

4.2  Justice Judgements can be Inclusive First, however, let us further defend the status of individuals’ judgements about social justice by pointing out that they need not result from parochial, sectarian, or blinkered reasoning. Democratic procedures can bolster their credentials for conferring permissible enforceability on their decisions by appealing to the fact that they combine the preferences or judgements of diferent individuals into a collective decision. At its most general, perhaps, the idea is that, to be legitimate, decisions must be, in the right way, inclusive. Tey must satisfy conditions that in one way or another include the preferences or judgements of those subject to the decision. Our response is to point out that individuals’ judgements about justice can also be inclusive in important ways. What is distinctive about democratic procedures is what

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

13

they include: the actual votes of actual citizens. But including and responding appropriately to the views and preferences of others may well be important to the correct identifcation of what social justice requires as that identifcation is attempted by each individual. To avoid confusion, let us call such procedures undertaken by an individual, in contrast to the democratic ones, processes.11 Such a process may be needed because for one to have any confdence that one’s input to the legitimate decision-making procedure correctly identifes social justice, one must have, and be able to ofer to others, certain kinds of reasoning or considerations as relevant. As an individual making one’s own judgement about what social justice requires, it is likely to be helpful to engage with others, consider their reasoning, see if their reasoning undermines one’s justifcations, and so on. But one must also satisfy an internal justifcatory process, have subjected it to critical refection, and so on (Goodin, 2000). Tinking about what people would or would not agree to, or what could be justifed to them under certain idealized conditions, or would emerge from a certain kind of procedure, were it undertaken, may be part of that process. In some senses the individual’s own process will be “collective” or “inclusive” in that it is likely to incorporate, or at least in some ways respond to, the views of others. But this concerns the individual forming her judgement about the correct input to the legitimating procedure. It does not concern the sum of those individual inputs being combined into a collective decision through an actual procedure for combining them.12 Crucially, then, it is possible to distinguish between (i) processes to which we as individuals (albeit perhaps through deliberation with others) must subject candidate principles in order to assess them correctly as principles of justice and (ii) procedures that individuals’ judgements about principles of justice must go through in order for that combined or collective judgement to be legitimate. Any contractualist approach to questions of justice will involve a variant of the idea that one arrives at correct answers to questions of justice by thinking about what would be agreed to by people, specifed a certain way, in circumstances specifed a certain way. As Estlund (2008: 16) says: “Contractualism . . . is a family of views that understand justice or rightness as constituted by facts about what would be agreed to in a certain imaginary collective choice situation.” Rawls’s original position is the paradigm case here, of course. Tis is a claim about (i). It does not require   We thank David Estlund for pressing us to clarify this.   For a similar suggestion, presented as a criticism of Forst (2014), see Caney’s (2014: 156) insistence on the importance of distinguishing between “the justifcation required for political legitimacy (political justifcation) and the justifcation required for a view to be philosophically correct (philosophical justifcation)” (his emphases). 11 12

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

14

Zofa Stemplowska and Adam Swift

conceiving justice as the outcome of any actual agreement between—or any “combined” or “inclusive” or “collective” judgement based on the input of—diferent people, but it allows justice to be inclusive in other important ways.13 Tis is not to deny the possibility that actual inclusive procedures can sometimes have epistemic advantages over the processes that individuals alone can undertake. But non-democratic processes can make use of at least some of the benefts (while also avoiding some of the pitfalls) of “inclusive” decision-making.

4.3  Reasonable Disagreement does not Ground the Conventional View

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Perhaps, however, we underplay the importance of reasonable disagreement. Te presence of such disagreement may mean that we owe respect to others’ views (and we discuss this in §4.4) but it may also suggest that our own views—even those that went through the process described in §4.2—are less worthy of respect. Te suggestion here is that the fact of disagreement— and in particular reasonable disagreement—should reduce confdence in one’s judgement to a level below that which, absent grave injustice, would justify acting on it in the face of a democratically legitimate decision. Put diferently, the proponents of the conventional view can appeal to the fact of reasonable disagreement to ground the conclusion that, whatever may in fact be the just decision, evidence will always warrant the conclusion that, except in cases of grave injustice, the legitimate decision is the permissibly enforceable one. 13   Amartya Sen (2009: 326) risks running the two modes—(i) and (ii)—together: “If the demands of justice can only be assessed with the help of public reasoning, and if public reasoning is constitutively related to the idea of democracy, then there is an intimate connection between justice and democracy, with shared discursive features.” Te thought here is that the way one goes about identifying justice, if one is doing it right, involves a kind of reasoning (which Sen calls “public reasoning”) that will itself play a role in any properly democratic way of making a collective decision. Even if that is true, we are not sure that Sen always holds on to the distinction between the two distinct elements in the picture. His emphasis on “social choice theory,” which is “deeply concerned with the rational basis of social judgments and public decisions” (95, our emphases), seems to involve a slide from the frst to the second. An individual’s judgement can be “social” in being about social matters and it can be “public” in the sense of being about public matters (or publicly available, or based on public reasoning). But talk of “public decisions,” and indeed the very idea of “social choice,” takes us across the analytical divide into the (for us) entirely distinct business of combining individuals’ judgements into a collective decision. Te legitimacy of such a decision may well depend on its democratic provenance (including perhaps considerations concerning the kinds of reasoning that citizens may properly engage in when deciding on public matters), but it remains a distinct question whether that decision is right about social justice.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

15

We do not need to deny that acknowledging reasonable disagreement about a judgement could appropriately lessen one’s confdence in it.14 We could concede that one’s commitment to one’s views should be reduced by the recognition that reasonable people can and do disagree, or that other views pass a “reasonableness” threshold. But this line of thought must have limits. One problem is that it becomes hard simultaneously to assert that such confdence is justifed in cases when the injustice at stake is grave. Tat is, if one’s confdence should track the views of others then it is difcult to see why we should make exceptions in the case of grave injustices. Unless grave injustice stood for “obvious injustice,” there is no good reason why one’s confdence would track gravity. If denying economic migrants settlement visas is unjust it is unjust, or if denying free access to life-saving drugs in order to use the resources for quicker epidural access on maternity wards is unjust, it may well be gravely unjust, but we may not be certain that it is unjust. By contrast, if stealing a pencil is unjust, it is not gravely unjust, but we may be justifed in confdently believing it unjust—and “obviously” so. Moreover, the thought that democratic decisions outweigh just ones because of lack of confdence in justice judgements can be met by analogous appeals to lack of confdence in judgements about legitimacy. If disagreement about justice should lead people to lack the confdence in their beliefs about what justice requires, then why shouldn’t disagreement about legitimacy lead them to lack confdence in their beliefs about what legitimacy requires? To be sure, in some instances people will disagree merely on what social justice requires rather than on what is the right procedure for deciding the matter (though note that on many accounts of reasonable disagreement what matters is not whether the disagreement exists but whether it could arise15). But in other cases, we should expect people to be less confdent about what would be legitimate than about what would be just. Te view that, society’s resources permitting, children with weak immune systems have the right that mass vaccinations be compulsory for the sake of herd immunity, or that social justice requires that proper resources be spent on protecting poor citizens from crime, or that a 5% pay gap between citizens who difer only by gender or ethnicity is unjust, can be held with greater confdence than a view about what would be the legitimate procedure for deciding those questions. Tese doubts can take the form of disagreement over the specifc democratic procedure that ought to be adopted or, more radically, whether a given issue should be decided by democratic procedures at all. 14   For discussion of the problem see D. Christensen (2009). We are grateful to David Estlund for the reference. 15   Jonathan Quong (2010).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

16

Zofa Stemplowska and Adam Swift

4.4  Equal Respect does not Ground the Conventional View Epistemic and credence considerations are, of course, not the only ones that may be invoked to motivate the need for inclusive democratic procedures. Some think that democratic procedures are needed to respect people’s general presumptive right or standing objection against coercion.16 Such a right or objection is notoriously difcult to establish. It is not clear that there is any wrong in coercing someone not to do evil. Nonetheless, in the words of Kolodny (2014b: 316):

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

. . . [I]t is a common view, perhaps the dominant view, in political philosophy that even the fact that the decision is substantively ideal is not enough to make it legitimate [permissible to implement]. Tere is, according to this view, some Further Objection to political subjection as such, even to substantively ideal decisions . . . For all the insistence on a Further Condition, however, it is elusive what Further Objection it is supposed to answer. When clearly stated, many of the candidates for this Further Objection seem weightless or confused. If a residual worry remains, as I argued elsewhere, it is that in being subjected to political decisions, we are somehow subordinated to other people: namely, those who make the decisions.

In what follows we do not pursue the objection from coercion but consider two more promising attempts to capture similar worries: the appeal to the need for equal respect and, in §4.5, the objection to subordination acknowledged by Kolodny. Advocates of the conventional view might argue that our view fails to recognize that disagreement is not simply an unfortunate fact—something for which, for practical purposes, we need an efective solution—or grounds for reducing one’s confdence in one’s own judgement. Te disagreement warrants a particular kind of response, one that treats each citizen as equally a source of authority. Unlike the individual judgements that are their inputs, the outcomes of democratic procedures may permissibly be enforced—even where they are mistaken—because those procedures are the way of responding to disagreement that accords citizens equal respect. Of course, the conventional view holds that the outcomes of otherwise legitimate procedures lack permissible enforceability when those outcomes are gravely unjust. So that view does not always prefer the “equally respectful” procedure to the just outcome; it allows content-reasons to outweigh procedure-reasons when the former reach a particular level of seriousness. In addition, many who ground the normative force of democratic decisions 16   See Fabienne Peter (2016) for discussion of how procedures could ground permissible enforceability because they deliver “sufciently justifed beliefs about normative authority” and/or “adjudicate between conficting wills.”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

17

in citizens’ disagreements with respect to matters of content confne their concern to that subset of disagreement which is deemed “reasonable.” What demands this particular moral response—this particular manifestation of “equal respect”—is not the fact of disagreement as such, it is the fact that some of the disagreement is reasonable. Tese two points might be connected: one might think that any citizen who would vote for a gravely unjust outcome is unreasonable, or vice versa, but neither version of the connection is necessary. Even with these two qualifcations in mind, we do not fnd this challenge persuasive. In the frst place, it is a mistake to think of our view as altogether neglecting the fact of (reasonable) disagreement. As we have explained, the  balancing view can acknowledge that individual judgements about the  content of social justice may themselves depend for their validity on deliberative procedures that give proper weight to the judgements of others. Te assessment of a principle as a correct principle of justice, as that assessment is conducted by each individual, may well involve consideration of whether it meets certain justifcatory criteria, whether it can ofer the right kind of response to those who disagree, or who reasonably disagree, and so on. So the diference between us and our opponent is not about whether disagreement demands proper acknowledgement and normatively appropriate response, it is about what constitutes that acknowledgement and response. Second, the appeal to the notion of “equal respect” fails to ground the conventional view. While allowing that justice is also a way of articulating that idea, Valentini (2012: 598) insists that it is the procedural ideal of legitimacy that is its relevant articulation in contexts where there is disagreement about justice: “what we typically call ‘justice’ articulates the demands of equal respect under the so-called circumstances of justice,” while “what we call ‘legitimacy’ articulates the demands of equal  respect under the circumstances of ‘political justice.’ ” Christiano’s justifcation of democracy as a legitimating procedure involves a similar claim. According to him (2004: 271, 273): “Each citizen has fundamental interests in being able to see that he is being treated as an equal in a society where there is signifcant disagreement about justice . . . Tere is a deep interest in having one’s moral standing among one’s fellows clearly recognized and afrmed.” Like Valentini, too, he frames this as a claim about justice: democratic decision-making is a publicly just and fair way of making collective decisions in the light of conficts of interests and disagreements about shared aspects of social life. Citizens who skirt democratically made law act contrary to the right of all citizens to have an equal say in making laws when there is substantial and informed disagreement. Tose who refuse to pay taxes or who refuse to respect property laws on the grounds that these are unjust are simply afrming a superior

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

18

Zofa Stemplowska and Adam Swift

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

right to that of others in determining how the shared aspects of social life ought to be arranged. Tus, they act unjustly.  (ibid.: 286)

Let us grant that there is a kind of injustice when a decision about social justice is enforced that difers from that which would have emerged from a  procedure that equally respected people’s judgement, or that publicly treated them as equal members of society. Tere is also a kind of injustice when a decision about social justice is enforced that fails to treat people equally with respect to those properties or capacities that would have been treated equally had the correct decision been enforced. Christiano (ibid.: 286) insists that “Te interests involved in being publicly treated as an equal member of society are the preeminent interests a person has in social life . . .” (his italics). We accept that those interests are weighty. But the idea of “interests involved in being publicly treated as an equal member of society” does not deliver the conclusion that procedural interests take priority over all other interests (subject to the not-too-grave-injustice clause)—after all, being publicly treated as an equal could involve both. Or, if the phrase is mean to denote merely the procedural interests, the claim is unpersuasive. People’s (non-procedural) interests are equally respected when people get what they are owed as a matter of social justice, and these interests are also important. How should we compare these two diferent aspects of what might be involved in respecting people equally, or treating them justly? Is it worse to be denied (i) that procedure for deciding social arrangements which treats those subject to it with equal respect, or (ii) those social arrangements that would result from the correct identifcation of the implications of each person’s claim to equal respect? We might think about this by contrasting the diferent properties or capacities at stake. On the one hand are those properties that are equally respected by the democratic procedure. Examples here would be the capacity to make judgements about which distributions should be enforced, or having that capacity recognized and afrmed by one’s fellow citizens. On the other hand, are those properties equally respected when the correct, socially just, decision is enforced? For example, the correct decision about allocation of resources to health care or education can respect people’s capacities to lead autonomous lives: democratic publics can decide to devote disproportionate resources to end-of-life care, or to university education, to the relative neglect of health care and education for young children. We see no reason to suppose that the capacities respected by the democratic procedure are so much more important than the capacities respected by the socially just education or health care policy as to make only grave mistakes grounds for denying the permissible enforceability of collective misjudgements. When we refect on what we ourselves care about, we readily accept that it may

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

19

sometimes be more important to cater to our capacities for health than our capacities to decide on how much to devote to the health budget. It is, moreover, far from obvious that one treats someone with respect by “respecting” their mistaken judgements about justice. When people vote for what they believe to be the just decision, we may assume that at least part of what matters to them—even in their role as democratic citizens, and putting aside their interest in being subject to just rather than unjust laws— is indeed achieving a just outcome. If they could see that they were mistaken, then they would want their mistake corrected; after all, what they care about is justice, not having their mistaken view about justice given equal weight with other, correct, views. If they do not see that they are mistaken, then overriding their view may indeed lead them to feel disrespected, and certainly their capacity to identify the right answer about justice, on this occasion at least, will be regarded as inferior to that of others. But if what they are trying to achieve through their democratic input is indeed a just outcome, then it is not obvious that the kind of respect that matters is accorded by putting aside knowledge of the very thing they were themselves seeking.17 We do not deny that there can be weighty reasons to enforce democratic decisions even when they are mistaken. We can grant that there is indeed a  kind of disrespect involved in substituting one’s judgement for that of one’s fellow citizens. One might even conceive it as manifesting a kind of contempt for their views, and as importantly fracturing our relationship of  democratic citizenship with them—though even this can go hand in hand with acknowledging, indeed publicly acknowledging, that those who made the incorrect decisions are generally good judges. But there is also a kind of disrespect involved in failing to grant people what they are, in fact, owed as a matter of social justice. We may not be warranted in seeing those who fail to grant people what they are owed as intentionally disrespectful; after all, by assumption, all our disagreement may be reasonable. But the disrespect involved in failing to see the democratic process as bestowing permissible enforceability on the collective decision need not involve any troubling intentionality either: we may simply think that on some issues people were given too little time to decide.18

4.5  Avoiding Subordination does not Ground the Conventional View Do we take the disrespect involved in substituting one’s judgement for that of one’s fellow citizens too lightly? What is at stake, perhaps, is not merely 17

  We are grateful to Victor Tadros for this suggestion.   We are grateful to an anonymous referee for making us clarify this.

18

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

20

Zofa Stemplowska and Adam Swift

which capacities are more worthy of respect but the fact that, as Niko Kolodny and, separately, Daniel Viehof have recently suggested, democratic procedures are necessary to sustain relations of social equality. Social equality, roughly put, means that people are neither socially superior nor inferior to one another. More precisely, according to Kolodny (2014b), social equality is missing—there is “subordination”—when some people have greater relative opportunity for power or de facto authority over others and are not “resolutely disposed to refrain from exercising it” (295).19 For Viehof (2014), social equality requires that “none of us has more of a say than any other” about the relationship at stake, e.g. marriage or citizenship in a democracy (340). People must “set aside, and not act on, unequal power advantages in shaping [their] . . . interactions and the norms and expectations governing them” (352); there must be “rough equality of power over the interactions that make up the relationship” (355), and the parties must be “committed to having equal power” (356). Te two accounts difer, but both Kolodny and Viehof argue that it is the role that democratic procedures play in sustaining relations of social equality that explains why those procedures can be the source of permissible enforceability.20 Social equality itself requires some social justice21 which means that some socially unjust decisions would advance neither social justice nor legitimacy. But both authors accept that social equality and democratic decisions, on the one hand, and social justice, on the other, can confict.22 Tat said, Kolodny’s account of the value of democratic procedures does not bolster the conventional view. He is explicit that his account is intended merely to capture the reason that democratic procedures may ofer for the permissible enforceability of their decisions even when those decisions are substantively mistaken. But, as he adds, “ . . . when a decision has substantive defects, what positive reason remains to overcome my objection? To this challenge, the answer is that if the decision was democratically made, then that is a positive reason that at least countervails against, even if it does not ultimately prevail over, my objection” (316; emphasis ours). He goes on to say that the reasons “weigh” against the objection.

  Tere is a further sufcient condition but we put this aside.   For Viehof they are also a source of authority. 21   Tough Kolodny’s caveat (2014b: 299) is that this is required only if we assume ongoing social interaction. 22   As Viehof (373) plausibly points out, the value of social equality would confict with social justice not only if the socially just decision was enforced despite a diferent democratic decision but also if someone merely threatened to enforce it (or even simply threatened to disobey the democratic decision). 19 20

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

21

Viehof may well be defending the conventional view, though his position is ambiguous.23 But even if his account were a version of that view, his endorsement of the value of social equality is not coupled with a defence of the value (since his main aim is rather to articulate how social equality, if accepted, bolsters the authority of democracy). As a result, he does not ofer reasons why legitimacy should outweigh social justice when the two confict (353).24 Whether it is Viehof’s view or not, we can consider whether avoiding social inequality must take precedence over the advancement of social justice (once grave injustice is of the table) when the two confict. We are told that social inequality can arise simply because some (genuinely) have  superior knowledge to that of others. Suppose, then, that some people have more time and capacity than others to research some policies (perhaps they do it for a living), and thereby acquire superior knowledge. Suppose they know that the democratic decision would unjustly deny additional life-saving treatment for the sake of extra pain relief on maternity wards. It is unclear why the value of equal social relationships among citizens must take precedence over the importance of providing the additional life-saving treatment. Of course, in this case it may matter whether the extra power is being wielded secretly or visibly, how often it is wielded, and whether the sources of the extra power are themselves innocuous or troubling; all these factors potentially bear on how deeply the wielding of the power undermines social equality. Tus the fact that we may accept as permissible the occasional enforcement of social justice, when it conficts with the democratic decision, does not commit us to accepting that enforcement if it were to be widespread. Nonetheless, it is implausible that the disvalue of subordination must always, except when grave injustice is at stake, take precedence over the value of greater social justice. Once our opponents accept that the value of 23   Viehof explicitly ofers a defence of democratic authority where citizens have a duty to obey (and, as we interpret him, where the decision is permissibly enforceable). He allows that “ . . . an egalitarian procedure may lack authority [and, we think, the ability to confer permissible enforceability on a decision] because it is insufciently likely to reach the correct conclusion . . . Te benefts of submission to the democratic authority, measured in terms of relational equality, may be insufcient to justify the expected costs measured in justice or the common good” (372). One passage suggests that he allows the injustice that takes precedence over the democratic decision to be merely more than “somewhat unjust” (373) but another suggests that the degree of injustice that can prevent democratic decisions from being authoritative should be decoded in terms of “severe injustice” (374). 24   He does suggest that only social equality can ground authority, which may count in favour of his account. To treat this as a reason why legitimacy should trump justice would, however, be to put the cart before the horse.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

22

Zofa Stemplowska and Adam Swift

social equality is to be weighed (or otherwise assessed) against the value of greater social justice in any given case, the position that one will invariably outweigh or take priority over the other becomes untenable. Not being ruled by others—standing as a social equal with others—is indeed valuable. But why think it should defer only to the avoidance of grave injustice? Tere may be no grave injustice, we assume, if some vaccination is not researched and provided, but some social inequality should be accepted, if necessary, for the sake of having it.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

5.  IS THE BALANCING VIEW TOO FRIENDLY TO LEGITIMACY? We have focused on the implausibility of legitimacy always outweighing less-than-grave injustice, but we also claimed that legitimacy may sometimes weigh against the advancement of justice. Tis last claim will come as no surprise to the proponents of the conventional view but it may be that both the conventional view and the balancing view are mistaken on this score. To see the force and the exact target of this worry, suppose that the democratic body that decides on the unjust policy is also the enforcer of the policy.25 Tere is something troubling in holding that it may be permissible for this body to enforce its unjust policy because of the value of legitimacy. Rather, it should have adopted the just policy to begin with. Put diferently, we might think that a third party may have a greater reason to bow to the fact that a policy has the right provenance than the party whose decision it is. By way of analogy, consider that your friend may have a greater reason to help you advance your autonomous decision to do something mildly stupid than you yourself have, since you should instead abandon the decision. Tis line of thought correctly suggests that the permissibility of enforcing less-than-just decisions may depend on whether the body that makes the decision is also that which enforces it. Tere will be cases where it will be impermissible for the body to enforce the decision when the same decision could permissibly be enforced by a third party on the grounds that it was democratically made. Nonetheless, we resist the conclusion that legitimacy can never outweigh justice in cases where the decision-maker and the enforcer are the same; it may be permissible for the body that made the unjust decision to enforce it. Why? Recall that we have so far been focusing on ideal legitimacy. We have been assuming that the democratically legitimate decisions have been made   We are grateful for discussion to Matthew Clayton and Andrew Williams.

25

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

23

by the best available procedure. Tat procedure need not, perhaps, be the best way of getting the decision right but it is, all things considered, the best way of deciding. Given this, we can see the unjust decision as nonetheless a reasonable decision to reach. Tis may be because the injustice is only factrelative (rather than evidence-relative) or because, say, the demos had so little time to reach the decision that the injustice is evident only to those who have had a longer time to consider the issue. In such cases we should think of the body as being in a similar position vis-à-vis its decision as a third party would: although the decision is unjust, the demos did not fall short in any of its duties when reaching it. If the loss in justice of enforcing the unjust decision is smaller than the loss in legitimacy of not enforcing it, it may still be permissible, all things considered, to enforce the unjust decision.26 We do not mean by that, of course, that the fact that no one failed in their democratic duties or obligations in reaching a decision makes the decision permissibly enforceable; this would contradict the balancing view. Rather, we are saying that the fact that no one failed means that appeals to the value of enforcing the legitimate decision are not blocked by the response: “if you wanted to enjoy legitimacy, you should have acted better than you have done.” Tus, just as long as the demos is not falling short in its democratic duties and obligations, legitimacy can weigh in favour of permissible enforceability. But this qualifed defence of legitimacy might be thought to weaken our case for the balancing view, because it suggests that those subject to the injustice have no grievance or complaint. If the best possible procedures have been followed, then nobody has done anything wrong, nobody has failed to act on any obligation owed to anybody else. One might say that, although the result is social injustice, nobody has acted unjustly, perhaps even that nobody has been treated unjustly. Perhaps all that citizens are owed—as a matter of justice—is the establishment of, and sincere and conscientious participation in, the best procedure for reaching political decisions. Tough attractive, that doesn’t seem right. Consider the permissible enforceability of a criminal justice decision rather than a democratic decision. Imagine an innocent convict who accepts that all concerned in reaching the verdict that convicted her deliberated and acted precisely as they should have done—except that they got it wrong (Estlund  2007; Overland and Barry 2011). She may have no complaint against the procedure; neither against those who decided what the procedure should be, nor against those 26   Tings might look diferent if, having taken the reasonable but mistaken decision, the demos learns of its injustice prior to enforcement. Even if the demos could not now reverse the decision, it may be that it should absorb some costs of the injustice it now knows itself to have decided upon.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

24

Zofa Stemplowska and Adam Swift

involved in it. But it is counter-intuitive to think that the lack of complaint means she has not been wronged by the conviction and imprisonment. Te mere fact that the decision was entirely legitimate may excuse its enforcement but it does not necessarily outweigh the reasons to secure a just outcome; enforcement of the just outcome may be permissible while enforcement of the legitimate one is impermissible. Te analogous thought applies in the political case. If the content of a democratically legitimate decision has indeed misidentifed what people owe one another, and what should in fact be enforced, then the absence of any complaint against those responsible for the misidentifcation does not mean that what has been misidentifed no longer qualifes as “social justice.” And it does not neutralize or annul the reasons to enforce the correct judgement.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

6.  DEMOCRATIC LEGITIMACY IN THE REAL WORLD We have articulated a “balancing” view on which both the legitimacy of  procedures and the justice of outcomes are pro tanto considerations relevant to the question of permissible enforceability—and observed that the conventional position, whereby gravely unjust decisions reached by democratically legitimate procedures may not properly be enforced, itself relies on that view without justifying why the view applies only in the case of grave injustice. We have rejected that conventional position as unduly restrictive: a democratically legitimate law or policy need not be gravely unjust for it to be wrong to enforce it. But there is a further reason to be wary of claims that legitimacy trumps justice in all but grave injustice cases. Our discussion so far has been concerned with the case in which perfectly democratically legitimate procedures produce unjust laws and policies. Like much of the literature, we have simply assumed that the procedures in question fully respect or promote the values that render their outcomes legitimate. But one of our motivations for articulating the “balancing” view is that this assumption is wildly false. Much has been written about whether—and in what way— “ideal” theories of justice have practical signifcance (Simmons,  2010; Stemplowska and Swift,  2012; Gaus,  2016). Tat signifcance is usually understood in terms of such theories being action-guiding; action-guiding not merely in principle but in practice; i.e., for us, here and now, in our own, far-from-ideal, circumstances. Te fact that our circumstances also fail to meet the demands of any plausible conception of “ideal” legitimacy, and the action-guiding implications (if any) of that second defcit, have received far less attention. If our ways of making political decisions are no more

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

25

legitimate than the content of those decisions is just, then procedural solutions to problems of injustice are as practically irrelevant as the ideal theories of justice they sometimes seek to replace. Even if the conventional view were right that, and even if, as Valentini and Pettit argue, “political justice” demands that decisions about what societal arrangements are implemented should be made democratically, none of that tells us how to weigh the competing considerations in polities, like the US and the UK, that fail to meet any plausible criterion of ideal legitimacy: polities whose decision-making procedures do not respect those very capacities or qualities of those subject to the decision that are appealed to by accounts of democratic legitimacy.27 For considerations of legitimacy to be properly taken into account here and now, we need non-ideal theories of legitimacy. What conditions must a decision procedure meet in order for it to permissibly enforce its somewhat mistaken judgements about social justice, even when that procedure does not measure up to a plausible specifcation of the value of, say, “equal respect” or “social equality”? Acknowledging the imperfections implied by feasibility constraints of various kinds—constraints that are often discussed in relation to justice but less so in relation to legitimacy—how much and what kind of democracy should be demanded as a condition for making the enforcement of injustice permissible? We might, that is, factor all the non-idealness into the concept of legitimacy itself. Perhaps what it means for a procedure to be “legitimate” is precisely that it is legitimate enough to make its decisions permissibly enforceable. Tat would allow us to say things like: “Te way laws are made in the UK fails miserably to realize the values we might hope to see realized by a political procedure, the values in virtue of which enforcing somewhat unjust political decisions would be permissible. Still, all things considered, it realizes them enough.” Appeals to the legitimacy of fawed procedures—where some cannot register to vote or lack basic literacy, where election promises are unreasonably broken, where bills do not get adequate scrutiny, where policies respond to media scares, where money can buy infuence, where politicians knowingly mislead—need to establish 27   Pettit (2015: 11–12) tantalizes the reader by promising to consider how “social and political issues . . . ought to be weighted in relation to one another,” asking “Should you be prepared to pay a cost in one form of justice for a beneft in another? And if so, how should the trade of go?” But he fails to ofer any kind of answer to those excellent questions. Te main kind of “priority” he accords the political—“Te very activity of proposing an ideal or theory of social justice presupposes under those assumptions that there ought to be politically just, suitably democratic procedures in place in the society targeted”—gives us no guidance. His discussion of “political justice” (aka “legitimacy”) is itself too idealized to give us any traction.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

26

Zofa Stemplowska and Adam Swift

that such compromised legitimacy should still trump the pursuit of social justice by other means. We fnd no argument that would ofer a blanket ruling in favour of nonideal legitimacy in, say, the UK, over and above all instances of apparent injustice. Tus the job centre ofcial who adds someone to the list of approved benefts claimants against the rules, the teacher who helps her under-resourced students “cheat” in tests, the local politician who secretly diverts police resources to deprived and unsafe neighbourhoods may not only be promoting social justice; it may be impermissible to coerce them into acting in accordance with democratically approved outcomes and, indeed, their use of coercive means to advance their own views of justice may itself be permissible. Tese matters will depend partly on the correct balancing of the relevant kind of (in)justice vis-à-vis the actual, non-ideally legitimate procedure that issued in a diferent decision. As we have said, both judgements of injustice and illegitimacy, and the confdence with which such judgements can be held, are scalar. It should go without saying that none of this licenses the unilateral promotion of one’s favoured conception of justice. Te fact that the (less or more) legitimate procedure cannot be relied on always to produce laws or policies that are permissibly enforceable, does not mean that anything goes. It just means that there is no avoiding difcult judgements.28

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Caney, S. (2014). “Justice and the Basic Right to Justifcation,” D. Owen (ed.) Justice, Democracy and the Right to Justifcation: Rainer Forst in Dialogue. London: Bloomsbury Academic, 147–68. Christensen, D. (2009). “Disagreement as Evidence: Te Epistemology of Controversy,” Philosophy Compass 4: 756–67. Christiano, T. (2004). “Te Authority of Democracy,” Journal of Political Philosophy 12: 266–90. Christiano, T. (2008). Te Constitution of Equality: Democratic Authority and Its Limits. Oxford: Oxford University Press. Estlund, D. (2007). “On Following Orders,” Journal of Political Philosophy 15: 213–34. Estlund, D. (2008). Democratic Authority. Princeton NJ: Princeton University Press. 28  We thank Matthew Clayton, David Estlund, Annabelle Lever, Victor Tadros, Andrew Williams, and an anonymous reviewer for discussion and detailed comments. We are also grateful for comments on more or less distant cousins of this chapter to audiences at Barcelona, Braga, Budapest, Edinburgh, Geneva, Manchester, Nufeld College, Oxford’s Centre for the Study of Social Justice, and Warwick’s Centre for Ethics, Law and Public Afairs.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Dethroning Democratic Legitimacy

27

Forst, R. (2014). “Two Pictures of Justice,” D. Owen (ed.) Justice, Democracy and the Right to Justifcation: Rainer Forst in Dialogue. London: Bloomsbury Academic, 3–28. Gaus, G. (2016). Te Tyranny of the Ideal: Justice in a Diverse Society. Princeton NJ: Princeton University Press. Goodin, Robert E. (2000). “Democratic Deliberation Within,” Philosophy and Public Afairs, 29: 81–109. Kolodny, Niko (2014a). “Rule Over None I: What Justifes Democracy?,” Philosophy and Public Afairs 42: 195–229. Kolodny, Niko (2014b). “Rule Over None II: Social Equality and the Justifcation of Democracy,” Philosophy and Public Afairs 42: 287–336. Overland, G. and C. Barry (2011). “Do Democratic Societies Have a Right to Do Wrong?,” Journal of Social Philosophy 42: 111–31. Parft, D. (2011). On What Matters: Volume 1. Oxford: Oxford University Press. Peter, F. (2016). “Two Grounds of Political Legitimacy,” working paper. Pettit, P. (2015). “Justice: Social and Political,” D. Sobel, P. Vallentyne and S. Wall (eds.) Oxford Studies in Political Philosophy, Volume 1. Oxford: Oxford University Press, 9–35. Quong, J. (2010). Liberalism without Perfection. Oxford: Oxford University Press. Rawls, J. (1993). Political Liberalism. Cambridge MA: Harvard University Press. Sen, A. (2009). Te Idea of Justice. London: Allen Lane. Simmons, A.J. (2010). “Ideal and Nonideal Teory,” Philosophy and Public Afairs 38: 5–36. Sinclair, T. (2018). “Te Power of Public Positions: Ofcial Roles in Kantian Legitimacy,” D. Sobel, P. Vallentyne and S. Wall (eds.) Oxford Studies in Political Philosophy, Volume 4. Oxford: Oxford University Press, 28–52. Stemplowska, Z. and A. Swift (2012). “Ideal and Nonideal Teory,” D. Estlund (ed.) Te Oxford Handbook of Political Philosophy. Oxford: Oxford University Press, 373–89. Valentini, L. (2012). “Assessing the Global Order: Justice, Legitimacy, or Political Justice?,” Critical Review of International Social and Political Philosophy 15: 593–612. Viehof, Daniel (2014). “Democratic Equality and Political Authority,” Philosophy and Public Afairs 42: 337–75.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

2

Te Power of Public Positions Ofcial Roles in Kantian Legitimacy Tomas Sinclair

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. Introduction A state is a system of institutional norms. Some of these norms create ofcial roles, purporting to give some people title to tell others what they may and may not do, and to physically compel them to comply. Agents of the state make and apply laws covering just about every aspect of our lives: not only such obvious areas as bodily integrity, but also housing, communications, roads, bequests, professions, pensions, and market exchanges, to name but a few. At some point, if you keep refusing to do as they say, you will be locked up. In this way, agents of the state take themselves to have far-reaching authority, that is, the moral power to alter people’s moral rights, duties, liabilities, and powers by issuing binding directives over a wide range of domains. How can anyone gain the right to do all this to anyone else? On the face of it, it’s a tremendous intrusion. And the state’s behavior is often felt that way. Tink of the expropriations involved in road-building or nationalization programs, or of recreational drugs laws, or think of government surveillance, or just of tax. Tis chapter explores the Kantian response to this question, which takes our innate equality and freedom as its starting point. Kantians argue that the state, and hence the ofcial positions that are partly constitutive of it, are in fact a necessary condition of the equal freedom of its members. Such freedom demands that we can have rights in external things—most obviously property rights, but also contractual and some other types of rights. Yet these “acquired rights” are not compatible with our innate equal freedom except under the jurisdiction of a (properly constituted) state, in which the agents of the state, occupants of its ofcial

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

29

positions, really do have the authority they claim. So the state’s authority is justifed. Tis conclusion is supposed to be independent of empirical speculation. If Kantians are right, defending the state’s authority calls for no knowledge of natural human dispositions to violence or partial judgment. Even perfect good faith and unanimous agreement in the “state of nature” would not be enough to render acquired rights compatible with the innate right. Nor do we need to show that we’re materially better of in the state. Tese features make the Kantian account attractive as a response to skeptics about political authority. But there is a puzzle about just how any concretely existing Kantian state could achieve what it is supposed to achieve. Our innate equal freedom makes the state of nature problematic because in the state of nature, a person’s attempt to acquire property involves subjecting everyone else to her will. But, as I explain, citizens of a state appear to be subjected to the will of legislators, judges, and agents of enforcement in a way that looks as if it should be no less troubling from the Kantian point of view. Moreover, anything that might be said of state ofcials in response seems sayable of individuals in the state of nature too—which would suggest that if the Kantian state is sufcient to solve the state of nature’s problems, it’s not necessary. Te particular forms of rule-following that constitute a state just don’t seem to have the power that Kantians attribute to them. Contemporary Kantians don’t do enough to explain how this puzzle is to be solved. But whether the Kantian account can live up to its promise depends on its success here. If it cannot solve the puzzle, then the Kantian account ofers no answer to skepticism about political authority after all. In this chapter, I sketch a solution to the puzzle. I set out the Kantian account, going into some detail about the problems Kantians identify with the state of nature and explaining the way in which the institutional norms and the coercive power of the state are supposed to address these. I go into detail here to make clear the nature and scale of a challenge to the Kantian account that I go on to describe in §2. In §3, I consider the most developed contemporary Kantian explanation of the way in which the state is supposed to meet the challenge. I argue that it leaves important questions unanswered. In the remaining sections I propose my own explanation. Key to the explanation is a distinction between legislative acts that efectively privilege me as by nature a special source of authority over others, and legislative acts that do not. Te former are not compatible with our innate equality, I argue, and only the institutional norms of the state can make possible the latter. For reasons of space, I focus solely on questions about the power of ofcial roles in the Kantian state. As I argue, it is the ofcial roles created by

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

30

Tomas Sinclair

the constitutive norms of the state that are key to solving the puzzle. Accordingly, I don’t say much about wider questions concerning the shape that the Kantian state must take. I also say little about the interpretation of Kant. It is not my aim to argue that contemporary Kantians have misread Kant, but to show how the account that they have infuentially defended must be understood if it is to fulfll its promise. 2. The Kantian Account

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

2.1  Trouble in the State of Nature Te recent revival of Kant’s political philosophy has made the outlines of the Kantian account familiar.1 Kantians argue that a properly constituted state is a necessary and sufcient condition for its members’ claims of property right, contract right, and “rights to [other] persons akin to rights to things” to be morally unproblematic (Kant 1996: 61). (Tese are a central class of claims purporting to confgure the appropriate distribution of individuals’ external freedom—roughly, freedom over the physical.) In the absence of a properly constituted state (i.e. in the state of nature) making such claims of acquired right inevitably puts claimants in morally objectionable relations with others.2 Te problem is that in the state of nature, acquired rights cannot be compatible with our innate equality, which is manifested in a universal innate right of “freedom as independence from being constrained by another’s choice, insofar as it can coexist with the [same] freedom of every other” (Kant 1996: 30). Suppose that we’re in the state of nature, and I fnd a clamshell, which I claim as mine. In doing that, I take there to be a law or rule (e.g. “fnders keepers!”) that assigns me rights in the clamshell and you (and everyone else) correlative duties to respect those rights. I presume myself authorized by this law to take the clamshell out of your reach and in that sense hinder your attempts to take possession of it, and so in that way authorized to physically constrain your choices in accordance with the law I (perhaps only implicitly) invoke. Now, since we are in the state of nature, there has been no authoritative promulgation of the law in question: I simply judge it to exist and apply here. But then to suppose that this law and its application are authoritative, that it gives moral reality to my claim and 1   See e.g. Flikschuh (2000,  2008); Ripstein (2009), Stilz (2009), Hodgson (2010), and Pallikkathayil (2010). 2  A distinct class of claims purporting to confgure the appropriate distribution of individuals’ external freedom are rights of bodily integrity. I set questions about such rights aside here.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

31

your correlative duties, is to suppose that my will can be “unilaterally lawgiving” for you—that it binds you more than yours binds me. Yet that is  inconsistent with our equal innate freedom as independence. So my claim  cannot be authoritative. Tus the innate right of freedom as independence is revealed to be a right of equal authority in the specifcation of the distribution of external freedom (Kant 1996: 30). Stated in this general way, the argument looks vulnerable to the objection that what I suppose is not that my will (in the form of my judgment about applicable law) is lawgiving for you, but that the truth is, so to speak.3 However, in acting on my judgment about the truth about applicable law in the state of nature—by pressing claims and constraining others on the basis of those claims—I efectively privilege my judgment in the determination of what each of us owes and is owed, and thereby take my will to be unilaterally lawgiving.4 Of course, if someone’s judgment were accurate (not only about the laws of external freedom but about her own accuracy in judging them), then it wouldn’t be obvious why she couldn’t authoritatively make claims of acquired right. Suppose that Lockean labor-mixing principles of frst appropriation are true. Why couldn’t someone accurately judge that they were and authoritatively make a claim to some land in accordance with them? After all, it’s not as if Kantians deny the applicability of moral principles to questions of external freedom, even without authoritative promulgation, as their commitment to the principle specifying the innate right of equal freedom shows. However, Kantians do deny the possibility of accuracy in the requisite sense here. Tough a priori principles governing acquired rights prescribe equal freedom, they are not specifc enough to determine a uniquely morally permissible concrete distribution of external freedom. As Anna Stilz notes, there are many possible systems of property that are potentially consistent with the principle of equal freedom, and under each of these there will also  be many possible particular rules about precisely which bundle of Hohfeldian incidents property consists in, how to get it, who gets which things, and so on (Stilz 2009: 40). Te principle of equal freedom doesn’t even determine a unique procedure for resolving the indeterminacies (Pallikkathayil 2010: 137).5   See Raz (1998: 27–8); Enoch (2010: 982; 2013: 159–60); Viehof (2014: 344–5).   See Gaus (1996: 122–9;  2015: 1091–3). Te same problem arises even if I act on someone else’s judgment. See §6. 5   Some Kantians would, I think, argue that the line of argument I’m ofering here misconceives the problem. Te problem, they would say, is that even if the only possible content of the only laws that could possibly be authoritative were fully determinately given, those laws could not be binding without the backing of some source of authority. 3 4

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

32

Tomas Sinclair

Contemporary Kantians distinguish three distinct aspects of the general problem of acquired rights in the state of nature. Te frst is the problem of unilateral choice, which concerns arrogation of authority. It is the problem most naturally suggested by the outline I gave above. Te problem is that if I could make acts of appropriation in the state of nature, then I would have to be able to confer authority on my acts so that they generated new rights for me and duties for everyone else, rights and duties that reconfgure the enforceable distribution of each person’s external freedom. But my being a source of the requisite sort of authority implies the capacity unilaterally to bind others through my acts. And this is inconsistent with the innate right of equal freedom. Again, it might be protested that I am not claiming to be the source of any authority: I am only channeling the authority of general principles of equal entitlement to freedom whose applicability and bindingness Kantians are happy to concede. But even if that’s true, I nevertheless assume special authority to resolve the legislative indeterminacies that the general principles leave open. And that privileges me and my judgment about the matter over  everyone else—it makes my judgment distinctively instrumental in the  confguration of constraints on equal freedom. So I cannot efect an authoritative appropriation. Te problem of unilateral interpretation is a problem concerning the application to particulars of the laws invoked by any claim of acquired right in the state of nature. Tis problem persists even if we set aside the problem of unilateral choice. Te problem is that the laws’ application to particulars would still involve the privileging of individuals’ judgments about it. For instance, you and I might agree on the authoritativeness of a law that says clamshells on the beach are mine and clamshells in the sea are yours, and yet disagree about this clamshell, which is being moved back and forth by waves on the beach. Once again, if my claim were authoritative, then you (and everyone else) would be subject to constraints privileging my judgment, but I would not be subject to constraints privileging yours. But that wouldn’t be consistent with our innate equality. So my claim can’t be authoritative. Te problem of unilateral interpretation is very similar to the problem of unilateral choice in that both arise as a result of indeterminacy engendered by general principles.6 But problems are importantly diferent all the same. (Compare Korsgaard 1996: 23–4.) But it seems to me that if the only possible content of the only laws that could possibly be authoritative were fully determinately given, then a unilateral acquisition according to those laws could not be objected to on the grounds that it lacked the backing of the right kind of authority, since the backing of such authority wouldn’t change anything—it would have no choice but to ratify such an acquisition. So the indeterminacy of the application of the general principle of equal freedom is crucial to the Kantian argument, as I read it. 6   As I observed earlier, some Kantians might resist this claim. See note 5.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

33

Te indeterminacy that gives rise to the problem of unilateral choice is an indeterminacy in the specifcation of the general principles for distributing external freedom that adequately respect innate equal freedom. Such indeterminacy isn’t inevitable, conceptually speaking. It might have been that the innate right analytically implied a uniquely applicable principle for  the distribution of external freedom. Te indeterminacy that gives rise to the problem of unilateral interpretation, by contrast, is an inevitable indeterminacy in the application of any general principle to concrete particulars (Ripstein 2009: 168–72). Even if there were a uniquely applicable principle for the distribution of external freedom, its application to concrete particulars would still generate the problem of interpretation.7 Te problem of assurance is a problem that arises because the laws invoked by a claim of acquired right in the state of nature purport to authorize the imposition by right-holders of physical constraints on what others can do. Te problem persists even if we set aside the problems of unilateral choice and interpretation. Te problem is that my actions in accordance with my claim in any given case may efectively constrain you in such a way that you are laid open to being taken advantage of by me (whether or not I will in fact take advantage of you). For I may not in turn be subject to efective constraint by you. For example: I may be able to take efective possession of the clamshell that the laws I invoke specify as mine (efective because you can’t take it for yourself ). But at the same time, it may be that you can’t take efective possession of a mussel, say, that the same laws specify as yours (even as I apply them), because you can’t stop me taking it. If the laws in question were authoritative, then you would be required to respect my claim to the clamshell even though there was no prospect—you had no assurance—of my respecting your claim to the mussel.8 So you would be authoritatively constrained to leave yourself open to being taken advantage of in this way—to depend upon me for the fulfllment of your claims even as I did not depend on you. But that would be inconsistent with our innate equality. So my claim can’t be authoritative.9   I thank Stefano Lo Re and an anonymous referee for pressing me to clarify this.   It is not part of the content of the laws that what people claim is theirs only on condition that they will respect others’ claims. One problem with such laws would be that—assuming that it is indeterminate at time t1 whether or not a person will respect others’ claims—they would be incapable of specifying anything as determinately anyone’s. Tanks to Tom Simpson for pressing me to clarify this. 9   Tis way of putting the problem owes much to Ripstein’s (2009: 159–68) discussion of it. Others, such as Stilz (2009: 51) and Pallikkathayil (2010: 139–40) emphasize individuals’ dependence on others’ being of similar will, in the state of nature, for the enjoyment of their acquired rights. I think the diferences are ultimately only a matter of presentation. 7 8

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

34

Tomas Sinclair

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

2.2  Te Kantian State Te properly constituted state is supposed to solve these problems. In the properly constituted state, property and other forms of acquired right are  institutionalized so that the laws under which individuals make their claims are no longer the expression of anyone’s unilateral will. Nor are the application and enforcement of those laws. Tey are instead the expression of a public, “omnilateral” will. As a result, individuals can enjoy the advantages of acquired rights without involving themselves in morally objectionable arrangements. Indeed, there is a general duty to establish the properly constituted state (Kant 1996: 86). It is a requirement of practical reason—which is, for Kant, a requirement of freedom itself—that rights in external things should be rendered possible (Kant 1996: 40–2). Hence, a properly constituted state exercises legitimate authority. Tere is no moral objection to its power, at least in principle, because its having that power is the only condition under  which its members can enjoy the acquired rights that must be possible for them. How is this supposed to work? Te idea is that the institutions of the properly constituted state make the law, its application, and its enforcement the issue of ofcials whose positions constitute a perspective that is not that of any private actor. At least in its ideal form, this makes the state perfectly representative (because constitutive) of the united will of its members, so that the laws under which individuals make their claims are the expression of that will, and the imposition, interpretation, and enforcement of such laws are morally unproblematic. Claims of acquired right under these laws  are therefore compatible with the innate rights of all. As Arthur Ripstein explains: Public acts are omnilateral because they are not any particular person’s unilateral choice, but instead are exercised on behalf of the citizens considered as a collective body. Tey are also omnilateral in a further sense: a unilateral will always has some particular end, some matter of choice. Te omnilateral will is diferent, because all that it provides is a form of choice, by providing procedures through which laws can be made, applied, and enforced . . . when the state authorizes the acquisition of private property, it does not make the having of property, or the accumulation of wealth, its purpose. Its purpose is to enable individual human beings to have things as their own as against each other.  (Ripstein 2009: 196)

Te most important means by which this is achieved is that there exist legislative, judicial, and executive roles in the Kantian state that are defned by obligations to create and sustain through law a rightful condition, so that when ofcials exercise their powers, they act with the Kantian public purpose alone. Tus, the authority invoked in any individual’s claims of acquired

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

35

right is that of the law as representative of the united will of the people, rather than of any particular person’s private judgment. Kant describes the state in which ofcials are genuinely representative of the united will of the people as a “true republic” (Kant 1996: 112). Tere are other substantive institutional requirements for this. Two that feature in the argument I will present are as follows. First, the laws are substantively just at least in the sense that their content specifes nothing that violates citizens’ innate rights—for instance, by classing some as serfs or the slaves of others (Kant 1996: 45). Second, the law is generally promulgated (Kant 1996: 89). But the key element in the Kantian account for our purposes is the ofcialrole-constituting institutions of the Kantian state. Before I turn to the puzzle that the Kantian account generates, let me stress two important points. First: the way in which the Kantian state solves the problems of the state of nature is not supposed to be a matter of degree. It’s not merely that the morally problematic character of interpersonal relations in the state of nature is diminished in the Kantian state. When the laws express the public, omnilateral will, as they do in the Kantian state, the problem goes away altogether (Kant 1996: 112). Second: the morally problematic character of interpersonal relations in the state of nature, as Kantians understand it, is not supposed to be a function of individuals’ tendency to judge partially or of their proneness to violence. Te fact that the state is a solution is not therefore supposed to be a matter of its limiting any such tendencies. Even a population of perfectly good-willed, peaceful, like-minded individuals could not live together in the state of nature consistent with their innate rights (Kant 1996: 89–90). As Ripstein explains, the need for the Kantian state is not supposed to refect human limitations. It is a matter of facts of reason “internal to the concepts of acquired rights” (Ripstein 2009: 146). 3. Two Objections To The Kantian Account Te Kantian account is attractive. It promises to dissolve doubts about state authority and power in part by appeal to the very values—individual freedom and equality—that make them seem suspicious in the frst place. And it claims the dialectical virtue of not relying on empirical speculation about the state of nature. But there are two simple objections that the account must address if it is to live up to its promise.10 Te frst is that 10   Tese objections are most clearly and powerfully articulated by Kolodny (n.d.). Of course, there are other objections to the Kantian account (see e.g. Sangiovanni  2012, Valentini 2012), but if it cannot address the two that are my focus here, it fails even by its own lights.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

36

Tomas Sinclair

the Kantian state isn’t sufcient to solve the problems of the state of nature, even as the Kantian conceives them. Te second is that the Kantian state isn’t necessary to solve those problems either.

3.1  Is the Kantian State Sufcient? Te insulation problem. As we’ve seen, the Kantian solution relies heavily on  the idea that legislative, executive, and judicial acts are expressions of  a  public, omnilateral will because the roles in question are defned by obligations to create and sustain through law a rightful condition. Te trouble is that in any existing or realistically possible state, these roles will be occupied by humans, whose acts in their ofcial capacities will presumably refect and so privilege their own judgments about the law and its application. Teir public judgments cannot be perfectly insulated from their private judgments. It seems impossible, therefore, that even a perfectly constituted state could eliminate the asymmetry of subjection to individual judgment that precludes the possibility of acquired rights in the state of nature.11 Te guarantee problem. Moreover, no existing or realistically possible state can guarantee that no one will take advantage of others’ compliance with the law by failing to comply herself, in just the way identifed by Kantians as problematic in their elaboration of the problem of assurance (see Kolodny n.d.: 16). So we have another reason to doubt that the Kantian state is sufcient to solve the problems of the state of nature.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.2  Is the Kantian State Necessary? Te ideal state of nature. Te second objection is that the Kantian state isn’t necessary to solve the problems that Kantians identify anyway. Te thought here is that there isn’t any reason why the state of nature couldn’t in principle be a place in which the following conditions obtained: a. everyone agrees upon both the specifcation and the application of the laws of external freedom; b. the content of the laws is substantively just; c. the specifcation and application of the laws aren’t infuenced by any objectionable bias towards any particular person’s private purposes; and d. each person is assured of everyone else’s compliance, thanks to universal, stable, independent, and completely efective dispositions to prevent violations of the rights specifed under those laws.  (Kolodny n.d.: 16)   See Kolodny (n.d.: 10–11).

11

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Te Power of Public Positions

37

In such an ideal state of nature, it would seem that neither individuals’ claims of acquired right nor their activity in accordance with their claims would be inconsistent with anyone’s innate rights. And so if the Kantian state is necessary at all, then that’s not something that can be shown without recourse to empirical speculation—it depends on the fact that the ideal state of nature, though conceivable, is practically unattainable.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.3  How the Two Objections Interact Note that anything Kantians might say by way of reply to the sufciency objection will tend to undermine their ability to give a satisfactory reply to  the necessity objection. Suppose, for example, that Kantians advance some consideration that explains why state ofcials can be relied upon not to make or apply law in a way that objectionably refects their private judgments. It’s hard to see why any such consideration could not also be invoked in respect of at least the ideal state of nature. If state ofcials can be in sufciently good faith, for example, why couldn’t private individuals in the state of nature be in sufciently good faith too as they apply the norms of organically evolved and long-established practices of acquired right? Perhaps it’s less likely that everyone in the state of nature could be relied upon to act from public motives than it is that ofcials of the state can, since the ofcials are presumably fewer in number. But unless it’s categorically impossible for them to act only from appropriately public motives, the problem remains. Similarly, the condition (d) of assurance of compliance in the ideal state of nature suggests that anything that can be said on behalf of the Kantian state’s powers to solve the problem of assurance can also be said on behalf of the ideal state of nature’s. And nor do the idealizations of power involved in imagining a genuinely omnipotent executive seem available only to those who are imagining a state. Even if omnipotence is what it takes, that seems no better for the Kantian than her opponent. 4.  Ripstein’s account To illustrate the force of these objections, consider Arthur Ripstein’s infuential presentation of the Kantian account in Force and Freedom. Ripstein goes into more detail than anyone else about the way in which ofcial roles are supposed to make the diference in the Kantian state. If the objections above seem troubling even for his view, then further elaboration of the account is needed. As part of what seems to be his response to the insulation problem, Ripstein stresses that ofcial positions in the Kantian state are defned by an

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

38

Tomas Sinclair

absence of private interests, embodied in the obligations associated with the positions, so that they constitute a public, objective standpoint (Ripstein 2009: 191–8). Accordingly, in the Kantian state ofcials are permitted to act only for the purposes of creating and sustaining a rightful condition, and there are prohibitions on the use of public ofce for private purposes, ruling out nepotism, bribery, and other forms of corruption (Ripstein 2009: 192–3). Ripstein suggests that thanks to these institutional provisions, “the distinction between an ofcial’s acting within his or her mandate and outside it does not depend on the ofcial’s attitude” (2009: 193). Now, this does highlight what may be an important advantage of the state over many states of nature—namely, that it ofers better protections against partiality and bias in the framing of laws of acquired right.12 It seems a likely defect of the state of nature that people’s attempts to act in accordance with claims of acquired right will tend to refect judgments that are biased in the sense that the laws invoked are framed, cynically or otherwise, to further the private ends of those invoking them. From a Kantian perspective, the natural way to think about this defect would be as a particularly grievous instance of the problem of unilateral choice—although that problem is supposed to persist in the state of nature even when everyone is in perfect good faith. But what Ripstein says doesn’t seem to show that any concretely existing Kantian state could adequately address such instances of the problem. Even perfectly enforced prohibitions on bribery and so on presumably can’t altogether preclude the use of ofcial powers to pursue private interests. So if that’s what’s needed, the Kantian state is no solution. Moreover, the partiality concern is distinct from the more general concern that was originally at issue. Tis general concern wouldn’t be adequately addressed by Ripstein’s remarks even if the partiality concern were. Te general concern wasn’t that ofcials of the Kantian state can use their ofces to pursue private interests, but that their activities express and so privilege their private judgments, making these judgments instrumental in the specifcation of the law and its application, and so in the confguration of the state’s system of constraints on external freedom. Yet how are legislators and judges supposed to reach their ofcial conclusions other than through the exercise of their own judgment? Insulation from such judgment seems not only practically impossible but also theoretically undesirable insofar as there seems to be no attractive alternative to it.13 So in any concrete Kantian 12   I do not think that there is any appeal to this advantage in Ripstein’s own account of why ofcial roles make the diference. But someone might suppose that it could be appealed to for that purpose. I thank an anonymous referee for pressing me to make this explicit. 13  As Kolodny (n.d.: 11–12) points out, even if the law is never formulated by a legislator (because it arises organically, like language), its application would still call for individual judgment. See also Sharon (2016: 149).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

39

state, the judgment of some individuals must inevitably be privileged over that of others. Now, Ripstein explicitly denies that the Kantian solution “rests on any claim about the ability of ofcials to do anything more than act on their own best judgment” (2009: 191). But he does not explain how the general concern is supposed to be alleviated. I think that’s because he doesn’t take it to be a concern at all. Te problem of unilateral choice is a problem about the arrogation of authority; Ripstein thinks that ofcials in the Kantian state really do have the authority to choose for all, and so the problem goes away no matter how the occupants of ofcial positions go about arriving at their decisions. Once you really are in authority, after all, it doesn’t matter if your private judgment informs your public decisions, so long as you comply with any accompanying obligations. But the reason that an individual doesn’t have the authority to choose for all in the state of nature, even if she takes herself to be bound to choose as a Kantian ofcial would be bound, was that her having that authority would involve privileging her and her judgment in the way that I described above. If that asymmetry undercuts any claim to authority in the state of nature, why doesn’t it also undercut any claim to authority in the Kantian state? It doesn’t seem satisfactory simply to assert that since ofcial roles defne a public rather than a private perspective, Kantian ofcials have authority that individuals in the state of nature lack. Te puzzle was to explain precisely how that is supposed to be compatible with the innate right of equal freedom, given the inevitable involvement of occupants’ judgments in the genesis of their public decisions. Te objections concerning the problem of assurance are also inadequately addressed by Ripstein’s account. Key to the Kantian state’s capacity to solve the problem, according to Ripstein, is its ability to “provide everyone with systematic incentives in relation to the possession of others,” which makes rights violations “prospectively pointless” (2009: 165). Even if you do manage to steal my property, it will remain my property in the eyes of the state, and the state will render my right in it efective by forcing you to give it back to me. “You have what is yours,” Ripstein says, “because if another wrongs you, you will be able to get it back” (2009: 167). So the assurance problem is solved. But this clearly assumes that the state’s enforcement is fully efective. If it wasn’t, not all rights violations would be prospectively pointless, and it wouldn’t be true that you were always able to get back what was taken from you. Yet it’s precisely this assumption that the guarantee problem calls into question, and Ripstein says nothing to explain what justifes it. Nor does he explain why whatever justifes it wouldn’t also be possible in the state of nature.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

40

Tomas Sinclair

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

5.  Where This Leaves Us As we saw, Kantians think there is a categorical diference between the state of nature and the civil state, and that showing this does not depend on empirical speculation. Te lesson of §§3 and 4 is that it’s difcult for the Kantian approach to vindicate this claim. Te best that can be said for the Kantian state as a solution to the problems that Kantians identify doesn’t seem adequate. What’s more, it doesn’t seem better than the best that can be said for the ideal state of nature either. Niko Kolodny deploys these objections to argue (among other things) that the Kantian concern with freedom should ultimately be interpreted as  a concern about status relations, and that only so interpreted can it justify state authority. A state can enshrine equality of status (e.g. through democratic procedures), even if it can’t provide assurance that no one’s judgment will be privileged in the determination of operative constraints on external freedom. It can do so because its institutions can secure equality of opportunity to infuence the state’s decisions, and this is enough to provide assurance that no one is subordinated to anyone else, even if some have decision-making power over others. Te institutional tribute to our equality is sufcient to allay anxieties about relations of social inferiority (Kolodny n.d.: 17–23; see also Kolodny 2014a, 2014b, 2016). But, whatever the merits of democracy or other institutional arrangements as a way of allaying such anxieties, this proposal doesn’t really do justice to Kantian concerns. It certainly doesn’t look as if Kantians are ultimately worried about relations of social inferiority. In Kant’s own work, the justifcatory story that leads from the freedom of the rational will via the “Universal Principle of Right”14 to the innate right of equal freedom includes no trace of a concern with status relations as they feature in the contemporary accounts of political egalitarianism that Kolodny draws upon.15 What’s at stake are more familiarly Kantian concerns about universality and equality— prohibitions on treating oneself as special, as an exception to the general rule, and on not valuing others as equals in that sense.16 And contemporary Kantians are all trying to stay fairly close to Kant’s account.17 14   “Any action is right if it can coexist with everyone’s freedom in accordance with a universal law, or if on its maxim the freedom of choice of each can coexist with everyone’s freedom in accordance with a universal law” (Kant 1996: 24). 15   See especially Miller (1997); Anderson (1999); Schefer (2003). 16   Of course, these concerns may resonate with contemporary Kantians because of deeper concerns about status relations. But equally, matters of social status may seem at stake in authority relations ultimately because of deeper, more familiarly Kantian concerns. Even in the former case, the distinctive form that the deeper concerns take in Kantian views may make for a distinctive account of political authority that is worth exploring in its own right. 17   Tough they may give more emphasis to notions of dependence and domination that the more fundamental Kantian concerns make salient.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Te Power of Public Positions

41

So it seems worth considering whether there is another way to avoid the objections and vindicate the Kantian account of political authority. I think there is. On my way of understanding the Kantian account, the fundamental problem that arises in the state of nature is that no one can make claims of external right without efectively taking herself to be exceptional in just the way that you would expect a Kantian to fnd objectionable. Te problems of unilateral choice and interpretation are instances of this more fundamental problem, which can be solved only by the constitutional norms of a Kantian state. Tus the necessity objection is met. Te problem of assurance is not an instance of the fundamental problem I have just described, and so it is a separate task to address it. Te key thing here, I believe, is to see that the needed solution is not a guarantee but equality of assurance. Now, although this may be achieved in the ideal state of nature, it may also be achieved in the Kantian state. Tus the sufciency objection is met. In the remainder of the chapter, I set out my proposed solutions in more detail.

6.  Why The Kantian State is Sufficient

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

6.1  Te Insulation Problem Te insulation problem was that no realistically possible state looks as if it can help with the problems of unilateral choice and interpretation. Ofcial roles are going to be occupied by humans, whose acts in their ofcial capacities will privilege their judgments about the law and its application. So if privileging of one person’s judgment over others’ is objectionable, the privileging of ofcial judgments over others’ is objectionable. In order to make progress with this problem, let’s reconsider the problem with privileging my judgment in the way that I would if I could efect authoritative appropriations in the state of nature. Obviously the idea is that it’s inconsistent with our innate equality, because it would involve an asymmetry in capacity to constrain others. But perhaps not all such asymmetries are inconsistent with our innate equality, when that is properly understood. My suggestion is this. Tere is a distinction between refexive and nonrefexive privileging of an individual’s judgment. An individual’s judgment is refexively privileged in the determination of the distribution of external freedom when she gives her judgment that role. An individual’s judgment is nonrefexively privileged when her judgment is given that role from without. Tis distinction can make the diference Kantians need. Te idea is that what’s wrong with claims of acquired right in the state of nature is that they

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

42

Tomas Sinclair

privilege individuals’ judgments refexively. And this unjustifably treats the person making the claim as if she is a special source of authority over her fellows: as if she has by nature a unique power to bind others. By contrast, in a properly constituted state, at least, the privileging of the judgments of legislators and judges is nonrefexive. How could this solve the problem? You might protest that even if the judgment of an ofcial role-bearer in the Kantian state is nonrefexively privileged, it would still seem that his choices constrain others more than theirs constrain his, in violation of the innate right. For when another person fnds her options constrained by the distribution of external freedom, the explanation of that appeals to facts about the role-bearer’s judgment but not to any facts about the other person’s. So the distinction between refexive and nonrefexive privileging of a person’s judgments seems not to be any help. But this way of expressing skepticism about the proposal begs the question against it. It’s true that the role-bearer’s choices constrain others in the envisaged situation, but it need not be true that the explanation of why his choices should constrain others need presuppose that he has by nature a unique power to bind others. To see this, consider the following sequence of fantastical scenarios. In computer 1, the laws of external freedom are made, applied, and (fully efectively) enforced by some kind of computer system, so that the judgment of a human is not called for at any point. Assume further that no human input was called for in the programming or setting up of the system—that it was created and switched on by a freak act of nature. Te computer’s algorithms model the perspective of the omnilateral will: it acts as if with the pure intention to create and sustain through law a rightful condition, and the specifcation of its laws meets the substantive justice condition of  compatibility with individuals’ innate rights. I take it that computer 1 would solve the problems that Kantians identify with the state of nature, even on our earlier analysis, which makes the sufciency and necessity objections bite in respect of any realistically attainable state.18 Next, in computer 2, we can introduce the following feature: the computer needs to route a signal internal to its operation via a human 18   Tis is not to say that it would be a realization of the Kantian ideal. In particular, the Kantian–Rousseauian ideal of the people as ultimate authority would not be realized. But as far as the problems that Kantians identify with the state of nature are concerned, it seems to me that the ideal is important only in virtue of the fact that any other concretely feasible candidate ultimate authority would be morally problematic. For any such candidate would be a rational agent other than the people as a whole, so that regarding it as the ultimate authority would simply give rise to the problems all over again. In the fantastical world, there is another candidate: the computer system.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

43

(“the Connector”), who is selected (by the computer) on the basis of her electrical conductivity from those volunteering for the role. (As it turns out, only one person satisfes the computer’s conductivity requirements.) Surely this makes no diference: computer 2 would also solve the problems of the state of nature. Tere is no Kantian objection to the Connector’s role in the  determination of the distribution of external freedom. After all, her judgment is not being privileged at all—it plays no role whatsoever. Next, imagine a slightly diferent version of the system in which the computer’s legislative output is sensitive to the content of the Connector’s judgment about the appropriate distribution of external freedom, because the relevant electrical signals pass through the physical structures on which the relevant states of her mind supervene—although she is still selected for her conductivity. In computer 3a, the computer’s legislative output does not match the Connector’s judgment, but is still sensitive to that judgment in the sense that a change in the content of the judgment implies a change in the legislative output and vice versa (all consistent with the conditions on the algorithms listed above). In computer 3b, the computer’s legislative output does match the Connector’s judgment. In all cases, the Connector’s judgment is substantively compatible with the general principle of equal freedom. If the problems of unilateral choice and interpretation are understood to  arise in any case in which one person’s judgment is instrumental in the determination of the system of constraints on external freedom, then neither computer 3a nor computer 3b solves the problems. In both cases, the resultant system of constraints on external freedom privileges the Connector’s judgment over others’ judgments. But in the case of computer 3a, at least, this is surely not objectionable from a Kantian point of view in the same way that taking the Connector’s acquisitions to be authoritative in the state of nature would be. In the state of nature, the only explanations that can be ofered for taking her claims of acquired right to be authoritative arbitrarily and unjustifably take her to be by nature exceptional, a natural authority over others. Te fundamental Kantian objection to this is obvious: no one is exceptional in this way by nature, and to treat oneself as if one is exceptional in this way is to act in a  way that cannot be universalized (Kant  1997: 33–4; O’Neill  1989: 94; Wood 1999: 108–9). By contrast, in computer 3a, no one objectionably treats herself as exceptional in this way, and no one needs to assume that the Connector is exceptional in this way to explain why the Connector’s judgment plays the crucial role in the determination of external freedom. It might be objected that the objectionable assumption is made in all of these cases, because of the way in which the computer picks people according to the criterion of conductivity. Surely it’s no more justifable to privilege a

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

44

Tomas Sinclair

person’s judgment because of her conductivity than it is to privilege her judgment because she is a natural authority? But this objection is mistaken. It would be unjustifable in the state of nature to privilege one’s judgment because of one’s conductivity. One’s explanation would have to appeal to some procedure that made conductivity relevant (perhaps because it is randomly distributed, for example), and, as I noted earlier, Kantian principles of equality are not sufciently determinate to uniquely pick out any such procedure. So, to privilege one’s judgment (or anyone else’s) about the distribution of external freedom because of one’s conductivity would be to presuppose that one’s judgment about the procedure for selecting whose judgment should count was specially authoritative. But in the computer-run state, privileging a person’s judgment because of her conductivity is not objectionable in that way, because the procedure for bringing about the rightful condition is not selected by anyone. Now, I think what can be said about computer 3a can also be said about computer 3b, even though the legislative output in that case refects the content of the Connector’s judgment. Tere is no relevant diference between them from the point of view of the innate right of equal freedom. Te point is once again that the privileging of the Connector’s judgment does not involve any arbitrary assumption that she has a natural power to bind her fellows. So even though the Connector’s judgments play a determining role in the system of constraints in force in the computerized state—a role that no one else’s judgments play—and are in that sense privileged, the privileging is not objectionable from a Kantian point of view. We can now see why the Kantian state need not give rise to the problems of unilateral choice and interpretation. For the Kantian state is relevantly like the computerized states. Te Kantian state is constituted, as we saw, by  institutional norms that give rise to ofcial positions. So long as the constitution’s selection of occupants of these ofcial positions is relevantly like the computer system’s procedures for selecting the Connector in that acting on them requires no presupposition that the occupants are by nature special sources of authority over their fellows, then there is no reason why it  should give rise to the problems. For that reason, perhaps hereditary monarchies are ruled out, as perhaps are “epistocracies” to the extent that the selection of the rulers (the epistocrats) is on the basis that they know best in the way Kantians deny they can know best (see §1.1).19 Democratic and lottocratic constitutions, meanwhile, seem more obviously favored, since (in at least some forms) they clearly do not presuppose that anyone is a special source of authority by nature.   I take the term “epistocracy” from Estlund (2008).

19

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Te Power of Public Positions

45

Tis does not mean that no ofcial may be selected at any level on grounds of moral sophistication or virtue, or that in resolving indeterminacies in the application of a priori principles of right ofcials must take care not to assume that their judgment is better than others’.20 It is important to stress this, because otherwise it may seem that the argument tends in the implausible direction of legislation by roll of dice. It does not tend in that direction. Te argument is neutral between constitutions—that is, the systems of norms that set up ofcial roles, their obligations, and the procedures of selection for them—so long as the explanation of the fnal authority (i.e. the last word on the choice, interpretation, or enforcement of the distribution of external freedom) that they bestow makes no appeal to natural authority. So it is neutral between randomly selected rulers and democratically selected rulers at least, though it constrains the rulers not to violate a priori principles of equality in their rule. And once the explanation of fnal authority is in this way freed of the objectionable assumption that anyone is a natural authority over her fellows (or that any group is naturally authoritative over any other), it may be delegated by those who are selected as rulers more or less freely, including on grounds of moral sophistication or virtue.21 And this is why the insulation problem is not a problem—and why Ripstein is right to say that whether an ofcial is acting within her mandate does not, on the Kantian account, depend on the ofcial’s attitude.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

6.2  Te Guarantee Problem Te foregoing shows that the Kantian state may be sufcient to solve the problems of unilateral choice and interpretation after all. But the problem of assurance remains. Since the problem of assurance concerns the security of our acquired rights rather than anything to do with the privileging of judgments, none of what I said in §6.1 looks as if it will help with this problem. Is there any other reason to think that the Kantian state could solve it? As with the insulation problem, the thing to do is review the way in which our innate equality is supposed to generate the difculty in the frst place. Te problem was one of dependence: if the laws I implicated with my claim of acquired right in the state of nature were authoritative, they would constrain others to respect my claims even in the absence of assurance for 20  Arguably, conditions of fair equality of opportunity must obtain, however, if selection on such grounds is to be consistent with the substantive justice condition. Perhaps this rules out even fully democratic selection on grounds of natural authority. Tat would make the reasons that voters have for voting as they do relevant to the legitimacy of democratic authority, though. Compare Viehof (2014: 373.) 21  Although there will be constraints on the delegation, such as a requirement of equality of opportunity and compliance with laws against nepotism, and so on (see §3).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

46

Tomas Sinclair

them that I would be likewise constrained by their claims. Te only assurance available to them would depend upon my will. In this way, the laws would authoritatively constrain some to leave themselves open to being taken advantage of by others. Tis way of setting out the problem, which is representative, makes it natural to think that what’s required is assurance of full and universal compliance with claims of acquired right—which of course gives rise to the guarantee problem. But that’s misleading. What is required is only that I am not more assured of compliance with my claims of acquired right than you are of yours. For only that makes for a problem of unequal freedom. Strictly speaking, a law with which no one had any hope of anyone’s compliance wouldn’t require anyone to lay herself open to being taken advantage of by others in a way that is incompatible with our innate equality. Nor does a law with which we can all expect the same degree of compliance by others, even if that’s not 100%. Now, in a state of nature where the only source of assurance that anyone has is her own strength and the willingness of others to refrain from interfering with her acquired rights, the condition of equality of assurance will most likely not be met, since some will be more able to ensure others’ respect for their claims of acquired right than the others are to ensure reciprocal respect. But the same need not be true in a well-functioning state, for the state’s power is sufciently great that it eclipses even that of the strongest individual. Superior strength therefore does not provide one with greater assurance of respect for one’s acquired rights, since the measure of one’s assurance is the much greater power of the state, which is the same for everyone. As a result, it does not matter that the state cannot guarantee everyone’s compliance with acquired rights. What matters is that no one has ex ante greater assurance than anyone else of respect for her acquired rights. And the state can secure this condition. Even if some are better positioned than others to violate acquired rights, they don’t have a better prospect than others of maintaining the resultant confguration of external freedom. So, for example, even if I’m more able to steal your car than you are to steal mine, I don’t have a better prospect of maintaining the situation in which I have your car and you don’t than you do of maintaining a situation in which you have my car and I don’t. So long as no one can reasonably expect to be at an advantage in respect of her capacity to evade justice (including the restitution of what is yours according to the laws of acquired right), the laws’ authoritativeness does not involve authoritatively constraining anyone to leave herself more open to being taken advantage of by others than they are open to being taken advantage of by her. And this despite the possibility that some will violate others’ acquired rights and get away with it. For although

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Te Power of Public Positions

47

in one sense the existence of this possibility entails that some must have been left open to being taken advantage of by others, it does not entail that they must have been more open to being taken advantage of than the others were to being taken advantage of by them—that they had better prospects ex ante than the others. I am not claiming that any existing state secures the condition of equal assurance.22 But it is not implausible to think that a realistically attainable state might. And so the guarantee problem is not the obstacle it appeared to be.23

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

7.  Why The Kantian State Is Necessary Te response to the assurance problem that I have just described shows that the associated sufciency objection is not fatal for the Kantian view. But the resources it employs are not the exclusive preserve of a state. And, as §2.3 predicted, this makes the necessity objection look all the more forceful. After all, the state’s strength supervenes on its citizens’ support. If a sufciently large number resist enforcement of the law, then it will no longer be true that it can secure the condition of equal assurance, and so it won’t solve the problem. By the same token, if a sufciently large number of people in the state of nature are disposed to enforce laws that they all accept, then their collective strength may eclipse that of any individual in just the way that’s sufcient to solve the assurance problem. And that is just what’s envisaged in condition (d) of the ideal state of nature that I described in §2.2. Kantians should concede this point. A feature of the way in which the problems of the state of nature are typically set out makes the concession look worse than it really is. Te problem of assurance is presented as something that arises even if there is unanimous agreement on both the laws of acquired right and their application to particulars.24 Tis suggests that unanimous agreement on the laws and their application is sufcient to solve the problems of unilateral choice and interpretation. But if that’s right, then the problem of assurance will have to explain why the state is necessary, 22   One clear worry concerns the way in which a state’s military forces may become sufciently powerful that they defeat the equal assurance condition. Another concerns inequality in access to justice (I thank an anonymous referee for pointing this out). 23   An anonymous referee suggests that this account of the assurance problem makes the way the Kantian state solves it a matter of degree. I am not sure about this, but, as the referee notes, what I say in §6 would make the concession nonfatal. 24   See e.g. Ripstein (2009: 159); Stilz (2009: 51); Pallikkathayil (2010: 137–40). My own presentation in §2.1 followed the example of these.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

48

Tomas Sinclair

since unanimous agreement on the laws and their application is clearly possible in the state of nature. Hence, conceding that the assurance problem can also be solved in the state of nature looks fatal. But unanimous agreement on laws of acquired right and their application is not sufcient to solve the problems of unilateral choice and interpretation. Even if we all agree, and act as if the laws we agree on are in force, and I recognize this in making a claim of right in the state of nature, in doing so I’m still assuming that I can bind others through my act, and expecting others to recognize my acts as binding. I act as if I am a ruler who is particularly attentive to the views and prevailing behavior of her subjects, or who takes the law of acquired right to be sensitive to existing practice. I act as if I occupy an ofcial role even though there are none. Tis might be about the least grievous instance of unilateral lawgiving imaginable, but it’s unilateral lawgiving nonetheless. It assumes that I’m a special source of authority over my fellows by nature. So, idyllic in many respects as the ideal state of nature may be, no acquisition in it can be authoritative.25

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

8. The Power Of Constitutional Norms I am claiming that the existence of a suitable state constitution is both necessary and sufcient, given equal assurance, for action under it to be authoritative. Te existence of a suitable constitution alone sufces because it alone creates suitably defned ofcial roles of authority in the determination of the distribution of external freedom through acquired rights, and the existence of these ofcial roles is what makes it possible to regard their occupants’ judgments as nonrefexively privileged, deployed for the purpose of sustaining a rightful condition by the people as a whole, rather than as imposed by the occupants themselves on the objectionable assumption of their own natural authority. What it takes for there to exist a state constitution is a complex matter of actual practice—of rule-following behavior among the relevant population. Tere must be a legal system, which means that there must be Hartian “rules of recognition” and ofcials who take the “internal point of view” of them (Hart 1994: 56–7, 116–17), for example, as well as appropriately supportive behavior from enough others.26 But I take it that the basic idea is clear enough. 25   For the same reason, Kantians may be able to concede that the Kantian state provides only a better approximation of equal assurance than the state of nature— i.e. that the solution is a matter of degree. I thank an anonymous referee for pointing this out. 26   For further discussion see Raz (1971, 1975); Copp (1999).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Te Power of Public Positions

49

However, this account may seem to give rise to a problem of regress. I argued that even unanimity in belief and practice about laws of acquired right wouldn’t be enough to confer authority on my claims under them. For in making such claims I would expect others to take my claims to be binding with ultimately no explanation of their authority but that I make them. And that objectionably refexively privileges my judgment. But if that’s the problem, isn’t it a problem for the higher-order authority-conferring norms that constitute a state too? After all, I have said that a priori principles do not pick out a unique set of norms conferring authority to determine the frst-order laws of acquired right. So there’s indeterminacy here too, which seems to raise just the same question of the authority to resolve it that arose in the case of the indeterminacy left by a priori principles at the level of frstorder laws of acquired right. Why isn’t anyone’s act in compliance with the authority-conferring norms an act of unilateral lawgiving? If widespread acceptance isn’t enough in the case of the frst-order claims of acquired right, an explanation is needed of why it would be here. Te explanation turns on the fact that acting in accordance with the authority-­ conferring norms is simply a matter of treating someone as authoritative (or, in the case of the authority herself, acting as the role specifes), which is not itself a matter of constraining others’ external freedom.27 If I make a claim of acquired right under circumstances in which there has been no authoritative promulgation of laws of acquired right, then unless I myself am a special source of the authority to constrain, my and anyone else’s actions in accordance with my claim are nothing more than unilateral constraint of others’ external freedom. But when I act in accordance with prevailing authority-conferring norms, even if no higher power has conferred authority on those norms, my actions are not like this. For authority-conferring norms are norms of deference that—unlike frst-order norms of acquired right—purport to authorize no direct constraint on anyone’s external freedom.28 Following them does not in itself involve taking oneself to have title to physically constrain anyone. So the problem that is generated by claims of acquired right in the absence of authoritative promulgation, even when everyone agrees on the laws of acquired right, is not generated by action in accordance with authority-conferring norms. Once the norms are in place—and note that they can have evolved organically, like a language, without anyone’s having imposed them—then ofcial directives and verdicts concerning acquired right do not refexively 27   Except where the roles of legislator or judge are combined with enforcement. But the Kantian state requires separation of these roles for precisely this reason. 28   See Raz’s (1990: 62–5) infuential account of subjection to authority as a matter of treating directives as “exclusionary reasons.”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

50

Tomas Sinclair

privilege the ofcial’s judgment, as we have seen. And claims of acquired right and the enforcement of such claims do not either, since they need presuppose no special authority on the part of the claimants or enforcers, thanks to the prior promulgation of the laws through ofcial directives and verdicts. (Tis is why the promulgation condition is so important.) So no one acts as if she is by nature a special source of authority in the determination of the distribution of external freedom, and everyone’s innate right is respected.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

9. Conclusion By analyzing more closely the problems of the state of nature, I hope to  have cleared a path to vindicating Kantian claims about the state’s capacity to solve those problems, and the need for it to do so. Space constraints and philosophical defciency prevent me from trying to defend the analysis against every objection, and much more would be needed in any case to fully vindicate the Kantian account of political authority. For instance, in expounding the notion of equal innate freedom that is at the heart of the account, Kantians often invoke ideas about being dependent on another’s will as a slave is dependent on a master. But we have seen that what ultimately drives the account is a worry about who gets to resolve indeterminacies in the application of a priori principles of right to the physical world, and that seems at best only a marginal element in  what makes slavery worrying. So it may be doubted whether this is  important enough to vindicate state authority, let alone notorious Kantian conclusions about the impermissibility of revolution in even an unjust state, for example.29 On the other hand, when, laboring under what Rawls (1996: 55–8) calls the “burdens of judgment,” we do disagree about what constitutes the right political expression of our equal freedom—even as we concede that others are not unreasonable in their own claims—it can matter very much how the decision gets made. Whether we can be reconciled to the arrangements that are chosen may depend on whether they are the upshot of a process that can be regarded as recognizing our freedom and equality, especially given that for every political choice that is made between arrangements, there will be people who think that those who prevailed got it wrong.30 In this light, 29   (Kant  1996: 6:320–3). For exposition and discussion see Hill (2002); Flikschuh (2008). 30   See Williams (2005: 13). For the idea that this kind of reconciliation is a key role of political philosophy, see Rawls (2001: 3–4).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Te Power of Public Positions

51

the considerations that drive the Kantian argument do not look so marginal after all. In any case, my account gives us a way to make sense of the faith that Kantians place in the particular kind of rule-following that constitutes the state, and it thereby shows how a Kantian account may escape one of the most signifcant challenges that it faces. On Kantian terms, even if not on any terms, public positions, and only public positions, have the power to realize our equality and freedom.31

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Anderson, E. 1999. “What is the Point of Equality?,” Ethics vol. 109, no. 2, pp. 287–337. Copp, D. 1999. “Te Idea of a Legitimate State,” Philosophy & Public Afairs vol. 28, no. 1, pp. 3–45. Enoch, D. 2010. “Not Just a Truthometer: Taking Oneself Seriously (but not Too Seriously) in Cases of Peer Disagreement,” Mind vol. 119, no. 476, pp. 953–97. Enoch, D. 2013. “Te Disorder of Public Reason,” Ethics vol. 124, no. 1, pp. 141–76. Estlund, D. 2008. Democratic Authority. Princeton: Princeton University Press. Flikschuh, K. 2000. Kant and Modern Political Philosophy. Cambridge: Cambridge University Press. Flikschuh, K. 2008. “Reason, Right, and Revolution: Kant and Locke,” Philosophy & Public Afairs vol. 36, no. 4. Gaus, G. 1996. Justifcatory Liberalism. Oxford: Oxford University Press. Gaus, G. 2015. “On Dissing Public Reason: A Reply to Enoch,” Ethics vol. 125, no. 4, pp. 1078–95. Hart, H.L.A. 1994. Te Concept of Law, Second Edition. Oxford: Oxford University Press. Hill, T.E. 2002. “Questions About Kant’s Opposition to Revolution,” Te Journal of Value Inquiry vol. 36, no. 2–3, pp. 283–98. Hodgson, L.P. 2010. “Kant on the Right to Freedom: A Defense,” Ethics vol. 120, no. 4, pp. 791–819. Kant, I. 1996. Te Metaphysics of Morals, edited by Mary Gregor. Cambridge: Cambridge University Press. Kant, I. 1997. Groundwork of the Metaphysics of Morals, edited by Mary Gregor. Cambridge: Cambridge University Press.

31  I presented these and related arguments at the Nufeld Workshop in Political Teory in Michaelmas 2015, the Oxford Studies in Political Philosophy workshop in April 2016, and the “Kant, Rights, and the State” conference in September 2016. I thank the organizers and participants for their comments. I am especially grateful to Richard Dagger, Cécile Fabre, Adam Kern, Nikolas Kirby, Niko Kolodny, Stefano Lo Re, Michael Otsuka, Tomas Simpson, Sandy Steel, Daniel Viehof, and several anonymous referees for their criticism and encouragement.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

52

Tomas Sinclair

Kolodny, N. 2014a. “Rule Over None I: What Justifes Democracy?,” Philosophy & Public Afairs vol. 42, no. 3, pp. 195–229. Kolodny, N. 2014b. “Rule Over None II: Social Equality and the Justifcation of Democracy,” Philosophy & Public Afairs vol. 42, no. 4, pp. 287–336. Kolodny, N. 2016. “Political Rule and its Discontents,” Oxford Studies in Political Philosophy, vol. 2, pp. 35–70. Kolodny, N. [No date]. “Being Under the Power of Others.” Available online at . Accessed 2 Jan 2017. Korsgaard, C. 1996. Te Sources of Normativity. Cambridge: Cambridge University Press. Miller, D. 1997. “Equality and Justice,” Ratio (new series) vol. X, pp. 222–37. O’Neill, O. 1989. Constructions of Reason. Cambridge: Cambridge University Press. Pallikkathayil, J. 2010. “Deriving Morality from Politics: Rethinking the Formula of Humanity,” Ethics vol. 121, no. 1, pp. 116–47. Rawls, J. 1996. Political Liberalism. New York: Columbia University Press. Rawls, J. 2001. Justice as Fairness: A Restatement, edited by Erin Kelly. Cambridge, MA: Harvard University Press. Raz, J. 1971. “Te Identity of Legal Systems,” California Law Review vol. 59, no. 3, pp. 795–815. Raz, J. 1975. “Te Institutional Nature of Law,” Te Modern Law Review vol. 38, no. 5, pp. 489–503. Raz, J. 1990. Practical Reason and Norms. Oxford: Oxford University Press. Raz, J. 1998. “Disagreement in Politics,” Te American Journal of Jurisprudence vol. 43, no. 1, pp. 25–52. Ripstein, A. 2009. Force and Freedom: Kant’s legal and political philosophy. Cambridge, MA: Harvard University Press. Sangiovanni, A. 2012. “Can the Innate Right to Freedom Alone Ground a System of Public and Private Rights?,” European Journal of Philosophy vol. 20, no. 3, pp. 460–9. Schefer, S. 2003. “What is Egalitarianism?,” Philosophy & Public Afairs vol. 31, no. 1, pp. 5–39. Sharon, A. 2016. “Domination and the Rule of Law,” Oxford Studies in Political Philosophy vol. 2, pp. 128–55. Stilz, A. 2009. Liberal Loyalty: Freedom, obligation, and the state. Princeton: Princeton University Press. Valentini, L. 2012. “Kant, Ripstein, and the Circle of Freedom: A Critical Note,” European Journal of Philosophy vol. 20, no. 3, pp. 450–9. Viehof, D. 2014. “Democratic Equality and Political Authority,” Philosophy & Public Afairs vol. 42, no. 4, pp. 337–75. Williams, B. 2005. In the Beginning Was the Deed. Princeton: Princeton University Press. Wood, A.W. 1999. Kant’s Ethical Tought. Cambridge: Cambridge University Press.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

3

Institutional Legitimacy

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Allen Buchanan

Tis chapter ofers a general theory of institutional legitimacy grounded in an account of the distinctive practical function of legitimacy assessments.1 Part I lays out the account of the distinctive practical function, the Metacoordination view. On this account, legitimacy assessments are part of a social practice that aims at achieving consensus on whether an institution is worthy of our moral reason-based support—support that does not depend solely on the fear of coercion or on a perfect ft between our own interests and what the institution demands of us. Part II makes two main points. First, it uses the Metacoordination view’s account of the practical function of legitimacy assessments to identify criteria of legitimacy that apply to a wide range of institutions, while at the same time allowing for diferences in criteria, depending upon the special function of the institution, and also allowing for variations in the stringency of the criteria, depending upon how badly we need the institution in question. Second, it shows that, for institutions that rule—those that back their rules with coercion—conformity to the requirements of the rule of law is a presumptive necessary condition of legitimacy. Part III explores the role of the virtue of law-abidingness in a viable social practice of making legitimacy assessments and shows how understanding this role explains features of legitimacy assessments that cannot be accounted for without it. Part IV uses the theory of legitimacy thus far developed to identify the inadequacies of two widely infuential views about legitimacy: the attempt to employ Raz’s Service conception of 1   Sometimes the term ‘institution’ is used in a very capacious, if not loose, sense, as when marriage or the French practice of going on vacation in August are said to be institutions. In this chapter, I will use the term in a somewhat more restricted sense, referring to what might be called political institutions, where this is understood broadly enough to cover not only governmental structures but also forms of social organization that include governing structures, even if their primary purpose or function is not government, such as religious institutions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

54

Allen Buchanan

authority as an account of institutional legitimacy and the assumption that a legitimate institution (of the sort that rules) has a claim-right to compliance with its rules. Te Conclusion sets out the attractions of the theory of institutional legitimacy developed in the chapter. I.  The Metacoordination View Of Institutional Legitimacy

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

I.I  Institutional Legitimacy Assessments as a Solution to a Practical Problem What is at stake when questions are asked or assertions are made regarding the legitimacy of an institution? What is the point of engaging in such discourse? Te term ‘legitimacy’ is used in a number of diferent senses and the point of legitimacy discourse can vary, depending upon the context. Te focus of this chapter is on the legitimacy of institutions. Even with this  restriction, it would be a mistake to assume that there is a single, unifed practice regarding the use of the term ‘legitimate’. In some cases, for example, the term ‘legitimate’, as applied to institutions, is used to mean legal or justifed or just. Such usage is unfortunate, because it obscures any distinctive value that legitimacy discourse may possess. In this chapter, I  ofer an account of institutional legitimacy that makes it clear that the term ‘legitimacy’ can have a distinctive sense and value. Without claiming that the concept of institutional legitimacy I articulate provides the only valid sense for the term, I identify an important function that legitimacy discourse can have, defne a concept of legitimacy whose employment best serves that function, and then develop general criteria for determining whether an institution is legitimate according to that concept of legitimacy. My hypothesis is that institutional legitimacy discourse can serve an important function: the achievement of a consensus on whether an institution is worthy of our moral reason-based support; that is, this discourse can help us ascertain whether we should accord the institution a certain kind of respect independently of the fear by which it will coerce us and independently of whether there happens to be a congruence of our own interests with its demands. Tis is a crucial question, because our predicament would be dire if the threat of coercion or the promise of congruence with self-interest were the only resources available for marshaling support for institutions—support they must have if they are to provide the benefts that make them valuable. Relying solely on a perfect ft between self-interest and institutional demands is unworkable, because institutions often rightly require one to act contrary to one’s interests. Relying solely on the threat of coercion would be unacceptable,

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

55

because the amount of coercive power required to render the threat of coercion an efective means of assuring sufcient compliance would make the institution too costly and too dangerous. Reasons that appeal to moral values or principles can provide a basis for supporting an institution that importantly complements the fear of coercion and appeals to self-interest. Let us call this sort of support ‘moral reason-based support’. Te key point is that moral reason-based support enables institutions to function without undue reliance on the threat of coercion or the inducements of self-interest. An institution may function well enough if a sufcient number of people who are able to facilitate or hinder its operations simply support it through unrefective habit. Under these circumstances the discourse of legitimacy may be absent. Te concept of legitimacy tends to perform a distinctive and valuable function when something has occurred that raises the question of how we ought to orient ourselves vis-à-vis an institution. Te perception that the institution is failing to perform its functions or a challenge from a rival institution can make the question of support salient, thereby prompting legitimacy discourse. Empowering an institution with our moral reason-based support can carry great risks: the institution may abuse or squander its power or it may operate in such a way as to cause discoordination, including interference with valuable activities of individuals and with the operation of valuable institutions. So we need to determine what an institution must be like if it is to warrant such support. Whether it does will depend on whether the benefts of supporting it outweigh the risks. According to the Metacoordination view, an institution is legitimate if and only if it is morally worthy of our support and an institution is morally worthy of our support only if the benefts of empowering it outweigh the risks of doing so. On this view, the assertion that an institution is legitimate represents an all-things-considered judgment regarding the moral reasons for supporting it, namely, that those reasons weigh in favor of supporting the institution, in spite of the risks that such support entails. With this understanding of the practical problem which legitimacy assessments address, the basic elements of the Metacoordination view can be outlined as follows. First, at least under reasonably favorable conditions, we should expect more of an institution than that it makes us somewhat better of than the noninstitutional alternative; yet we should not be so demanding as to expect it to be fully just. Full justice is usually too demanding, for two reasons: (1) in our world, it may simply be unattainable and (2) we often need institutions to help us make further progress in justice, so refusing to regard an institution as legitimate unless it is fully just would be self-defeating from the standpoint of justice. Te best would be the enemy of the better.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

56

Allen Buchanan

Second, generally speaking, for an institution to deliver the goods that render it valuable, there must be sufcient moral reason-based support for it: Enough of those who are in a position either to facilitate or to hinder its operation must regard it as worthy of respect, as having the special standing that is generally required if it is to deliver the goods that make it valuable and do so without excessive recourse to coercion. In other words, for the distinctive practical function of institutional legitimacy assessments to be successful in  any particular instance, what might be called a workable consensus, not unanimity, is required. What is needed is that enough people support the institution, on the basis of moral reasons, sufciently to empower the institution to supply its distinctive benefts. Sufcient moral reason-based support reduces the need to rely on coercion and thereby lowers the material costs and the threats to freedom that a massive enforcement apparatus poses. A workable consensus that the institution is morally worthy of support need not be grounded in a thoroughgoing and deep agreement on all relevant moral values or principles. For example, persons who have diferent views about what justice requires may still converge in their assessments of whether an institution is morally worthy of support, if they agree that the institution has certain characteristics and believe that those characteristics count in favor of supporting the institution, even if they hold diferent views about why those characteristics matter. Tird, although diferent institutions supply diferent goods, they all facilitate coordination and their value depends on their doing so. But if they are to play this coordinating role, we need to solve a prior coordination problem: we need to be able to converge in our moral reason-based support on some particular institution. It is not enough that we all recognize that we need an institution; we must coordinate our support on one institution among the alternatives. We need to achieve convergence on judgments that this institution deserves our support, is worthy of our respect. Tus the title ‘Metacoordination view’. Fourth, a proper showing of respect—the appropriate response when an institution is judged to be legitimate—involves diferent behaviors for diferent parties, depending upon their relationship to the institution. For those to whom the institution addresses its rules, it involves a presumption of compliance that operates independently of an assessment of the content of any particular directive. Te disposition to comply is defeasible, because some rules might be so patently and egregiously unjust or so at odds with achieving the proper goals of the institution that they give addressees no reason whatsoever for acting, regardless of how admirable the institution is that issues them. For other parties, respect will simply be a matter of not interfering with the institution’s operations, of according its agents a kind of  impersonal respect, and of proceeding on the assumption that if the

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Institutional Legitimacy

57

institution is fawed, then the proper initial response is to try to reform it, rather than to destroy or bypass it. In the case of rule-addressees, an institution’s legitimacy does not imply that addressees owe to the institution a moral duty of compliance. In other words, on this view, a legitimate institution need not have a claim-right to our obedience—we need not owe obedience to it. Instead, all that is necessary is that rule-addressees have a defeasible, content-independent moral reason to comply.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

I.II  Content-independent, Exclusionary Reasons If the point of the practice is to achieve convergence on moral reason-based support for the institution and if what support requires in the case of those to whom the institution’s rules are addressed is that they operate on the presumption that the rules as such are to be complied with, then the fact that the institution is worthy of support constitutes a content-independent reason for those individuals to comply with its directives. Te contentindependent reasons feature, then, is explained by the Metacoordination view’s characterization of the nature and point of the practice of making legitimacy assessments. Further, this content-independent reason is ‘exclusionary’ in the sense that the judgment that the institution is worthy of our moral reason-based support implies that some reasons for not complying with its directives are inappropriate, that is, that they have no weight at all. In particular, the fact that it would be in one’s best interest not to comply and the fact that one would not be coerced for failing to comply are excluded reasons, and the Metacoordination view explains why they are. Te mere fact that in this particular instance I would do better, from the standpoint of my own interests, by ignoring its rules, or can avoid being penalized for noncompliance, is not a reason for noncompliance, because the point of securing moral reason-based compliance is to enable a valuable institution to function without excessive coercion and without the unsatisfable demand of a perfect congruence between self-interest and what the institution demands. It is important not to misconstrue this point: the claim is not that considerations of self-interest are rendered wholly irrelevant—the fact that compliance would further one’s interest can count in favor of compliance—but rather that the mere fact that one would do better from the standpoint of one’s interests by not complying is not a reason for not complying. Furthermore, the claim that the moral reasons are exclusionary is compatible with acknowledging that the fact that compliance would be extremely costly in terms of one’s self-interest can carry weight as a reason for noncompliance. Tat, too, is compatible with saying that the mere fact that one would do better by not complying

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

58

Allen Buchanan

is  not a reason for not complying. Te Metacoordination view not only accommodates the idea of exclusionary reasons, but also explains why certain reasons are excluded, and this is a point in its favor. As with the content-independent reasons feature, the exclusionary reasons feature is explained, on the Metacoordination view, by its account of the nature and point of the practice of making legitimacy assessments. Fifth, well-founded legitimacy assessments refect a reasonable balancing of the benefts and risks of empowering an institution by giving it our moral reason-based support. By granting institutions the special standing that legitimacy connotes, we empower them so that they can provide us with benefts we could not otherwise obtain. Such empowerment carries the risk that the power will be abused or squandered or will have unacceptably deleterious efects on the activities of individuals or the operation of other valuable institutions. Te practical problem is that we need to develop a shared conception of what an institution must be like to warrant our moral reason-based support, one that refects a reasonable balancing of the benefts of having a functioning institution and the risks that empowering it to perform its functions entails.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

I.III  Preserving the Distinction between Sociological and Normative Legitimacy Although the Metacoordination view takes sociological legitimacy (the widespread belief that an institution is legitimate) seriously and explains why it is important, it is an account of legitimacy in the normative sense. According to the Metacoordination view, an institution is legitimate if and only if it is worthy of our moral reason-based support. It recognizes the importance of achieving sociological legitimacy, but does not confuse it with normative legitimacy.

II. General Criteria For Institutional Legitimacy Suppose we succeed in coming to a shared understanding of what normative standards an institution should meet if we are to run the risks of empowering it through our moral reason-based support. If more than one feasible institution satisfes those standards, then we face two alternatives: we can either allow that each is legitimate, though neither enjoys exclusive legitimacy, that is, neither has the right to prohibit the other from operating in the domain; or, if allowing two institutions to operate in the same domain

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

59

would cause serious problems, we can choose to support one over the other  on the basis of something other than a legitimacy assessment (for example, tossing a coin or throwing our support behind the temporally frst institutional proposal or choosing on grounds of comparative efciency or transition costs). If only one institution is already operating in a certain domain, we need to know whether it measures up—whether it is worthy of the standing that it needs to have to deliver the goods, keeping in mind that thus empowering it carries risks. Yet, as I said earlier, often we cannot reasonably expect it to be fully just. Institutions are typically valuable because they enable coordination. Yet in anything other than the most unfavorable circumstances, to decide whether an institution is worthy of support we need to look to other considerations—not merely whether, were it to be accorded the standing that a judgment of legitimacy implies, it would achieve coordination. So the question arises: are there general criteria whose satisfaction warrants our according institutions forms of respect that they need to have if they are to provide their distinctive benefts and do so without excessive reliance on coercion? Can anything general be said about what features are likely to make such support a good bargain, given the risks of empowering institutions with our support? In an earlier work (Buchanan and Keohane  2006), I argued for an afrmative answer to those questions: across a broad range of diferent kinds of institutions, including those that attempt to rule and those that do not, certain criteria are typically especially relevant to legitimacy assessments, because satisfaction of these criteria is generally a reasonable requirement for an institution to warrant the various forms of support that I noted in Section I. None of them is strictly necessary for legitimacy, but the more of them that are satisfed, and the greater the extent to which they are satisfed, the stronger the case for concluding that the institution is legitimate. 1. Comparative beneft: the institution provides signifcantly better benefts than the noninstitutional alternative. 2. Institutional integrity: there is not a large disparity between the institution’s most important professed goals and procedures, on the one hand, and its actual performance, on the other; or if there is such a disparity there is an explanation of it that shows it to be beyond the institutional agents’ control and allows for a signifcant prospect that the disparity will be reduced over time. 3. Minimal moral acceptability: the institution does not engage in violations of basic rights (or at least violations are not pervasive and there is good reason to believe that efective eforts to reduce violations are under way).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

60

Allen Buchanan

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4. Acceptable origination (or sound pedigree): the institution came to be in a morally acceptable fashion, through an appropriate process. 5. Sound procedures: especially in cases where there is likely to be serious disagreement about outcomes or where outcomes will signifcantly advantage some and disadvantage others, an institution’s procedures should be sound—that is, reliably apt for achieving the institution’s aims—and in particular should be sufciently fair that those who disagree with an outcome can reasonably conclude that the procedures did not stack the deck against them. Generally speaking, only when an institution satisfes these criteria is empowering it likely to be a good bargain, that is, to be worth the risks. Tese general criteria of institutional legitimacy make good sense, then, from the standpoint of the Metacoordination view. With the exception of criterion 1, it would be wrong, however, to construe these criteria as strictly necessary conditions for institutional legitimacy, for one simple reason: how demanding we should be in setting the conditions for empowering an institution by granting it our support will depend upon  how bad the noninstitutional alternative is. If the noninstitutional alternative is awful, it will be reasonable to be less demanding: the expected benefts of the institution will outweigh risks of empowering it that would be unacceptable under more favorable conditions. At least in circumstances that are less than utterly grim, these fve criteria are presumptively necessary for institutional legitimacy. Te acceptable origination criterion may not be satisfed by some institutions that many would deem legitimate. Tis might be true, for example, of some of the frst democratic, rights-respecting states if they originated from violent political changes, or without democratic constitutional foundings. Relaxing this requirement in such cases makes sense because states are so vitally important that it would be unacceptable to withhold our support simply because, at some point in the remote past, they came to be through unjust actions. Moreover, constitutional democracies have sufcient virtues to ofset the worry about their tainted origins; in other words, we have reason to expect that their sullied pedigree will not seriously afect their current performance. Nonetheless, it may be reasonable to expect the creation of new states to be a cleaner process than that by which even the best existing states came to be, and in some cases, extremely unsavory origination may rob a new institution of legitimacy. For some institutions it is not enough that they merely refrain from violating human rights. Given their distinctive roles, they must also act afrmatively to promote human rights. Tis would be true, presumably, not just of the state, assuming that one of its primary justifying functions is to

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Institutional Legitimacy

61

protect human rights, but also of constitutional courts in domestic systems, of supranational tribunals that are explicitly designated as human rights courts (like the European Court of Human Rights), of international criminal courts, and of the International Court of Justice. For other international courts or quasi-judicial institutions, depending upon their distinctive domain of competency, legitimacy might require only nonviolation, not active promotion, of human rights. Over time, if human rights concerns come to penetrate such a domain, as may already be beginning to occur in the case with international commercial law and environmental law, then the legitimacy of the institution may come to require active promotion of human rights, not merely nonviolation.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

II.I  Legality, Lawfulness, and the Legitimacy of Institutions Tat Attempt To Rule Te fve general criteria apply to a wide range of institutions, including those that attempt to rule and those that do not (such as international institutions that merely issue guidelines that serve as coordination devices for states). But in the case of institutions that attempt to rule— that issue rules and attempt to secure compliance with them through the  use of coercion—something else is required, namely, reasonable satisfaction of the requirements of the rule of law. To see why this is so, let us begin with the paradigmatic case of an institution that attempts to rule: the state. Because a distinctive feature of states is that they create and sustain public order through law, their conformity with the requirements of the rule of law is of critical importance for their legitimacy. When backed by coercive power, as in the case of the state, law is profoundly dangerous—unless it is  constrained within the requirements of the rule of law. Satisfying the requirements of the rule of law does not guarantee that the power of institutions that attempt to rule will not be abused or squandered, but it signifcantly reduces the risk that it will be. Further, the protections it provides can be efective across a wide range of types of legal systems and across substantively diferent types of laws within a given system. Te general applicability of the requirements of the rule of law is a refection of the fact that they are tailored to reducing risks that are common to law generally. Some of the requirements, in particular 2. and 7., are also valuable because they enable lawlike systems of rules to facilitate coordination. Lawfulness (conformity to the requirements of the rule of law) is important not just for the legitimacy of states, but also for any institution

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

62

Allen Buchanan

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

that wields signifcant power through ruling. Satisfaction of the requirements of the rule of law helps to prevent institutional power from being misused and in so doing it makes more acceptable the risk of empowering such institutions with our support. Te more statelike an institution is—that is, the more closely its activities approximate ruling in the strictest sense—the greater the peculiar risks that arise from attempts to wield power through rules and hence the more relevant lawfulness is as a requirement of its legitimacy. Te requirements of the rule of law can be satisfed to a greater or lesser degree. Te more statelike an institution is—the more comprehensive its sphere of operation is and the greater the coercive power it wields—the more fully it should satisfy the requirements, other things being equal. However, if all of the feasible alternatives are even worse from the standpoint of satisfying these and other relevant criteria, and if converging in support of this type of institution is morally necessary, an institution could still be legitimate even if it is seriously defcient. Tis conclusion follows, given the fundamental idea that legitimacy assessments, on their best construal, are part of a practically valuable social practice. Tere is considerable consensus that the requirements for the rule of law include the following. 1. Rules are to be sufciently general to reduce the risk that they will be weapons for discrimination or tools for favoritism. 2. Rules are to be sufciently long-standing to facilitate the predictability required for rational planning on the part of individuals and groups. 3. At least in the case of rules that specify criminal ofenses, there is a strong presumption, if not an outright prohibition, against retroactivity. 4. Rules are to be interpreted and applied impartially, with like cases being treated alike. 5. All who are subject to the rules are to have efective access to them for the protection and advancement of their interests. 6. No one is to be a judge in his own case and all are to be accountable under the rules. 7. Rules are to be public, that is, available and understandable, without undue costs, to all who will be held accountable under them. Where these requirements are satisfed, the risks of living within an order in which power is exercised through coercively backed rules are signifcantly lessened. Accordingly, whether these requirements are satisfed is relevant to determining whether an institution that rules is worthy of our empowering it with our moral reason-based support, given the extraordinary risks that such institutions pose. Te Metacoordination view helps to explain why satisfaction of the requirements of the rule of law is presumptively necessary for the legitimacy of statelike institutions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Institutional Legitimacy

63

III.  Legitimacy And The Virtue Of Law-Abidingness William Edmundson (2006) has suggested that an appreciation of the virtue of law-abidingness is important for understanding the authority of law. I want to suggest that the notion of law-abidingness, if this virtue is understood as a disposition to value the rule of law, is crucial both for understanding the legitimacy of institutions that attempt to rule and for the actual achievement of moral reason-based support for such institutions. I will show that the virtue of law-abidingness is important for both normative and sociological legitimacy.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

III.I Law-abidingness A person who has the virtue of law-abidingness has a deep and stable recognition of the importance of the rule of law and is thereby disposed to respect institutions that conform to and promote it, and this recognition is motivationally potent for her. She is committed to living in public order that satisfes the various requirements of the rule of law listed in Section II, though she may not be able to articulate them precisely. At the deepest level of conviction, she is convinced that human beings ought to live together in  a way that recognizes and promotes their responsible agency and the important role that plans play in their lives, on terms of equality and mutual accountability, and that law is crucial for achieving this form of association. But she also understands that if the law is to enable this form of association, it must meet certain standards. Tat is why her commitment is ultimately to lawfulness, not mere legality. Under conditions in which a legal order is either absent or underdeveloped, this commitment will be expressed by her willingness to cooperate with others to create institutions that will achieve the rule of law. Where there is a legal system but it systematically violates the requirements of the rule of law, she will not regard it as worthy of respect. Its laws will not activate her disposition to comply with the law.

III.II  Law and Law-abidingness What distinguishes law is not simply that it achieves coordination through norms that are presented as authoritative nor that it is in principle enforceable, but that it achieves coordination by appealing to the virtue of  law-abidingness. Tis claim is not circular, because the virtue of lawabidingness can be characterized, as I have just done, without reference to the law. All that is necessary is the idea of a public order under rules, along with the notion that the rules must satisfy certain criteria if they are to enable a form of association that publicly acknowledges that we are all

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

64

Allen Buchanan

reason-responsive, plan-making agents who are accountable under those rules on terms of equality.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

III.III  Courts and Legitimacy Tis understanding of the intimate connection between lawfulness and legitimacy in the case of statelike institutions also sheds light on the distinctive contribution that courts can make to the legitimacy of states. In political discourse the legitimacy of a state is often thought to depend, inter alia, on whether it includes an independent judiciary. Te judiciary is independent when it is sufciently immune to political infuence, especially infuence exercised by the state itself, to achieve a reasonable approximation of the requirement that the law is to be interpreted and applied impartially and, above all, that no one (including especially the state) is to be above the law or a judge in her own case. Courts facilitate coordination in a special way, by appealing to the virtue of law-abidingness while protecting and exemplifying it; not just by specifying certain rules that may be liable to enforcement (which any mechanism of unprincipled, predictable coercion could do), but by helping to ensure that those rules, and the judicial processes through which they are identifed, conform to the requirements of the rule of law and are publicly shown to do so. When they operate properly, courts play the crucial role of authoritatively identifying the proper object of the disposition to law-abidingness, while at the same time exemplifying, in their own conduct, what it is to be law-abiding. Tis last point merits elaboration. To be law-abiding is to recognize the moral importance of living together in a social order that satisfes the requirements of the rule of law. It is also to understand that actual legal systems and their constituents, including courts, can conform well enough to the requirements of the rule of law to be regarded as legitimate, and hence to be accorded the standing that legitimacy implies, without their being able to close the gap between morality and legality—that is, without having the power to create moral duties simply by issuing rulings. Because law-abidingness is a moral virtue, a disposition to expect the law to satisfy certain standards whose ultimate importance is grounded in moral values, those who possess this virtue can never fully cede their capacity to ascertain how they ought to act, morally speaking, to a court or to any other legal institution. Te law-abiding person recognizes the importance of the law for morality, but does not confate legal obligation and moral obligation. Her disposition to comply with law (or more generally with rules) is constrained by her sense of justice. When law-abidingness is not thus constrained, it becomes the vice of unrefective obedience and can result in complicity with, or even participation in, grave injustices.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

65

Te virtue of law-abidingness and the distinctive role of courts in a legitimate state are in a kind of symbiosis. Courts, when they operate as they should, aid individuals in their exercise of the virtue of law-abidingness by helping them to understand what being law-abiding requires in particular circumstances; and individuals, to the extent that they have this virtue and therefore possess an independent sense of what it is to be law-abiding, are in a position to evaluate the legitimacy of courts, to determine whether they are doing well enough in protecting, promoting, and exemplifying the rule of law to warrant our moral reason-based support. If an institution is a court, then the proper way of showing respect—of acknowledging the legitimacy of the institution—is to view its decisions as  legally binding. If one possesses the disposition to law-abidingness, one’s recognition that one has a legal duty will create a presumption that one is to comply, independently of the content of the particular legal duty, at least if one assumes that the law that imposes the legal duty is part of a system of law that approximates the rule of law sufciently to be worthy of respect—and if the particular law is not egregiously unjust. For example, if a law arbitrarily stated that certain parties were to be immune to accountability under the criminal law or that a certain group of people have no legal right against enslavement, no law-abiding person would have any reason whatsoever to comply with it, even if it issued from a legitimate legislative process and was approved by a hitherto legitimate court. However, if the particular law was sufciently unjust, the fact the legislature had passed it and the court had approved it would count heavily against either institution’s being legitimate (in both cases one of the fve general criteria for institutional legitimacy, minimal moral acceptability, would be violated). I have suggested that accounts of the legitimacy of states or statelike institutions that make no mention of law, much less of lawfulness, are incomplete because such institutions must at least seriously approximate the rule of law if they are to be legitimate. An adequate account of state legitimacy should capture the intuition that the lack of an independent judiciary is usually a serious challenge to the legitimacy of a state and should accordingly include an account of the distinctive contribution that courts make to the legitimacy of the state. Te present account does this by linking the importance for state legitimacy of lawfulness and the distinctive role of courts in helping to ensure that states rule lawfully, that is, in conformity to the requirements of the rule of law. My account’s emphasis on the connection between legitimacy and the virtue of law-abidingness is also attractive for another reason. It shows how important the virtue of law-abidingness is for the viability of a sound social practice of legitimacy assessments in the case of institutions that rule. Unless

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

66

Allen Buchanan

this virtue is widespread, it may not be possible to secure sufcient moral reason-based support for valuable institutions. Focusing on the virtue of law-abidingness has yet another beneft. It shows why foreigners ought to regard themselves as presumptively bound to comply with the laws of a legitimate state. A presumption of compliance grounded in the virtue of law-abidingness does not depend upon consent or upon any associative ties or any special relationship whatsoever between the individual and the institution. In contrast, consent or associativist theories cannot explain (at least not in any direct and plausible way) why foreigners should feel obligated to comply with the laws of a state that we intuitively regard as clearly legitimate. A theory of legitimacy that includes a role for the virtue of law-abidingness can explain why those who have no special relationship to an institution that rules can nonetheless have a defeasible, content-independent, prima facie obligation to comply with its rules.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

III.IV  Democracy and Legitimacy In an earlier article (Buchanan 2002), I suggested that in democratic states citizens owe each other a defeasible duty of compliance with democratically created laws. Tomas Christiano (2008) holds a similar view and feshes it out more fully. He argues that people living together under a legal order that determines the basic terms of their association owe one another a public recognition of their equal status and that democracy is the only form of government that can enable them to fulfll this obligation. He then concludes that in a democracy each citizen owes the people of the democracy as a collectivity a duty of compliance with the laws. To evaluate this argument adequately would require probing the theory of equality that Christiano develops at length. For present purposes, it will sufce to say that the Metacoordination view is consistent with the thesis that in democracies citizens can owe each other or the people as a collectivity compliance with the laws, while still maintaining that institutions, including the state, can be legitimate in the absence of any claim-right on its part to obedience— that is, without any moral duty of compliance being owed to it. Tey are consistent because, on my account, even if it is true that democracy is a necessary condition of legitimacy in the case of statelike institutions, it is  only a necessary condition where democracy is feasible. If democracy is not presently feasible for international institutions that in some signifcant sense rule, they may nonetheless be legitimate, if they satisfy other relevant legitimacy criteria sufciently well to realize the basic values that make democracy valuable. Two of the most important of these values are physical security and nondiscrimination. As Amartya Sen (1981), Tomas Christiano (2011), and

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

67

others have argued, human rights to physical security tend to be better realized in developed democracies owing to the fact that democratic accountability, achieved through elections and competition among political parties, makes governments more responsive to the basic interests of citizens. Democracies publicly manifest a recognition of equal basic status by according all citizens the right to participate as equals in the most important political processes. Satisfaction of the fve general criteria of legitimacy listed earlier—especially the minimal moral acceptability condition—helps to ensure that governments are responsive to the basic interests of citizens, in the absence of democratic accountability stricto sensu, that is, even if there is no accountability through elections and competition among parties. Satisfaction of the requirements of the rule of law, even in the absence of democracy, contributes to physical security by preventing the most dangerous misuses of the law, while at the same time afrming basic equal status at least insofar as it erects barriers against discrimination and favoritism and ensures that all are accountable under the law and all have access to its resources. Te theory of institutional legitimacy advanced here explains how, under  conditions in which democracy is not feasible, states and statelike institutions can be legitimate, though not democratic, if they sufciently realize the values that underlie democracy to warrant our noncoerced support. So, even if democratic institutions—and they alone—feature a claim-right to obedience, this does not in itself impugn the legitimacy of nondemocratic institutions. Tis is an important conclusion, because it shows that my account can accommodate three intuitions: frst, that for statelike institutions democracy is a necessary condition of legitimacy only where it is feasible; second, that democracies are a unique form of political association in which citizens stand in thick moral relations to one another (having a claim-right on each other or owing obedience to the democratic collective); and third, that even when there is agreement that democracy is infeasible for international institutions, it still makes sense to argue about whether they are legitimate. Finally, my account explains why, even if democracy is a necessary condition for the legitimacy of statelike institutions where it is feasible, it is not sufcient. In principle, a state could be democratic, in the sense of  having competition among political parties and majoritarian electoral processes in which all can participate, and even in the extended sense of also featuring constitutional individual rights, but still fall short in satisfying the requirements of the rule of law. Tis would be the case, for example, if the actual efectiveness of individual constitutional rights for some citizens was undercut by discriminatory interpretation or enforcement of the law. Given the importance of the rule of law in making the formidable power of states

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

68

Allen Buchanan

morally acceptable, signifcant departures from the rule of law can undercut the legitimacy of even a democratic state.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

IV.  Two Rival Accounts Of Legitimacy With the main ideas of the Metacoordination view and its application to the case of institutions that attempt to rule on the table, we can now begin to see why this approach is superior to two infuential alternative accounts of legitimacy, the Claim-Right view and the Razian view. Te former holds that an institution (of the sort that attempts to rule) is legitimate vis-à-vis those to whom it addresses its rules only if they have a moral duty, owed to the institution, to comply with those rules, independently of their particular content (Simmons 1981). Te latter holds that an institution (of the sort that attempts to rule) is legitimate vis-à-vis an individual A if and only if A  would do better, according to the best reasons that apply to her, by complying with the institution’s rules than she would if she did not (that is, than if she determined how to act in any other way) (Tasioulas 2010). Both of these alternative views are presented by those who endorse them as accounts of the concept of legitimacy, on the understanding that this concept applies to institutions. Both are implicitly ofered as the most philosophically illuminating understanding of legitimacy for institutions that attempt to rule and for the state in particular. My aim here is to evaluate them as alternatives to the practical concept of legitimacy that I have articulated, the Metacoordination view. I will seek to determine whether they illuminate what I have identifed as a distinctive practical function or point of institutional legitimacy assessments. My assessment will also proceed on the assumption that although there is a distinction between the concept of institutional legitimacy and the criteria an institution must satisfy to instantiate the concept, an account of the concept ought to provide some guidance as to what the criteria are.

IV.I  Against the Claim-Right View It may be that state ofcials sometimes say that we owe a duty of obedience to the state, but the Metacoordination view shows that the truth of such a claim is not necessary for states to be legitimate. Te state may garner the moral reason-based support it needs in order to provide the coordination and other benefts that make it valuable and do so without excessive reliance on coercion, even if no one has or believes she has a duty, owed to the state, to obey the laws. Sufcient support may be achieved if enough people generally believe they should refrain from interfering with the state’s

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

69

operations and if addressees of the state’s laws believe they have a (nondirected) defeasible, prima facie, exclusionary, content-independent reason to comply. In fact, all that may be required is that enough addressees have a disposition to comply, grounded in the virtue of law-abidingness. Tere are many reasons for compliance with the law, and taken together they can motivate sufcient support for institutions even if no one has or believes she has a moral duty owed to the state. In some cases, there are straightforward moral reasons to comply, as when the law mirrors moral norms. (Te chief reason for complying with a legal prohibition against murder, for example, is that murder is morally wrong.) In other cases, the prospect of penalty may be sufcient, if the costs of compliance are sufciently low. Further, if one has the virtue of law-abidingness and believes that the system of law under which one lives meets certain moral requirements, and in particular that it reasonably approximates the requirements of the rule of law, one will have a strong motivation to obey the law, without believing that one is morally obligated to obey it simply because it is the law. Exercise of the virtue of law-abidingness does not require the belief that any institution has a moral claim-right to obedience, but it can nonetheless contribute to moral reason-based support for institutions. Tere are two other reasons for thinking that the Claim-Right view is too demanding to be of much use in real-world disputes about legitimacy. First, the claim-right requirement is satisfed only (if at all) in the case of democratic institutions, and if (under present conditions) it is infeasible to require international institutions to be democratic, then it follows that no international institution can be legitimate, no matter how well it satisfes what are generally thought to be relevant criteria for legitimacy (such as the fve criteria listed in Section II) and no matter how substantively just its actions are. Tis conclusion should give one pause, because it conficts with the intuition that one can still argue meaningfully about what international institutions would have to be like in order to be legitimate after acknowledging that they cannot presently be democratic. Given how valuable some international institutions are, one should be reluctant to accept a view that rules out the possibility of their being legitimate, simply on the grounds that they are not democratic. One should suspect that the Claim-Right view is not of or for this world. Second, if the standard for legitimacy is set so high as to be unsatisfable in our world, the concept of legitimacy will be incapable of performing its valuable and distinctive practical role. Tat role is to achieve moral reasonbased metacoordination—convergence of support, grounded in moral reasons, capable of empowering institutions to function efectively without excessive reliance on coercion. A conception of legitimacy that is unsatisfable cannot perform this function.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

70

Allen Buchanan

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Of course, as a matter of linguistic stipulation, one may say that an institution that attempts to rule is legitimate only if it can create duties merely by issuing directives or for some other reason has a claim-right to compliance. As far as I can ascertain, however, no one who opts for this usage, including Simmons, has shown why this understanding of legitimacy is most important for political philosophy. In contrast, I have an explanation of why it is not the conception of legitimacy that is of most philosophical interest: it is incapable of guiding a social practice of legitimacy assessments in which the concept of legitimacy can play its distinctive role, the role that distinguishes it from conceptions of justice, on the one hand, and of mere advantage relative to the noninstitutional status quo, on the other. Nor have proponents of the Claim-Right view given us any reason to think that it is often at stake in real-world disputes about the legitimacy of institutions that rule. On the contrary, what seems to matter in realworld disputes is whether an institution ought to be accorded a certain standing—whether it is worthy of respect, based on moral reasons. What respect amounts to will vary, as I have already noted, depending upon which parties are involved. For those to whom the institution addresses its rules, the proper mode of showing respect will be to presume that the institution’s directives are to be complied with, unless they are patently unjust. Tis falls short of taking oneself to have a directed duty to comply, but there is no reason to think that more is required for institutions to  garner the support they need to function efectively without undue reliance on coercion.

IV.II  Against the Razian View According to Raz (1986), A has authority over B just in case B does better, according to the reasons that independently apply to her, by complying with A’s directives than she would by determining on her own how to act. Tasioulas asserts that the Razian view provides the best general conception of institutional legitimacy. Tere are two main problems with the attempt to deploy Raz’s understanding of authority as an account of institutional legitimacy: (1) it is unilluminating, because it yields little insight about the features that make institutions legitimate in the absence of intuitions about those features that we have independently of accepting the Razian view; and (2) it is ill-suited for the practical task for which a notion of legitimacy is distinctively valuable (solving the metacoordination problem) because of its individualistic understanding of legitimacy: an institution whose directives may enable me to act better according to the best reasons that apply to me need not perform the same service for you, and vice versa.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

71

To appreciate objection (1) we need only ask: how does the Razian view help us identify the features of institutions that are relevant to assessments of their legitimacy? If we did not already have an idea of what those features are, how would the Razian view enlighten us? In the absence of a comprehensive account of what the reasons are for all agents (an account which, I suspect, no one possesses), we must rely on speculation as to which reasons are relevant to the assessment of institutional legitimacy. But to the extent that such speculation yields plausible results, it relies implicitly on intuitions about what makes for institutional legitimacy that we have independently of and prior to our introduction to the Razian view, not upon the Razian view. In words, we must already have some idea  of which sorts of reasons are relevant to assessments of institutional legitimacy. Te problem is that the idea of ‘best reasons’ does no work in helping us to determine what we should require of an institution if we are to confer that standing on it that legitimacy connotes; it sheds no light on what an institution should be like if the risks that accompany empowering it with our uncoerced support are acceptable. It does not even indicate that what is at issue is whether the benefts of empowering the institution are worth the risks. Tere is one possible exception to the conclusion that the Razian view provides no independent insight into what characteristics an institution needs to have to be legitimate. Te Razian view does perhaps suggest that a legitimate institution will be one whose policies are substantially informed by input from experts. Tat would make sense, because an institution having that feature would, other things being equal, enable those who follow its directives to do better than they would were they to rely on their own (nonexpert) judgment. But beyond that, the Razian conception appears to shed little light on criteria for institutional legitimacy. Further, expertise is relevant, presumably, only insofar as it contributes to the satisfaction of other, more basic legitimacy-making features, and about these the Razian account is silent. Whenever someone proposes an alternative account of institutional legitimacy which, like the Metacoordination view as I have feshed it out here, specifes some intuitively plausible substantive criteria for institutional legitimacy, the proponent of the Razian view may say, ‘Well, if those are the proper criteria, then one will in fact do better, according to the best reasons that apply independently to one, by complying with the directives of an institution that satisfes those criteria. For example, if acceptable origination or institutional integrity or minimal moral acceptability are criteria of institutional legitimacy generally or if conformity to rule-of-law requirements is a criterion of legitimacy for statelike institutions, that is because one does best, according to the reasons that independently apply to one, if one complies with the rules of institutions that satisfy those criteria. One’s best

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

72

Allen Buchanan

reasons will refect the importance of those criteria for determining whether the institution is legitimate. So the Razian view is vindicated.’ Rather than rebutting my claim that the Razian view provides little guidance, this reply confrms it. Criteria of legitimacy are identifed independently of any consideration of the Razian view and then, if they are deemed acceptable on independent grounds, or enjoy strong intuitive plausibility, the Razian view simply puts the stamp of approval on them and claims them for its own. In other words, the notion of applicable reasons is an empty shell that provides no guidance as to what makes an institution legitimate. If one has an independent notion of what the legitimacy-making features of an institution are, then one can fll the shell, but the shell adds no value. It contributes nothing of substance to an account of legitimacy to say ‘such and such makes an institution (of the statelike type) legitimate and, oh, by the way, if it is legitimate, then one will do better according to all the reasons that independently apply to one if one takes its directives as content-independent reasons for acting.’ Most of us, if we are honest, will admit that we don’t know what all the reasons are that apply to us across the board, though we are likely to have some idea of what sorts of reasons are relevant to assessing the legitimacy of institutions before we ever encounter the Razian view. Te good news is that we don’t have to know what reasons apply to us across the board; all we have to know is what sorts of characteristics institutions need to have if they are to be worthy of our moral reason-support, given the point of the practice of making legitimacy assessments. Suppose the Razian replied as follows: ‘Your frst objection trades on  a  failure to distinguish between the concept of legitimacy and the criteria for legitimacy. Te Razian view characterizes the concept, not the criteria’ (Tasioulas 2010: 99). Perhaps, but the problem is that the Razian characterization of the concept provides no guidance for what to look for in criteria. In contrast, the Metacoordination view characterizes the concept in  a way that helps us identify the criteria: It tells us to look for criteria which, if satisfed, would make the institution worthy of our moral reasonbased support, given that we must balance the benefts of empowering it with such support against the risks of thus empowering it. Given that we know (through bitter experience) a good deal about what the risks of empowering various types of institutions are, this tells us quite a lot. For example, it tells us that for statelike institutions, those that wield substantial power through coercively backed rules, the criteria for legitimacy should be more demanding than for institutions that do not wield coercive power. And in directing us to identify the peculiar risks of empowering that type of  institution, it points toward the conclusion that satisfaction of the requirements of the rule of law is required for their legitimacy. Once we understand the practical function of legitimacy assessments, the problem of identifying the criteria for legitimate institutions is (thankfully) diferent

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

73

from and much more manageable than the problem of determining what one’s best reasons are. Now for objection (2): Te Razian view does not capture the point of the social practice of making public legitimacy assessments, namely, to try to arrive at some shared standards that institutions are to satisfy if they are to receive the kind of morally based widespread support that they need to have to achieve coordination and produce other benefts, without excessive reliance on coercion. On the Razian view, legitimacy is a strictly individual matter. Terefore, there is no reason to think that if  complying with an institution’s directives will enable me to do best according to the reasons that independently apply to me, the same will be true for you or for others. Te Razian might nonetheless say that her view is compatible with a recognition that the practical importance of legitimacy assessments lies in the possibility that a social practice of assessments can yield sufcient consensus to achieve moral reason-based coordinated support for valuable institutions. She could say that whether a particular individual should regard an institution’s directives as authoritative will depend, in part, on whether it is likely that enough others will do so, because the achievement of coordination is of great practical importance for all of us and because coordination requires that most of us, at least, regard the institution’s directives as authoritative. Notice, however, that if enough others take the institution’s rules as authoritative to enable the institution to deliver the relevant goods, then the individual may plausibly conclude that this will occur whether she takes the rules as authoritative or not. Whether the individual’s decision is determined by the desire to free-ride or simply by an impartial desire to avoid costs of compliance that would contribute nothing to successful coordination, she will conclude that, according to her own best reasons for acting, it is not the case that she should regard the institution’s rules as authoritative. So, it appears that the individualistic character of the Razian account of legitimacy cannot accommodate the idea that legitimacy judgments function to achieve principled, coordinated support of institutions. One fnal diference counts in favor of the Metacoordination view: it can explain the exclusionary reasons feature; Raz’s view assumes it without explaining it.

V. Conclusion Te Metacoordination view of institutional legitimacy has several advantages. 1. It explains how legitimacy assessments can play a vital practical role— that of achieving the coordinated moral reason-based support that is

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

74

2.

3.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4.

5.

6.

Allen Buchanan needed to allow institutions to facilitate benefcial coordination without excessive recourse to coercion. In so doing, the Metacoordination view explains (i) what sociological legitimacy is (namely, widespread moral reason-based support, as opposed to support based solely on the fear of coercion or on self-interest) and (ii) why sociological legitimacy is so important (it enables us to reap the benefts of institutions without excessive reliance on coercion), but (iii) it does both of these things without confating sociological and normative legitimacy. It explains why the ability to facilitate coordination is an important contributor to an institution’s legitimacy without suggesting that coordination is sufcient for legitimacy. It emphasizes that, except in the bleakest of circumstances, we can often reasonably demand more of an institution than that it achieves coordination and the benefts that depend on coordination. It provides a unifed account of legitimacy that applies to a wide range of institutions, including those that rule and those that do not, but in such a way as to acknowledge that the full set of criteria for legitimacy will vary depending upon the function of the institution. It explains the intuition that legality is important for legitimacy, at least in the case of statelike institutions, but without making the mistake of thinking that mere legality confers legitimacy. Unlike some other recent philosophical accounts of legitimacy, it does not ignore the role that lawfulness and hence the law play in legitimacy, but at the same time does not reduce legitimacy to mere legality, that is, to conformity to whatever the law happens to be, because it acknowledges that it is important for the law to satisfy the requirements of the rule of law. It explains the intuition that whether a state is legitimate can depend, not just on whether it is democratic, but also upon whether it has an independent judiciary, because it illuminates the distinctive role of courts as guardians of the rule of law and recognizes the importance of the rule of law for state legitimacy. It explains why we should take the directives of legitimate institutions seriously; that is, why we should presume that they are to be complied with (unless their content is patently unjust or patently at odds with the proper functioning of the institution). And it does this without implying that we have a content-independent directed moral duty to  comply, owed to the institution or its agents. It can also accommodate the view, however, that in democratic states citizens morally owe one another, or the people as a collectivity, obedience to the laws.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

75

7. It shows that the Claim-Right view, as a general requirement of legitimacy for institutions that rule, ought to be rejected for two reasons: because, at least for institutions for which democracy is not feasible, it makes a successful practice of legitimacy assessments impossible under realistic conditions, and because such a practice can get along quite well without such a demanding notion of legitimacy. 8. It shows what is unsatisfying about the attempt to construe the Razian understanding of authority as an account of institutional legitimacy: that understanding provides little independent insight into what the substantive criteria for legitimacy are because it ofers no account of the point of the practice of making legitimacy assessments; and its individualistic character does not acknowledge the distinctive social role of legitimacy assessments as attempts to converge on shared understandings of what institutions must be like if they are to warrant our uncoerced support. Without bringing in the notion of the point of the practice, there is no guarantee that an institution that satisfes Raz’s criterion for legitimacy for me or for you will be worthy of our support. But if the notion of the practice and its point are available, the Razian criterion adds little of substance to our understanding of how to make legitimacy assessments. 9. In emphasizing the role of the virtue of law-abidingness in assessments of institutional legitimacy, the Metacoordination view has a signifcant advantage over any theory that defnes legitimacy in terms of a special relationship, such as consent or associative ties, between individuals and institutions that rule. Unlike such theories, the theory provided here explains the intuition that foreigners should regard the laws of a legitimate institution as presumptively to be complied with. If an institution is worthy of moral reason-based support, then anyone capable of complying with its rules has a defeasible, contentindependent reason to comply, regardless of any special relationship. Individuals who possess the virtue of law-abidingness will acknowledge this reason and act accordingly. 10.  It accommodates the fact that the legitimacy of an institution sometimes depends upon its relationships to other institutions, including relationships of reciprocal legitimation, because those relationships can be relevant to whether the institution warrants our moral reason-based support. In contrast, theories that understand legitimacy solely in terms of the relationship between an institution and  those within its sphere of operation cannot account for the phenomenon of reciprocal legitimation. For example, it has been argued that states contribute to the legitimacy of international courts by creating them through treaty-making (one way of satisfying the

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

76

Allen Buchanan

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

acceptable pedigree criterion of legitimacy) while international courts in turn contribute to the legitimacy of states by providing an impartial determination of the law that binds them, thus preventing a situation in which states would be judges in their own case (Buchanan 2016). Nor, consequently, can theories that understand legitimacy solely in  terms of how institutions relate to individuals subject to them explain  the fact that institutional legitimacy is often an ecological afair, that is, that in many cases the legitimacy of an institution cannot be assessed without understanding its relationship with other institutions (Buchanan 2013). 11. It accommodates the intuition that, where democracy is feasible, institutions that rule must be democratic to be legitimate, while also accommodating the institution that even though it is not currently feasible for international institutions to be democratic, it still makes sense to argue about whether they are legitimate. I will conclude with three fnal points about the highly practical, thoroughly social conception of legitimacy I have outlined in this chapter. First, on this view, the legitimacy-making features of an institution are not, as it were, out there, to be discovered by normative analysis that pays no attention to the requirements of achieving sociological legitimacy. Instead, it explains why achieving sociological legitimacy is important from a normative standpoint, namely, because it enables us to empower institutions so that they can provide their distinctive benefts without giving them dangerous levels of  coercive power. Accordingly, the proper task of the philosopher who  theorizes institutional legitimacy is not to ignore the importance of achieving sociological legitimacy concerning valuable institutions, but to articulate the distinctive practical role and social character of legitimacy assessments and help clarify and render more accurate our intuitions about how demanding we should be in determining whether to accord an institution the various forms of respect that it must enjoy if it is to deliver the benefts we believe we cannot achieve noninstitutionally and do so without excessive reliance on coercion. Second, my view lies between two extreme views of the legitimacy of statelike institutions: on the one hand, a very weak notion which identifes legitimacy with merely being justifed in wielding power, where being justifed means simply not acting wrongly (having a mere Hohfeldian liberty-right); and on the other, the very strong Claim-Right view. Elsewhere I have noted that the Weak view fts some uses of the term ‘legitimacy’ (Buchanan  2002). Te Weak view, however, does not attach any uptake condition to legitimacy: it does not imply that when an institution is legitimate, we are morally required to show it respect, for example by not interfering with it or by presuming that its directives are to be complied

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Institutional Legitimacy

77

with, or by frst trying to reform it if it is fawed, rather than destroying or abandoning it. Te Weak view tells us only that if an institution is legitimate, its agents are justifed in attempting to rule; it says nothing about how others should respond. Yet surely in many cases, when the term ‘legitimacy’ is used, an uptake condition is implied. In fact, as I have argued, legitimacy assessments can only function as part of a social practice that can solve the practical problem that makes the concept of legitimacy distinctively valuable if it implies an uptake condition, because moral reason-based support is a kind of uptake. Te Weak view cannot accommodate such uses of the term. Tus my account of the practical function of legitimacy assessments explains exactly why the Weak view is unsatisfactory. Te Claim-Right view remedies this defciency of the Weak view, but at a prohibitive cost: it holds that if an institution is legitimate, then not only is it justifed in ruling but also a very robust uptake is required on the part of those it addresses, namely, they ought to regard themselves as having a moral duty to obey its rules, regardless of their particular contents. I have suggested that at most there is only one kind of institution that satisfes the claim-right requirement: democracy, if Christiano’s account is correct. If this is so, then the Claim-Right view  entails that no nondemocratic institutions, no matter how just they  otherwise are and no matter how well they satisfy other criteria for legitimacy, can be legitimate. Tat is an unpalatable conclusion, at least in the case of institutions we desperately need and for which democracy is not presently feasible. My view lies between these extremes. Unlike the Weak view it includes an uptake condition, specifying a range of respectful behaviors whose appropriateness depends on the relationship between the individuals in question and the institution. But unlike the Claim-Right view, it does not hold that the only proper uptake on the part of rule-addressees is to regard themselves as having a moral duty to comply, owed to the institution, independent of content. Tird, the view of institutional legitimacy ofered here is both variable and dynamic. It is variable in the sense that it allows that less-demanding criteria for legitimacy may be appropriate if we need an institution very badly, because in those circumstances the risks of empowering the institution with our uncoerced support may be lower than the risks we face if we withhold our support because the institution does not measure up to some higher standard. It is dynamic in the sense that as conditions improve, it will be appropriate to demand more of an institution, if it is to be worthy of our uncoerced support.2 2  I am grateful to Samantha Besson, Tomas Christiano, Julian Culp, Andrea Sangiovanni, Bas Van der Vossen, and Christopher Wellman for valuable comments on a draft of this chapter.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

78

Allen Buchanan

References

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Buchanan, Allen, ‘Political Legitimacy and Democracy,’ Ethics 112 (2002): 689–719. Buchanan, Allen, Te Heart of Human Rights. Oxford: Oxford University Press, 2013. Buchanan, Allen, ‘Te Legitimacy of International Courts,’ unpublished paper (2016). Buchanan, Allen and Robert O. Keohane, ‘Te Legitimacy of Global Governance Institutions,’ Ethics and International Afairs 20 (2006): 405–37. Christiano, Tomas, Te Constitution of Equality. Oxford: Oxford University Press, 2008. Christiano, Tomas, ‘An Instrumental Argument for a Human Right to Democracy,’ Philosophy & Public Afairs 39 (2011): 142–76. Edmundson, William A., ‘Te Virtue of Law-Abidance,’ Philosophers Imprint 6 (2006): 1–21. Raz, Joseph, Te Morality of Freedom. Oxford: Oxford University Press, 1986. Sen, Amartya, On Famines: An Essay on Entitlement and Deprivation. Oxford: Oxford University Press, 1981. Simmons, A. John, Moral Principles and Political Obligations. Princeton: Princeton University Press, 1981. Tasioulas, John, ‘Te Legitimacy of International Law,’ in Samantha Besson and John Tasioulas (eds.), Te Philosophy of International Law, 97–118. Oxford: Oxford University Press, 2010.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

PA RT I I

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

E G A L I TA R I A N I S M

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

4.

‘(Luck and Relational) Egalitarians of the World, Unite!’ Kasper Lippert-Rasmussen

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. INTRODUCTION In recent years, egalitarian political philosophy has been marred by a family dispute, often fought with the intensity of, well, family disputes.1 For years, luck egalitarianism was dominant. Ten in 1999 came Elizabeth Anderson, who wrote an extremely infuential critique in which she also proposed an alternative account of egalitarian justice: social relational egalitarianism (for short, relational egalitarianism).2 Generally, luck egalitarians have dismissed Anderson’s critique, arguing, for instance, that it is pitched at a lower level of abstraction than luck egalitarianism (Arneson 2000, 345; Cohen 2008, 271; but cf. Cohen 2013, 193). Such responses have been only partly successful, and the number of egalitarian political philosophers who are relational egalitarians and reject luck egalitarianism appears to have increased. Believing that there is truth to both sides, this chapter explores the core of the diference between two views and how exactly they are incompatible. I also propose 1   Cf. ‘[i]f much recent academic work defending equality had been secretly penned by conservatives, could the results be any more embarrassing for egalitarians?’ (Anderson 1999, 287). Te ‘academic work’ Anderson refers to is that of luck egalitarians. I defne luck and relational egalitarianism in §2. 2   Predecessors of Anderson’s article that make analogous criticisms are David Miller (1995; 1998; cf. Arneson 1995) and Jonathan Wolf (1998; cf. Hinton 2001; Firth 2013; Wolf  2010). Wolf argues that egalitarians care not just about distributions, but also about respect. Miller favors a similar pluralist position, inspired by Michael Walzer’s notion of spheres of justice (cf. Anderson 1999, 326). Interestingly, neither Wolf nor Miller thinks that the ideal of equality is exhausted by a certain view of justice, and both appear to think that a concern for social relations refects broader concerns about the good society. Important relational egalitarian works following Anderson’s  1999 article include Anderson 2008; 2010; O’Neill 2008; Scanlon 2003, 204–18; Schefer 2003; 2005; 2015; and Schemmel 2012.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

82

Kasper Lippert-Rasmussen

a way of settling some part of the dispute, or at least elucidating that this part is a family dispute, as it were. Section 2 thus defends a certain view about what constitutes the core of the diference and disagreement. Section 3 responds to the reductionist challenge that relational egalitarianism can be construed, without loss, as a subspecies of distributive egalitarianism that has a certain view on the proper equalisandum that focuses on social goods.3 I defend a guarded reductionist view, claiming that substantively, though not formally, the disagreement between luck egalitarians and relational egalitarians can be construed as a disagreement about the proper equalisandum and thus might be seen as a family dispute within distributive egalitarianism. Te next two sections address two challenges to this claim. Section 4 addresses Schefer’s recent account of how the two views are diferent. I argue that the view that he describes is not best construed as a relational view, but rather should be seen as a third view—that is, dispositional egalitarianism. Section 5 addresses the objection that luck egalitarianism is concerned with responsibility in a way that relational egalitarianism is not, and §6 sets out an egalitarian position that incorporates insights from both views. Section 7 argues that this ecumenical egalitarian position—‘ecumenical’ in the sense that it incorporates signifcant aspects from luck, relational, and dispositional egalitarianism—is not essentially disjunctive but can be grounded in fairness, understood as the claim that inasmuch as individuals are diferently situated—whether in terms of how well of they are or in terms of how they are treated relative to others—this is unfair, unless the relevant diference is motivated by their diferential exercise of responsibility.4 Section 8 concludes. First, however, I ofer one preliminary clarifcation: By ‘justice’ I mean a state in which everyone provides others with what they owe them. Tis way of construing the right theory of egalitarian justice is congenial to relational egalitarians like Elizabeth Anderson, since this notion is close to her contractualist conception of justice and diferent from the common luck egalitarian use of ‘justice’ to refer to a state in which comparative fairness is fully realized, perhaps irrespective of any exercise of agency. However, along with Vallentyne (2015, 40–9), I shall assume that a theory of what we owe one another can be based, at least in part, on an account of comparative fairness (cf. §7).   Luck egalitarianism is another subspecies of distributive egalitarianism.   If the ecumenical view can be grounded in fairness, this strengthens my reconciliatory ambition. Moreover, assuming a refective equilibrium view of justifcation, it also strengthens the justifcation for egalitarianism given that the grounding of the diferent egalitarian subprinciples in the overarching value of fairness increases the coherence of the proposed egalitarian view relative to a pluralistic version of egalitarianism that ascribes value to an equal distribution, egalitarian relations, and egalitarian dispositions separately. 3 4

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

83

2.  THE CORE DISAGREEMENT

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

In this section, I identify what I take to be the core diference and disagreement between relational and luck egalitarians. I place a lot of weight on two salient statements by Anderson and Schefer. Te need for identifying a core diference is due to the fact that luck and relational egalitarian theorists have several disagreements, not all of which are logically tied to one another. Tese are thus best resolved separately. Hence, the conjecture is that some of these disagreements are not refective of whether these theorists are luck egalitarians or relational egalitarians, but rather are  issues which could as well be discussed as intraluck egalitarian or intrarelational egalitarian issues. Here is how Anderson (1999, 288–9) contrasts two views about the aim ‘of egalitarian justice’: luck egalitarianism on the one hand (or ‘equality of fortune’ in her terminology below), versus her own relational view, democratic equality, on the other: democratic equality . . . is a relational theory of equality: it views equality as a social relationship. Equality of fortune is a distributive theory of equality: it conceives of equality as a pattern of distribution. Tus, equality of fortune regards two people as equal as long as they enjoy equal amounts of some distributable good—income, resources, opportunities for welfare, and so forth. Social relations are largely seen as instrumental to generating such patterns of distribution. By contrast, democratic equality regards two people as equal when each accepts the obligation to justify their actions by principles acceptable to the other, and in which they take mutual consultation, reciprocation, and recognition for granted. Certain patterns in the distribution of goods may be instrumental to securing such relationships, follow from them, or even be constitutive of them. But democratic egalitarians are fundamentally concerned with the relationships within which goods are distributed, not only with the distribution of goods themselves.  (Anderson 1999, 313–14)

Anderson’s point that ‘equality of fortune’ entails an instrumental view of social relations applies—if it applies to luck egalitarianism (see §3)—to any view that takes justice to be concerned exclusively and noninstrumentally with the distribution of goods. Hence, the efective target of Anderson’s critique is much broader than its ofcial target, i.e. luck egalitarianism. It applies to views that, unlike luck egalitarianism, ascribe no signifcance to people’s exercise of their responsibility, e.g., outcome egalitarianism, and it applies, mutatis mutandis, to views that, unlike luck egalitarianism, do not take equality to be the distributive pattern mandated by justice. For instance, according to resource sufcientarians, justice is satisfed when, in a distribution involving two people, both of them have enough resources, whatever their social relations are.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

84

Kasper Lippert-Rasmussen

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

In any case, Anderson’s main point is that luck egalitarianism focuses on the distribution of goods, while relational egalitarianism focuses on the nature of social relations. Admittedly, her formulation is guarded in that she claims only that luck egalitarians take a ‘largely’ instrumental view of social relations—implying that she does not rule out that in part they may take a noninstrumental view—and that relational egalitarians might think that certain distributions of goods may be ‘constitutive of social relations’ and that they are fundamentally concerned ‘not only [emphasis added] with the distribution of goods’. However, in the interest of a clean cut, I shall ignore these qualifcations (cf. Anderson 2010, 1–2). Tis would be problematic if Anderson provided reasons for not making a nonqualifed contrast, or if she provided examples of luck egalitarians who in part take a noninstrumental view of social relations and of relational egalitarians who think distributions are constitutive of suitably egalitarian social relations or fundamentally important. However, she does not.5 Because luck egalitarians are concerned with distributions, Anderson submits that they fail to address the objectionable inegalitarian nature of oppressive social relations. For instance, they fail to note how the unfreedom of gays and lesbians ‘to publicly reveal their identities without shame or fear’ (Anderson 1999, 320) is a concern of egalitarian justice, because the relevant unfreedom is not a matter of inequality in terms of resources or other privately owned ‘divisible goods’ (Anderson 1999, 319). Schefer draws a contrast between the ideal of distributive equality and relational egalitarianism which is similar to Anderson’s: According to [distributive equality, the most infuential version of which is luck egalitarianism: KLR], equality is an essentially distributive value. We can directly assess distributions as being more or less egalitarian, and justice requires that we strive to achieve fully egalitarian distributions, except insofar as other values forbid it. Tis is the view taken by Jerry Cohen when he says, ‘I take for granted that there is something that justice requires people to have equal amounts of, not no matter what, but to whatever extent is allowed by values that compete with distributive equality’6 . . . According to the . . . relational view, equality is an ideal governing certain kinds of interpersonal relationships. It plays a central role in political 5  Some theorists suggest that certain kinds of distributive equality, e.g. an equal distribution of ‘the power to make decisions about the society’s common activities’, is a feature of the social relations between members of this society (Norman  1998, 40; cf. Brighouse and Swift 2014, 27). 6   If ‘values that compete with distributive equality’ are not justice values, then it is puzzling, given Cohen’s pluralism, how non-justice values can determine what justice requires. Hence, these values must be justice values. If so, luck egalitarianism does not, in Cohen’s view, exhaust the value of justice—a conclusion which motivates the fact that my defnition of luck egalitarianism below only gives a sufcient condition for injustice (cf. Parft 1998, 3; Temkin 1993, 13, 17).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

85

philosophy because justice requires the establishment of a society of equals, a society whose members relate to one another on a footing of equality. For those who accept this view, one important task is to consider what kinds of institutions and practices a society must put in place if it is to count as a society of equals. Te relevant institutions and practices will include those that govern the distribution of goods within society, and so the ideal of equality, understood as an ideal that governs the  relations among the members of society, will have important distributive implications. But, according to this view, equality is a more general, relational ideal, and its bearing on questions of distribution is indirect  (Schefer 2015, 21–2)

(cf. Anderson  2008, 142; Schefer  2003, 21–2; Schemmel  2012; Tan 2012, 97–9). On that basis, in part, I defne luck and relational egalitarianism as follows:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Luck egalitarianism:  It is unjust if some people are worse of than others through no responsibility of their own. Relational egalitarianism:  It is unjust if some people do not relate to each other as equals, and it is not the case that it is unjust if some people are worse of than others, not even when that is through no responsibility of their own.

Te two views are incompatible by virtue of the second conjunct—the negative relational claim—in the statement of relational egalitarianism. Call this incompatibility the core disagreement. Luck egalitarianism and the frst conjunct—the positive relational claim—are consistent though distinct claims. Call the diference between those two claims the core diference. One might wonder why luck egalitarianism only states a sufcient condition for injustice and does not include a negative conjunct comparable to the one included in the defnition of relational egalitarianism (cf. Miklosi 2018). I delineate luck egalitarianism in this way, frst, because this is the way luck egalitarianism is normally delineated (cf. Temkin  1993, 13). Second, Patrick Tomlin (2015, 154; cf. Lippert-Rasmussen 2015b, 87–98) has shown that there are elements of relational egalitarianism present in the position of the paradigm luck egalitarian, G. A. Cohen. If Tomlin is right and Cohen was not inconsistent on this matter, this is an exegetical reason to delineate luck egalitarianism in the way I do.7 7  Admittedly, one could take the non-standard view that Cohen is not a luck egalitarian, but rather holds an interesting hybrid position. Moreover, some luck egalitarians might think that unequal social relations are unjust if, and only if, they result in unjust inequality of welfare etc. (Arneson 2000, 341–2). Such luck egalitarians might add that, as a matter of fact, hierarchical social relations tend to result in objectionable distributions. However, in possible worlds quite diferent from ours, where hierarchical social relations are a required means of bringing about an egalitarian distribution, hierarchical social relations are preferable to egalitarian social relations. Te strength of the arguments below does not hinge on the present issue.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

86

Kasper Lippert-Rasmussen

At this point it is relevant to introduce the reductionist challenge to relational egalitarianism. According to this challenge, whenever people do not relate to one another as equals, then there is some good that is unequally distributed, and whatever is unjust about the former can be expressed in terms of a claim about what is unjust about the latter. If so, relational egalitarianism is reduced to a subspecies of distributive egalitarianism, which focuses on a particular kind of equalisanda—those goods that derive from, or are constituted by, the existence of certain social relations.8 Moreover, relational egalitarianism as delineated above is incoherent, since it implies that it is unjust if some people are worse of than others in terms of the pertinent good which relational egalitarians care about. If so neither the core disagreement—relational egalitarians do think that some distributive inequalities are unjust—nor the core diference—the positive relational claim might be equivalent to a version of the luck egalitarian claim (cf. §5)—exists. Call this the reductionist challenge to relational egalitarianism. Section 3 expands upon and assesses this challenge.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.  A RESPONSE TO THE REDUCTIONIST CHALLENGE To support the reductionist challenge, consider a two-person case where X and Y relate to one another. We can then defne a certain good—the good of social standing (cf. Miller 1995, 206)—as a good which X would have more of than Y insofar as X relates to Y as Y’s superior; less of insofar as Y relates to X as X’s superior; and equal amounts of insofar as they relate to each other as equals. We can then add that distributive equality is perfectly satisfed if it is true of all pairs of people in a certain society that the relevant two individuals have equal amounts of social standing.9 It seems that, essentially, this sort of move transforms a concern for the egalitarian nature of social relations into a concern for the distribution of a certain good, 8   Tere is no comparable reductionist challenge to luck egalitarianism, since this view applies even in situations with no social relations (cf. Arneson 1999, 226). Te reverse situation—that people have social relations, but no distributive issues arise between them—is not possible. At least one theorist, Ronald Dworkin, who is normally seen as a paradigm luck egalitarian, believes that distributive inequality is unjust only when it obtains between citizens and, thus, refects failure on the part of the state to treat its citizens with equal concern and respect. Accordingly, his view does not apply to situations without social relations. 9   Egalitarians are not committed to there being precise and elaborate measures of the amount of unjust inequality. Te view I sketch here is very simple and incomplete, but this does not mean that it is not a version of the ideal of distributive equality.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

87

i.e., social standing. If generalizable, any concern for social relations can be transformed into a concern for the distribution of a certain good. In an early response to the reductionist challenge, David Miller argued that equality of status obtains

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

when each member of society regards him- or herself as fundamentally the equal of all others, and is regarded by others as fundamentally their equal. It should be obvious that status in this sense is not a good that can be directly distributed equally. Tere is nothing one can hand out to individual people in the way that one can hand out titles to income or property. Nor does it seem likely that there will be some other good Y, the appropriate distribution of which will lead to equality of status.  (Miller 1995, 199)

In Miller’s view, equality of welfare is a distributive ideal of equality despite the fact that welfare, e.g., in the sense of preference satisfaction, cannot be ‘directly distributed’. Hence, his argument for not regarding equality of status to be a distributive ideal hinges entirely on whatever arguments can be built upon the last claim in the quoted passage; yet such arguments are, as I shall now argue, implausible. First, there is no empirical asymmetry between equality of status and equality of welfare. Just as there is no good such that a suitable distribution thereof will result in equality of status, there is no good such that a suitable distribution thereof will lead to equality of welfare. Moreover, he mentions ways in which very unequal distributions of certain goods, e.g., money, are likely to lead to inequality of status (Miller  1998, 34). Tis implies that less unequal distributions of money are likely to lead to less inequality of status. Second, an ideal could be distributive in nature even if it is unlikely that there is any good which we can distribute in such a way that the ideal is realized, e.g., equality of welfare if our access to the minds of others is indirect and imperfect at best.10 Miller’s reasoning presupposes that whether an ideal counts as distributive depends on what we are able to do, and not on the intrinsic nature of the ideal. Tird, even if whether an ideal is distributive depends on what we are able to achieve, Miller’s implicit requirement is too strong. Instead, we might suggest that an ideal is distributive if there is some good which is such that some distributions thereof are likely to result in less inequality (of status) than are other distributions. 10   Tere is a sense of ‘distribution’ on which only material things that can be handed out physically and directly can be distributed (cf. Young 1990, 8–9; but see Gheaus 2016, 3). Tis is not the sense of ‘distribution’ that welfarist luck egalitarians care about.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

88

Kasper Lippert-Rasmussen

Schefer is another relational egalitarian who has responded to the reductionist challenge, which he puts (in somewhat diferent terms) as follows: [I]t may seem that the relational view, if fully spelled out, must itself take a distributive form. For suppose that the members of society are committed to the ideal of a society of equals and are determined to structure their relations in accordance with that ideal. How could they go about doing this? Te answer, it may seem, is that they would take care to ensure that certain important goods, such as status, power, or opportunity, were distributed equally within the society. Tat is what it would mean for them to achieve a society of equals. But if that is correct, then the relational view is not really an alternative to the distributive view but rather a version of it.  (Schefer 2015, 22)

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

If the reductionist challenge is sound, then relational egalitarians cannot succeed in challenging distributive egalitarianism per se, since their view is simply a version thereof. Tis is not to say that luck and relational egalitarian theorists do not disagree, or to say that the former are right on those issues about which they disagree; but, surprisingly, their disagreement falls within the scope of the much-discussed distributive ‘equality of what?’ question. Tis result is interesting, even if it is not prejudicial to how all of the disagreements between luck and relational egalitarians should be resolved.11 Tere is a way in which the reductionist challenge can be answered and in which the concern for the nature of social relations is diferent from a concern for the nature of distributions. To explain it, I need to take a closer look at relations between individuals. Some relations between individuals are not social, e.g.:12 __is taller than__ __has more hedonistic pleasure in her life than__ __lives longer than__ Since relational egalitarians are concerned with the egalitarian character of social relations, presumably, they are not concerned as such with the nature of these other types of relations. Examples of social relations are, e.g.: __has authority over__ __oppresses__ __defers to__

11   It does bear on some of the disagreements. If the reductionist challenge is vindicated, there is pressure on relational egalitarians to explain why the distribution of goods other than the good of social standing, say, is irrelevant to justice. 12   For X and Y to be in a certain social relation, it has to be the case (but is not sufcient) that they interact with one another or that X interacts with Z who, at an overlapping time interval, again interacts with Y, and so on and so forth. Even if this is quite an undemanding condition—e.g. it can be satisfed by people who are chainconnected through a series of partly overlapping pairs of interacting persons—it might still be true of two persons not so related that one is taller than the other, etc.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

89

Relational egalitarians are concerned with the nature of social relations. Tey want these relations to have a suitably egalitarian character.13 No doubt there is a large family of diferent views as to what exactly is the nature of egalitarian social relations, but they are diferent—obviously— from relational antiegalitarians, who believe that social relations should have a suitably hierarchical nature, e.g., in the way that Edmund Burke (1987 [1790]) believed that the nonvirtuous lower classes should defer to the nobility. Relational egalitarian views are also diferent from the views of relational nonegalitarians, who do not care about whether social relations are egalitarian or hierarchical, but think that justice requires that they have some alternative features, e.g., that they are informed by the common good. Tey also difer from the views of social relations sufcientarians, e.g., people who think that while justice does not require that people relate to one another as equals, their social relations should be such that in any social relation all parties are treated in such a way that their social standing does not fall below a certain threshold—defned in absolute or relative terms. Simplistically put: relational sufcientarians think justice is incompatible with slavery, but compatible with the relation between a political leader and a deferential follower, who greatly admires and trusts the leader, provided that their relation involves a minimum level of mutual respect. Undoubtedly, there might be other relational views, e.g., relational prioritarians.14 For present purposes, my main point, for which the relatively clear ideas of relational sufcientarianism and so on will do, is that relational egalitarianism is one member of a broader view of social relationalism, albeit the only one that attracts the attention of contemporary political philosophers. Some relations may be trickier to categorize as social or nonsocial than the samples above, e.g.: __is wealthier than__

On the one hand, this seems not to be a social relation, since X and Y might never interact with one another, so ‘being wealthier than’ is not a feature of 13   Suppose we adopt an objective list account of well-being (Grifn 1986, 40–55). If so, it seems plausible that having reciprocal, egalitarian social relations to others is noninstrumentally valuable, e.g., because ‘deep personal relations’ are non-instrumentally relevant to our well-being and, at least in the case of social relations between adults, depth is in part constituted by equality (cf. Tomlin 2015, 164). 14   Burke believed that social relations are not a zero-sum matter—hierarchy, instead of egalitarian social relations, gives those who end up at the lower ends of the hierarchy the chance of virtuous modesty and deference and that might be better for them, social relationswise, than some bland form of equality, which homogenizes and vulgarizes everything: ‘the order of civil life establishes [the true moral equality of mankind, i.e. natural hierarchy: KLR] as much for the beneft of those whom it must leave in a humble state, as those whom it is able to exalt to a condition more splendid’ (Burke 1987, 124; cf. 170).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

90

Kasper Lippert-Rasmussen

their social relations to each other. We can imagine that X is an Inca king in the Americas prior to Columbus and Y is a peasant in medieval Europe. On the other hand, it is impossible for X to be wealthier than Y without there being certain social relations between X and other individuals in X’s society, on the one hand, and social relations between Y and other individuals in Y’s society, the diference between which constitutes X’s greater wealth and Y’s lesser wealth.15 What we should say at this point is that the relevant relation is a nonsocial relation; yet, unlike ‘is taller than’, it is a relation that exists between persons in virtue of their possession of a good—wealth—which is inherently social in nature. In other words, there is a distinction between, on the one hand, the distinction between social and nonsocial relations and, on the other hand, the distinction between relations between individuals by virtue of their possession of social goods and relations between individuals by virtue of their possession of nonsocial goods. Typology of possible objects of egalitarian concern

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

c. Social relations d. Nonsocial relations

a. Social goods

b. Nonsocial goods

X has a superior rank to Y X has more wealth than Y

NA X is taller than Y

Nothing prevents distributive egalitarians from being concerned with the distribution of inherently social goods, i.e., what goes on in the lower left-hand box. However, the real issue, one might argue, is the nature of  the relations with which luck egalitarians and relational egalitarians are concerned, and here the claim is that the relation which friends of the distributive view are concerned with is not a social relation. Tat threeplace relation: ‘__and__have equal amounts of__’ is not as such a social relation, though it may pertain to social, as well as nonsocial, goods.16 Tis diference seems to sufce as a basis for rejecting the reductionist challenge, since relational egalitarians are interested in the nature of social relations, not in the abstract, triadic nonsocial relation of ‘__and__have equal amounts of__’, whether applied to social or nonsocial goods. Tis rebuttal of the reductionist challenge, however, establishes nothing of substance. It merely establishes a formal diference between the distributive 15   Cf. Cohen’s (2011, 182) view that a ‘sum of money is tantamount to (≠ is) a license to perform a disjunction of a conjunction of actions’. 16  Cf. Cohen’s (2011, 3) remark in the quote from Schefer above that ‘there is something which justice requires people to have equal amounts of ’.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

91

and the relational views. However, for any social relation there is a pair of a  corresponding nonsocial relation and social good such that the relevant corresponding distributive egalitarian and social egalitarian views will condemn or endorse exactly the same situations. Take, for instance, one of Elizabeth Anderson’s (1999, 318) favorite examples of an injustice which she thinks luck egalitarians cannot accommodate because of their focus on the distribution of goods: namely that of gays and lesbians who, unlike heterosexuals, do not enjoy the ‘ability to appear in public without shame, and without being ascribed outcast status’. Tis injustice no doubt involves a particular set of social relations. However, it can also be characterized as a matter of an unequal distribution across homo- and heterosexuals of the relevant presentational ability which Anderson mentions. Tis ability is a good, in the sense that it is desirable to have it and undesirable not to have it. Also, although you cannot distribute that good in the way you can distribute money or land, like welfare it is a good that one can have more or less of. Indeed, that this state of afairs obtains seems to be Anderson’s central complaint about the situation. Moreover, the relevant objectionable social relations could not exist in the absence of an unequal distribution of the relevant presentational skills, just as this inequality could not exist in the absence of the relevant homophobic social relations. Hence, at least in the case of one of Anderson’s paradigmatic examples of an injustice, we can naturally translate it into an injustice in terms of the distribution of a certain good. Take next the relation of ‘__exploits__’, which may initially seem more resistant to a similar analysis. Tis is a social relation which has an objectionable inegalitarian form according to relational egalitarians (cf. Cohen 2011, 5; Norman 1998, 44). Tis is the frst view—the relational egalitarian view. However, we can stipulatively introduce a certain good—call it exploitation standing—such that, ceteris paribus, if X exploits Y and Y does not exploit X, then X has more of the good of exploitation standing than Y does. We can also extend the notion to cover cases where X is exploited by a greater number of people than Z is exploited by; cases where they are exploited by the same number of people but to unequal degrees; cases where X exploits Y in a major way and Y exploits X in a minor way, e.g., introducing exploitation standing relative to diferent dimensions; and to cases where X exploits some people at the same time as X gets exploited by others, etc.17 We might then say that the relation we are interested in is the abstract triadic relation ‘__and__have equal amounts of__’—a relation which, owing to its abstract nature, is not a social relation—where the particular 17   While ‘exploitation standing’ is a term of art, it is natural to say, e.g., that members of one group are exploited more than members of another group, i.e. a group of people who exploit but do not get exploited themselves.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

92

Kasper Lippert-Rasmussen

value of the third relatum, ‘__’, that we happen to be concerned with is the  social good of exploitation standing. Tis is the second view—the distributive egalitarian view. Te diference between these two views appears merely formal—at least, the two views overlap, extensionally speaking. Tis suggests that while the reductionist challenge can be answered, formally speaking—there is an analytical distinction to be drawn between the two kinds of views, and relational egalitarianism as I stated it in §2 is not incoherent—the nature of  this distinction is such that it is morally irrelevant. More generally, substantively speaking, any relational egalitarian view is equivalent to a certain form of distributive egalitarianism pertaining to a social good, defned on the basis of the relevant social relation.18 Admittedly, the fact that, in substance, relational egalitarianism is identical to a form of distributive egalitarianism does not show that, in substance, it is identical to distributive luck egalitarianism. In short, while the reductionist challenge might undermine the core disagreement, it does not undermine the core diference. For reasons that will become apparent in §5, however, I do not think this reply weakens the force of the reductionist challenge. Some might concede that the distinction between an inequality in a social relation and an inequality in a nonsocial relation pertaining to the relevant corresponding social good is morally irrelevant, but deny that this casts light on the nature of relational egalitarianism on the following ground: What, at a fundamental level, relational egalitarians object to is not that people do not relate to one another as equals overall. Tey object to any unequal social relation between individuals, independent of whether it is counterbalanced by an unequal, reverse-direction social relation. Relational egalitarians subscribe to a deontological view proscribing relating to others as unequals, and that view is not extensionally equivalent to any view to the efect that individuals should have equal amounts of a certain kind of social good.19

18  Te present way of drawing the distinction between relational and distributive egalitarianism rests heavily on a distinction between relations that are social and relations that are not. Above I simply relied on paradigm cases that fall in either of the two categories. For present purposes all I need to rely on is the claim that some relations between persons are social and others are not without having to submit any claims about the nature of these relations. 19   A diference challenge—the Walzerian challenge—says that relational egalitarians do not just think that, from an overall perspective, social relations should be egalitarian. Rather, or in addition to that, they think that there are diferent spheres of justice and that within each of these spheres people should relate to one another as equals. Tis challenge, however, brings up a diferent distinction from the one that is at stake here. In principle, luck egalitarians could embrace the idea of diferent spheres of justice and

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

93

In response to this challenge, I concede the truth of the previous sentence, but deny its signifcance. First, the distinction of signifcance here is the deontological/non-deontological distinction, not the social relations/ distributions distinction. One way of seeing this is to remind oneself of a comparable deontological luck egalitarian norm: ‘Do not act in ways that will result in someone being worse of than others through no responsibility of their own’. Tis view, too, is not extensionally equivalent to a certain view enjoining that everyone has equal amounts of some good, whether social or not. Second, the nonextensionally equivalent relational view is implausible, since surely counterbalanced unequal social relations are less bad, from the point of view of relational equality, than noncounterbalanced unequal social relations, when all else is equal. Tird, most relational egalitarians are  concerned with inequality of overall social standing, not with the inegalitarian nature of each and every social relation. David Miller, for instance, submits that relational equality does not require that ‘people should be equal in power, prestige, or wealth . . . What matters is how such diferences are regarded, and in particular whether they serve to construct a social hierarchy in which A can unequivocally be ranked as B’s superior’ (Miller 1998, 31; cf. Anderson 1999, 312; Schefer 2003, 36). My claim is simply that such a concern for the absence of hierarchical social relations is extensionally equivalent to a concern that people have equal amounts of whatever goods hierarchical social relations essentially involve, and that whatever nonextensional diference might exist between these two views is morally irrelevant.20 I do not claim, in general, that there can be no morally relevant diferences between extensionally equivalent views. Rather, I argue by way of a counterchallenge here. Tat is, I challenge anyone to explain the morally relevant diference between rejecting hierarchical social relations in the contend that within each no one should be worse of in terms of the relevant good specifc for that sphere through no responsibility of her own. 20   Accordingly, if, on the ground that there is a deontological, relational egalitarian constraint against exploiting anyone, someone objects to a non-hierarchical situation where X exploits Y, Y exploits Z, and Z exploits X in a way that leaves the three persons involved equally well of in terms of any possible equalisandum, the scope of my substantive reductionism does not extend to this view. Nor does its scope extend to a similar view regarding acts whose objective meaning is that some individuals have an inferior status (cf. Anderson and Pildes 2000; Hellman 2008; Schemmel 2012). In the latter case, I doubt, however, that the view is best construed as a form of relational egalitarianism. After all, one can act in ways the objective meaning of which is that a certain group of people, with whom the agent has no social relations, has an inferior status, e.g., a state refuses to accept responsibility for a tragically ‘successful’ genocide (= no present descendants of the group subjected to genocide with whom the state interacts socially) it committed almost a century ago.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

94

Kasper Lippert-Rasmussen

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

way that Miller and others do and an unequal distribution of whatever good hierarchical social relations consist of. Specifcally, I do not see the force of the objection: ‘True, if people have hierarchical social relations, then they have unequal amounts of social standing, but what makes the situation unjust is the former, not the latter’. I maintain that to be at the lower end of a social hierarchy simply is to have less social standing than others have. One cannot ask: ‘Suppose people’s relations were hierarchical and yet they had equal amounts of social standing. Would that be in any way unjust?’ By defnition, if people have equal amounts of social standing, their relations are not hierarchical. Finally, I fnd it difcult to conceive of the possibility that someone is superior to someone else in terms of their social relations, but yet there is no good in virtue of which the former is better of than the latter. Presumably, there is a diference between having a position in a web of social relations which is superior to that of others and simply having a position which is diferent from that of others, and the best way to account for that distinction appeals to some good that the former implies inequality of—the one in the superior position has more of that good than someone in an inferior position—and which the latter does not. Te Secretary of State has a superior position in the Department of State compared to an intern. Tis is due to the fact that she has much more of the good that we normally call ‘authority’ than the latter has. Te fact that some people are ‘above’ or ‘below’ others implies an ordering, which in turn sufces for the notion of distribution being applicable. I tentatively conclude that, in a substantive sense, the reductionist challenge is vindicated. 4.  SCHEFFLER’S DELIBERATIVE CONSTRAINT In response to my cautious endorsement of the reductionist challenge in §3, some might suggest that we can do better and that relational egalitarianism cannot be (substantively) reduced to a view about distribution. In an insightful elucidation of the relational egalitarian ideal, Schefer takes his point of departure in a simple personal relationship between two people—a marriage—and asks what it is for such a relation to be a relation between equals. One component of such a relationship draws on values other than equality. So, for instance, an egalitarian relation is one in which individuals treat one another with respect and each sees the other as a moral agent with the rights and responsibilities accruing to moral agents.21 Another, and 21   I put less emphasis on this component than Schefer does. One could combine the other and ‘distinctively egalitarian component’, which I address in a moment, with values

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

95

distinctively egalitarian, component is what Schefer dubs the egalitarian deliberative constraint: If you and I have an egalitarian relationship, then I have a standing disposition to treat your strong interests [understood broadly to include the person’s needs, values, and preferences] as playing just as signifcant a role as mine in constraining our decisions and infuencing what we do. And you have a reciprocal disposition with regard to my interests. In addition, both of us normally act on these dispositions. (Schefer 2015, 25)22

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Schefer (2015, 28–9) explicitly notes that satisfying the deliberative constraint is compatible with reaching decisions that do not leave the parties, who relate to one another as equals, equally well of.23 Indeed, he thinks it is unlikely that participants in egalitarian personal relationships will ‘attempt to satisfy the [deliberative] constraint through the self-conscious application of a fxed distributive formula’, though, for reasons he does not expound, the constraint will ‘exert pressure in the direction of egalitarian distribution’ (Schefer 2015, 33, 34). Conversely, even if a strict distributive formula of equality were continuously satisfed, the ideal of relating to one another as equals would fail to be satisfed if one of the parties to the relationship continually fouted the deliberative constraint. Tis connects with a more general point that he makes; to wit, that equality, as he construes it, is ‘a form of practice rather than a normative pattern of distribution’ (Schefer 2015, 31).24 Hence, we cannot, as the reductionist challenge presupposes, helpfully defne a certain good—say, social standing—as something that, all other things being equal, two people have equal amounts of if, and only if, they relate to one another as equals, and as one that one person has more of other than the ones Schefer espouses and still hold a combined view which is no less relational (or, as I shall argue, dispositional) egalitarian than Schefer’s. Hence, these diferences between such views and Schefer’s are not really of the essence in addressing the reductionist challenge per se. Also, one could similarly tie luck egalitarianism to the values that Schefer mentions, e.g., reciprocity and mutual respect (cf. Schefer 2015, 24–5), or other values for that matter. 22   Some might deny that the deliberative constraint can be part of an account of what we owe one another—what I took to be my topic in this chapter—on the ground that we do not owe one another to be disposed to act or refrain from acting in certain ways. But whether this is so is a substantive issue, not one that is resolved by the concept of what we owe one another. 23   Tis is compatible with luck egalitarianism, which, like relational egalitarianism, does not focus on distributive outcomes per se. Te more difcult question is whether the deliberative constraint is satisfed in a situation where all parties have the relevant dispositions, but some people have better opportunities than others of having their interests etc. promoted. 24   Tis passage suggests a fourth way—a practice-focused view—of construing the ideal of equality—one which I do not discuss in this chapter.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

96

Kasper Lippert-Rasmussen

than the other person to the extent that the former relates to the latter as  superior and the latter relates to the former as inferior (cf. LippertRasmussen 2015b, 195–6). Doing so would leave out the ‘deliberative and practical dimensions’ of relational equality. More generally, it shows that there is a deep and genuine diference between distributive views of equality and those which relational egalitarians subscribe to. Schefer’s attempt to rebut the reductionist challenge fails for two reasons. Te frst reason is that, however it is intended by Schefer, the deliberative constraint is not best seen as a constraint on the nature of social relations. My claim is not that, given his description thereof, Schefer’s deliberative constraints cannot reasonably be understood to pertain, in part at least, to social relations. My claim is that the deliberative constraint so interpreted is less plausible than it is when interpreted as one that pertains to dispositions independent of social relations. Consider the following two questions: frst, can the constraint (on its most plausible construal) be satisfed in the absence of social relations; and, second, can it be violated in the absence of social relations? If the answer to both questions is ‘yes’, this vindicates my main claim in the previous paragraph. It is natural to think that the form of the constraint is a conditional, where the antecedent says that X and Y are socially related and the consequent then states what features this relationship must have to be an egalitarian one. If so, the antecedent is false if X and Y have no social relationship, and, thus, the constraint is trivially satisfed in the absence of social relations. An additional and more substantive point is that even if X and Y are not socially related, we might still think that it is valuable if they are disposed in the way that the deliberative constraint highlights.25 Suppose that X and Y both live alone on desert islands. One day Y’s autobiography and a computer with Internet access wash ashore on X’s island. A similar thing happens on Y’s island. Tey read about each other and set out to chat with each other through Facebook, starting, so they think, to form all sorts of social conventions and to coordinate plans. Both believe that they interact with one another, but in fact they both interact with a sophisticated communications robot. Hence, they believe that they have social relations with each other, but in fact they are as solitary as they always were. My contention is that, disposition-wise, this is no less objectionable than an otherwise comparable situation in which they do chat with each other on Facebook and, thus, do succeed in forming social relations. Accordingly, if 25   We might either think that their having inegalitarian dispositions is disvaluable, or we might think (additionally) that their having egalitarian dispositions is valuable.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

97

so, on its most plausible version (which then is not the conditional-style interpretation suggested in the previous paragraph), the deliberative constraint can be violated even in the absence of social relations. It might be replied that there is no egalitarian reason to care about people’s dispositions in the absence of their having social relations and that this shows that the constraint cannot be violated in the absence of social relations.26 Hence, the constraint really is a constraint on the nature of social relations—for it to apply to a set of people, they must be socially related. Hence, if they are not socially related, the deliberative constraint is neither violated nor satisfed, in the same way as some would argue that if there is no King of France, then ‘Te King of France is bald’ and its negation are neither true nor false. We can strengthen this interpretation by noting that, as Schefer describes the constraint, the object of the relevant dispositions, which we normally act on, is what we do and decide, and that this might be understood to mean what we do and decide together. Surely, doing and deciding together is a way of relating to one another, and my two Robinson Crusoe-like characters do not act or decide anything together, even though they falsely believe that they do. In response, however, note frst that at least some of the features of egalitarian social relations that Schefer thinks make them valuable can also be instantiated in my scenario of people who falsely believe they have social relations. For instance, according to Schefer, hierarchical social relations ‘distort people’s attitudes toward themselves, undermining the self-respect of some and encouraging an insidious sense of superiority in others’ (Schefer 2005, 19). Tese defects might arise in my social-relationsfree, Robinson Crusoe-like scenario. Tus, part of what makes Schefer care about egalitarian relations makes it unmotivated, by Schefer’s own lights, to restrict the scope of the deliberative constraint to settings involving social relations. Second, it is odd to present a constraint that pertains to people’s dispositions as a constraint on social relations. True, people who are disposed in ways compatible with Schefer’s constraint are likely to have relations of a certain egalitarian kind (just as it is unlikely, but not impossible, that they will form suitable egalitarian relations even in the complete absence of egalitarian dispositions). However, strictly speaking, the constraint constrains people’s dispositions, not their relations, even if we assume its scope does not extend beyond situations where people are socially related. By way of analogy: a constraint saying that if people are socially related, they should have equal amounts of resources does not constrain people’s social relations, even if it applies only when they are socially related, and it would be odd to 26

  My initial formal point supports this claim as well.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

98

Kasper Lippert-Rasmussen

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

appeal to this constraint in an account of how relational egalitarianism difers from the distributive ideal. By way of further support for this claim, consider having a peaceful relationship. It is much more likely that X and Y have such a relationship if they are peacefully disposed, but it is conceptually possible for them to be non-peacefully related even if they are peacefully disposed, e.g., if they misunderstand each other’s intentions, and it is possible that their relations are peaceful even if they are belligerent, e.g., if through sheer luck no occasion for confict arises. Te criteria for relations being peaceful do not include anything about the dispositions of the involved parties. Arguably, the same is true about having an egalitarian relationship. Having certain egalitarian dispositions is not part of what defnes egalitarian relations, even if the latter is very unlikely to obtain in the absence of the former. Two hierarchically disposed people, who through sheer luck always treated each other’s interests as equally signifcant whenever deciding what to do together, should not conclude, once they learn about each other’s dispositions, that, despite appearances, they did not relate to one another as equals. Rather, they should simply infer that the egalitarian relationship which they managed to have was an extremely fragile one. I conclude that the deliberative constraint is not really a constraint on the nature of social relations. However, Schefer’s constraint is not reducible to a form of distributive egalitarianism. In the attempt to show that the distributive and the relational views are distinct, Schefer has in efect brought to our attention a third view, dispositional egalitarianism. Tis view is neither about distributions nor about social relations, but holds that justice is a matter of how individuals are disposed to act:27 Dispositional egalitarianism: It is unjust if some people (who are suitably related) are not disposed to relate to others in a way that respects the deliberative constraint.28

Dispositional egalitarianism is diferent from, but also compatible with, both distributive and relational egalitarianism. While I think dispositional egalitarianism is interesting and might capture part of what egalitarians are concerned with—concerns that might not be reducible to a concern for 27   Tis view has an honorable history and Hume’s (2004, 83–98) view that justice is a(n artifcial) virtue strikes me as a potential (non-egalitarian) predecessor (cf. Cohen’s 2008, 317–18 example of the ‘remarkably just person’, who would rather forgo an advantage than unjustly enjoy more than do others who are no less deserving). Anderson also makes some—to my mind, confusing—remarks to the efect that for relational egalitarians justice is fundamentally a virtue of agents (Anderson 2010, 22; cf. Miklosi forthcoming, 22). 28   Te ‘suitably-related’ qualifcation refects that Schefer does not think that the deliberative constraint governs all social relations.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

99

distributions or the nature of social relations (whether the latter is substantially reducible to the former or not)—the existence of a third distinct egalitarian view cannot show that relational egalitarianism is not reducible to a particular form of distributive egalitarianism. My second reason for thinking that Schefer’s response to the reductionist challenge is unsuccessful derives from the fact that we might defne what seems to be a luck egalitarian deliberative constraint which is diferent from, but in certain ways similar to, Schefer’s deliberative constraint:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Te luck egalitarian deliberative constraint: If you and I have an egalitarian relationship, then I have a standing disposition to treat the concern that you are not worse of than I am, in terms of the satisfaction of our strong interests for reasons not suitably refecting our diferential exercise of responsibility, as constraining our decisions and infuencing what we do. You have a reciprocal disposition with regard to my interests. In addition, both of us normally act on these dispositions.

Perhaps Schefer would say that while this constraint will often constrain agents in the same way as his own deliberative constraint does, it will sometimes constrain us diferently; and when this happens we see—perhaps for reasons relating to how people can live in an egalitarian marriage without wanting to level down—that his deliberative constraint is a more plausible constraint than the luck egalitarian deliberative constraint. However, my present concern is not to discuss which of the two constraints is the more plausible one. Rather, it is this: the fact that there is a luck egalitarian deliberative constraint shows that the issue between luck and relational egalitarians—or, strictly speaking (despite Schefer’s belief that what he ofers is an indirect argument for relational egalitarianism), dispositional egalitarians—is not whether the locus of egalitarian justice is distributions or dispositions. After all, luck egalitarians might have luck egalitarian views on both matters. Rather, the disagreement concerns which distributions (if any) and which dispositions justice requires. Hence, I conclude that Schefer’s response to the reductionist challenge is unsuccessful, and therefore now turn to a diferent way in which my response to the reductionist challenge might be resisted.

5. RESPONSIBILITY As the negative relational claim refects, much of the early work of relational egalitarians focused on the way in which luck egalitarians think that distributions should refect the diferential exercise of choice and responsibility (cf. Tomlin 2015, 155). Take, for instance, Elizabeth Anderson’s version of

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

100

Kasper Lippert-Rasmussen

the harshness objection, according to which luck egalitarianism is committed implausibly to requiring that victims of bad option luck—e.g., a motorcyclist who deliberately chose to take the risk of driving without a crash helmet uninsured and now, due to a trafc accident, will die on the roadside if unassisted—sufer the bad consequences of their choices (Anderson 1999, 295–302). Te role of responsibility also plays a large role in Anderson’s critique of how luck egalitarianism is committed to making and expressing intrusive and demeaning views regarding people’s exercise of their choices (Anderson 1999, 289). Similarly, much of Schefer’s early work on luck and relational egalitarianism focuses on showing that the limited role Rawls (1971) ascribes to responsibility is incompatible with an interpretation of A Teory of Justice as being a proto-luck egalitarian work (Schefer 2003, 24–31) and, more generally, with establishing that ‘the project of developing a responsibility-based conception of egalitarian justice is misconceived’ (Schefer 2005, 5). In Schefer’s view, ‘the presence or absence of choice’ simply cannot plausibly have the ‘make-or-break signifcance’ that luck egalitarians ascribe to it (Schefer 2003, 19). An objection to my discussion of the reductionist challenge so far might be that while it shows that relational egalitarianism is substantively reducible to a form of distributive egalitarianism, it does not show that it is reducible to distributive luck egalitarianism. Because, and only because, of the latter’s view of the signifcance of responsibility, the core diference and the core disagreement persist. I have three responses to this challenge. First, it is striking that the very contrast between luck egalitarianism, on the one hand, and relational egalitarianism, on the other hand, is heterogeneous in the sense that the two terms invoke two diferent ways of cutting the cake. ‘Luck egalitarianism’ involves a cut between those egalitarian views for which luck plays a role and those for which it does not. Te view that justice requires that everyone should have equal levels of welfare however they exercise their responsibility is one such view. Tis view is not luck-ist, but it is not relational either. As noted in §2, this view is open to many of the complaints that relational egalitarians direct against distributive views in general. ‘Relational egalitarianism’ involves the cut between those views according to which the locus of egalitarian justice is social relations and those views according to which it is something else, e.g., the distribution of certain goods or dispositions.29   In efect we have four possibilities: Luck egalitarianism has two separate elements: luckism and distributive egalitarianism. Anderson and Schefer reject both, but one can mix and match, so various hybrids are possible. I thank Richard Arneson for a helpful formulation of this point. 29

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Non-luckist distributive egalitarianism Luckist distributive egalitarianism

101

Relational egalitarians Anderson (ofcial version)

Non-relational egalitarians Outcome egalitarians

Relations ought to be informed by the luck egalitarian deliberative constraint

Luck egalitarianism

Tese two distinctions cut across one another. As I shall argue in my third response below, there are diferent accounts of what it is for social relations to have a suitably egalitarian nature, but, surely, some of those accounts imply that responsibility plays an important role, e.g., I do not treat my co-citizens as equals if I constantly make negligent choices and demand that they pick up the tab for me, ignoring how this takes resources away from their pursuit of their life-plans. I would exploit them if I were to proceed on this basis. Hence, the issue of responsibility is simply orthogonal on the issue of social relations versus distribution, and settling the issue of responsibility still leaves us with the issue of the site of equality. My second response to the responsibility challenge is that, despite the fact that relational egalitarians have criticized luck egalitarianism for its focus on responsibility, it is not as if relational egalitarians think that responsibility plays no role in a society in which social relations have a suitably egalitarian character. For instance, Anderson believes that justice does not require that citizens who violate the law should have ‘efective access to the conditions of their freedom at all times’ (Anderson 1999, 289; cf. Wolf 2010, 348).30 Similarly, Schefer (2003, 6) is open to consequential responsibility playing a justifed role—though one which is much more restricted than that ascribed to it by luck egalitarians—in a society of equals. Accordingly, luck and relational egalitarian theorists do not disagree about whether responsibility should play any role. Teir disagreement concerns the type and importance of that role. More specifcally, if the reductionist challenge is true, there are  some social goods which are such that relational egalitarians think inequality without diferential responsibility would be unjust while diferential responsibility might justify unequal distributions. My third response is that it is simply an open question what role considerations about responsibility should play when social relations have a suitably egalitarian character (cf. Schefer 2003, 31). One contender says that individuals treat one another as equals by letting them bear the costs of their (risky) choices (whatever those are). On this view, failing to pick up the tab for a victim of bad option luck is not treating this person as inferior. Tis is not saying that it is morally permissible or not morally disgraceful to   See also Anderson’s (1999, 299) remarks on ‘lazy, able-bodied surfers’.

30

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

102

Kasper Lippert-Rasmussen

not come to the rescue of the dying, negligent motorcyclist. Indeed, it is not to say that actually assisting the dying negligent victim is not fully compatible with treating one another as equals. In support of the former positive claim, some might say that one who refrains from helping might do so fully endorsing the thought that were he and the motorcyclist to switch places, it would be permissible for the motorcyclist to leave him to die. Moreover, Schefer’s own deliberative constraint does not rule out not helping the negligent victim. Tis is because Schefer understands ‘interests’ broadly so that interests include values, and if everyone subscribes to the value of consequential responsibility, then potential rescuers reasoning in accordance with the deliberative constraint might conclude that they are not required to ofer help. I conclude that it is justifed to focus on the locus of egalitarian justice as the core diference and disagreement between luck and relational egalitarians.31 6.  AN ECUMENICAL VIEW

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

In §4 I noted that dispositional egalitarianism is compatible with both of the former two views. In light of the reductionist challenge, this suggests that one might simply take the ecumenical egalitarian view that injustice obtains if the following disjunction is true: Ecumenical (luck) egalitarianism: it is unjust if, avoidably, any of the following three conditions are satisfed: 1) some people are worse of than others (through no responsibility of their own); 2) some people do not relate to each other as equals (through no responsibility of their own);32 or 3) some people, who are suitably related, are not disposed to avoid some people being worse of than others (through no responsibility of their own) or to relate to others as equals (through no responsibility of their own).33

If you ignore what is within the parentheses, you get ecumenical egalitarianism. If you do not, you get ecumenical luck egalitarianism. In   For further defense of this claim, see Lippert-Rasmussen 2015a; 2015b.   If the reductionist challenge is substantively vindicated in the way I have argued, then 2) is substantively contained in 1). However, stating 2) still has a presentational point. Moreover, some might reject my substantive vindication, but nevertheless see the attraction of the ecumenical view. 33   I include ‘avoidably’, i.e. would not have obtained had some agent who had it in his power to act diferently acted diferently, because ecumenical (luck) egalitarianism is intended to be an account of justice understood as an account of what we owe one another (see last paragraph, §1). Tis, however, is consistent with the ecumenical luck egalitarian principle being grounded in a concern for fairness, as suggested in §7 (cf. Vallentyne 2015, 49; Norman 1998, 40). 31 32

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

103

either case, we have a view which captures core elements from all three egalitarian views that I have discussed. In setting this out, the hope is that philosophers will see these elements as possibly combinable, perhaps even coherent, and not simply as parts of diferent, incompatible theories of egalitarian justice. Admittedly, ecumenical egalitarianism is not perfectly ecumenical. Unlike luck egalitarianism as defned in §2, it does not imply that unavoidable inequalities are unjust.34 Also, it does not capture the luck egalitarian concern for responsibility, and although it accommodates the positive component of relational egalitarianism as defned in §2, it is inconsistent with the negative relational claim. However, my cautious embrace of the reductionist challenge suggests that this component must be rejected anyway. Moreover, the view is ecumenical in terms of the locus of justice. Ecumenical (luck) egalitarianism is perfectly consistent. Compared to any of the three views of which it is formed, it takes a broader view of the locus of justice. According to ecumenical egalitarians, the locus of justice encompasses not just distributions, but also social relations (whether reducible to distributions of social goods or not) and people’s dispositions (independent of whether each afects any of the other loci of justice). Tis means that ecumenical egalitarians might say that, for instance, a certain situation is just in terms of its distribution, but nevertheless imperfect justice-wise, since some individuals are not disposed to relate to others as equals or are not disposed to respect the luck egalitarian deliberative constraint. Ecumenical egalitarianism is more complex than each of the three views upon which it draws. For instance, ecumenical egalitarians must say whether a slightly more unequal distribution which involves people relating to one another as equals to a signifcantly higher degree is better in terms of egalitarian justice than one in which the distribution is less unequal, but people less often relate to one another as equals. However, this sort of tradeof might have to be made anyway. Suppose egalitarian justice is fully captured by a concern for distributions being equal. However, there is also an independent moral concern that people relate to one another as equals. If so, we would still have to weigh these two concerns against one another; it is just that, unlike the ecumenical egalitarian, this weighing does not take 34  My defnitions of relational and dispositional egalitarianism also imply that unavoidably hierarchical relations are unjust and that unavoidably inegalitarian dispositions involve injustice. Whether unavoidable things can be unjust is an issue that is orthogonal to the issue between luck and relational egalitarians. Moreover, it is not true of luck egalitarian theorists that they all object to unavoidable inequalities (cf. Dworkin 2000, 105). Relational egalitarian theorists have not, to my knowledge, addressed the issue of what to say about unavoidable hierarchies, assuming, possibly falsely, that such there are.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

104

Kasper Lippert-Rasmussen

place within egalitarian justice, but is one that pertains to the concern for egalitarian justice versus other concerns, all of which, presumably, might bear on moral permissibility. 7. FAIRNESS Suppose we agree that ecumenical egalitarianism is an internally consistent position. Even so, it might still seem like an arbitrary concatenation of diferent views with no signifcant relation to one another. To respond to this suggestion, I want to suggest that ecumenical luck egalitarianism is rooted in the value of fairness: It is unfair if people are diferently situated if the fact that they are diferently situated does not refect their diferential exercise of responsibility.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Fairness so construed—and admittedly there are other ways to understand it (Hooker 2005)—grounds the distributive luck egalitarian component (since having unequal amounts of the relevant equalisandum is a way of being diferently situated) and the relational luck egalitarian component (since not being treated as an equal is a way of being diferently situated) of the ecumenical view. To reach the dispositional luck egalitarian element, one could appeal to the transmission principle: If a certain state of afairs or a certain way of relating to one another is unfair, then, all else being equal, a person is unfair to the extent that she is not disposed to act under a constraint of neither bringing about such an unfair state of afairs nor relating to others in this way.35

To see the need for the ‘other things being equal’ qualifcation, suppose X and Y are partners in love. X treats Y in a sexist manner. X treating Y in that way is an unfair state of afairs. I could have prevented this. Yet the transmission principle does not imply—implausibly, some would say—that I am unfair, for it might be the case that not all else is equal. Either I have the authority to interfere in the way I could have interfered with the couple’s relations or I do not. If the former, the transmission principle implies that I am less of a fair person for not interfering. Given that I had the authority to act in a way that prevented this state of afairs from obtaining and given that there was no moral justifcation for my not so acting, this implication strikes me as plausible. If the latter, then not all other things are equal and,

35

  ‘Bringing about’ refers both to doing and allowing.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



‘(Luck and Relational) Egalitarians of the World, Unite!’

105

accordingly, the transmission principle does not imply that I am unfair for not intervening. Te transmission principle seems plausible in itself. In what sense could a person not be unfair (or, at any rate, not fail to be fair) if she was in no way disposed to bring about an unfair state of afairs? In assessing this claim, observe that one might be somewhat disposed not to bring about a certain unjust state of afairs and yet fail to prevent it from obtaining because of a stronger, countervailing disposition, e.g., a disposition to avoid unreasonably large costs. Moreover, there seem to be quite similar and equally plausible transmission principles, e.g., ‘If a certain state of afairs is morally bad, then a person is morally bad to the extent she is not disposed to avoid bringing about such a state of afairs’. If we care non-instrumentally about persons not being unfair and we care about fairness, then we care non-instrumentally about persons being disposed to neither countenance that some people are worse of than others through no responsibility of their own, nor that people do not relate to each other as equals through no responsibility of their own. Ex hypothesis, to care about this simply is part of what it is to care about fairness. Te transmission principle ties fairness of states of afairs (where that includes unfairness of relations) to fairness of persons. Tis strikes me as plausible. Suppose we think that luck egalitarianism captures what fairness requires distribution-wise, but, Hume-like, also think that a fair person is simply one who respects the laws and keeps her promises. In that case, we could not say that two persons who are equally law-abiding and promisefulflling are unequally fair, even if one knowingly causes, and the other one does not, inequalities for which the worse of are not responsible. But it would seem odd, say, for the two persons who share the pertinent view about fair distribution and despite their knowledge of the way in which they difer, to regard each other as being equally fair persons.36 Still, the transmission principle does not say that what makes a fair person a fair person is that she is disposed to act under a constraint of not bringing about unfair states of afairs—something many virtue ethicists would deny. Te principle is consistent with the view that what makes it unjust that some people are worse of than others through no responsibility of their own is that such a state of afairs is one that unfair persons are disposed to bring about. Nor does the principle say that we have reasons to care about whether persons are fair.

36  A similar argument applies to the view that fair relations between people are relations in which no one relates to others as unequals (through no responsibility of their own) and its connection to views about what it is to be a fair person.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

106

Kasper Lippert-Rasmussen

No doubt, some who are attracted to ecumenical luck egalitarianism will not embrace the grounding in fairness that I have proposed. For instance, if one holds, as Anderson does, that the ideal of equality is grounded in some contractualist notion of fair cooperation, one might reject the broad, non-agency-presupposing notion of fairness of the view proposed here. My aim here is not to show that ecumenical luck egalitarianism grounded in the way I have just proposed is attractive to any egalitarian. Nor is it my aim to defend ecumenical luck egalitarianism against other forms of ecumenical egalitarianism beyond whatever defense lies in the suggestion that it is rooted in fairness. My aim is to show that there is a rationale for a luck egalitarian view which extends to all three sites of justice discussed in this piece.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

8. CONCLUSION In this chapter I identifed the core diference and core disagreement between luck egalitarianism and relational egalitarianism. I then argued that while, in a purely formal sense, the reductionist challenge to relational egalitarianism can be rebutted, locus-of-justice-wise there is no substantive signifcance to the relevant diference between the two views, in efect vindicating the reductionist challenge. I then addressed two objections to this line of argument—one appealing to the centrality of the relational egalitarian critique of the luck egalitarian ‘obsession’ with responsibility, and the other appealing to Schefer’s recent introduction of a certain deliberative constraint in his attempt to undermine the reductionist challenge. In the course of responding to Schefer, I defned a third egalitarian position—one which pertains neither to distributions nor to relations (whether the concern for relations is reducible to a concern for relations or not), but to dispositions—which is not fully explicit in Schefer’s work. In response to the responsibility challenge, I noted that the responsibility issue is orthogonal to the locus of justice issues and that, in light of the reductionist challenge, responsibility is relevant even to paradigm relational egalitarians. I then proposed an ecumenical luck egalitarian view, suggesting that it can be grounded in a concern for fairness understood in a certain way. Not all family disputes can be settled; yet if I am right, some issues within this particular one can. Admittedly, there are diferent egalitarian views that all fall within the category of ecumenical egalitarianism and, while I have briefy sketched a luck egalitarian variant thereof that accommodates some of Anderson’s and Schefer’s insights, I have not defended this version against other versions. Nor has this been my intention. My main intention has been to clear the underbrush such that the debate

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

107

between luck and relational egalitarians can be seen in a diferent and non-distorting light.37

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Anderson, E. 1999. ‘What is the Point of Equality?’ Ethics 109 (2): 287–337. Anderson, E. 2008. ‘Expanding the Egalitarian Toolbox: Equality and Bureaucracy.’ Proceedings of the Aristotelian Society Suppl. Vol. 82 (1): 139–60. Anderson, E. 2010. ‘Te Fundamental Disagreement between Luck Egalitarians and Relational Egalitarians.’ Canadian Journal of Philosophy Suppl. Vol. 36: 1–23. Anderson, Elizabeth and Richard H. Pildes. 2000. ‘Expressive Teories of Law: A General Restatement.’ University of Pennsylvania Law Review 148 (5): 1503–75. Arneson, R. J. 1995. ‘Against “Complex” Equality.’ In Pluralism, Justice and Equality, edited by David Miller and Michael Walzer, 226–52. Oxford: Oxford University Press. Arneson, R. J. 1999. ‘Egalitarianism and Responsibility.’ Journal of Ethics 3 (3): 225–47. Arneson, R. J. 2000. ‘Luck Egalitarianism and Prioritarianism.’ Ethics 110 (2): 339–49. Brighouse, H. and A. Swift. 2014. Family Values: Te Ethics of Parent-Child Relationships. Princeton, NJ: Princeton University Press. Burke, E. 1987 [1790]. Refections on the Revolution in France. Harmondsworth: Penguin Classics. Cohen, G. A. 2008. Rescuing Justice and Equality. Cambridge, MA: Harvard University Press. Cohen, G. A. 2011. On the Currency of Egalitarian Justice: And Other Essays in Political Philosophy. Princeton, NJ: Princeton University Press. Cohen, G. A. 2013. Finding Oneself in the Other. Princeton, NJ: Princeton University Press. Dworkin, R. 2000. Sovereign Virtue: Te Teory and Practice of Equality. Cambridge, MA: Harvard University Press.

37   Earlier versions of this chapter were presented at the Danish Philosophy Conference, March 4, 2016; Moral Philosophy Seminar, University of Oxford, May 9, 2016; Department of Philosophy, University of Reading, May 10, 2016; and the 4th Annual Oxford Workshop for Studies in Political Philosophy Conference, University of Pampeu Fabre, Barcelona, June 1, 2016. I thank Cristina Astier, Ian Carter, Andreas Christiansen, Roger Crisp, Jens Damgaard Taysen, David Estlund, Jessica Flanagan, Søren Flinch Midtgaard, Anca Gheaus, Nils Holtug, Brad Hooker, Kristian Jensen, Sune Lægaard, Zoltan Miklosi, Viki Møller Lyngby Pedersen, Charlotte Newey, Geraldine Ng, Lasse Nielsen, Serena Olsaretti, Tore Vincent Olsen, Tom Parr, Tomas Sinclair, Adam Slavny, David Sobel, Rasmus Sommer Hansen, Zofa Stemplowska, Philip Stratton-Lake, Victor Tadros, Patrick Tomlin, Peter Vallentyne, Alex Voorhoeve, Steve Wall, and Andrew Williams for their helpful comments. I am grateful to Mike Otsuka, who acted as a respondent on the  last occasion mentioned, and to Richard Arneson, who was one of two reviewers for Oxford Studies of Political Philosophy.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

108

Kasper Lippert-Rasmussen

Firth, J. 2013. ‘What’s So Shameful About Shameful Revelations.’ Law, Ethics, and Philosophy 1: 31–51. Gheaus, Anca. 2016. ‘Hikers in Flip-Flops: Luck Egalitarianism, Democratic Equality and the Distribuenda of Justice.’ Journal of Applied Philosophy Online frst: 1–16. Grifn, James. 1986. Well-Being. Oxford: Clarendon Press. Hellman, Deborah. 2008. When Is Discrimination Wrong? Cambridge, MA: Harvard University Press. Hinton, T. 2001. ‘Must Egalitarians Choose Between Fairness and Respect?’ Philosophy & Public Afairs 30: 72–87. Hooker, Brad. 2005. ‘Fairness.’ Ethical Teory and Moral Practice 8 (4): 329–52. Hume, David. 2004 [1751]. An Enquiry Concerning the Principles of Morality. Oxford: Oxford University Press. Lippert-Rasmussen, K. 2015a. ‘Luck Egalitarianism Versus Social Relations Egalitarianism.’ Canadian Journal of Philosophy 45 (2): 220–41. Lippert-Rasmussen, K. 2015b. Luck Egalitarianism. London: Bloomsbury Press. Miklosi, Zoltan. 2018. ‘Varieties of Relational Egalitarianism.’ In Oxford Studies in Political Philosophy, Vol. 4, edited by David Sobel, Peter Vallentyne, and Steven Wall, 110–37. Oxford: Oxford University Press. Miller, David. 1995. ‘Complex Equality.’ In Pluralism, Justice and Equality, edited by David Miller and Michael Walzer, 197–225. Oxford: Oxford University Press. Miller, David. 1998. ‘Equality and Justice.’ In Ideals of Equality, edited by Andrew Mason, 21–36. Oxford: Blackwell. Norman, R. 1998. ‘Te Social Basis of Equality.’ In Ideals of Equality, edited by Andrew Mason, 37–51. Oxford: Blackwell. O’Neill, Martin. 2008. ‘What Should Egalitarians Believe?’ Philosophy & Public Afairs 36: 119–56. Parft, D. 1998. ‘Equality and Priority’. In Ideals of Equality, edited by Andrew Mason, 1–20. Oxford: Blackwell. Rawls, J. 1971. A Teory of Justice. Oxford: Oxford University Press. Scanlon, T. M. 2003. Te Difculty of Tolerance. Cambridge: Cambridge University Press. Schefer, S. 2003. ‘What is Egalitarianism?’ Philosophy and Public Afairs 31 (1): 5–39. Schefer, S. 2005. ‘Choice, Circumstance, and the Value of Equality.’ Politics, Philosophy, and Economics 4 (4): 5–28. Schefer, S. 2015. ‘Te Practice of Equality.’ In Social Equality: On What It Means to Be Equals, edited by C. Fourie et al., 21–44. Oxford: Oxford University Press. Schemmel, C. 2012. ‘Distributive and Relational Equality.’ Politics, Philosophy & Economics 11 (2): 123–48. Tan, K.-C. 2012. Justice, Institutions, and Luck. Oxford: Oxford University Press. Temkin, L. 1993. Inequality. Oxford: Oxford University Press. Tomlin, Patrick. 2015. ‘What is the Point of Egalitarian Social Relationships?’ In Distributive Justice and Access to Advantage: G. A. Cohen’s Egalitarianism, edited by Alexander Kaufman, 151–79. Cambridge: Cambridge University Press. Vallentyne, P. 2015. ‘Justice, Interpersonal Morality, and Luck Egalitarianism.’ In Distributive Justice and Access to Advantage: G. A. Cohen’s Egalitarianism, edited by Alexander Kaufman, 40–9. Cambridge: Cambridge University Press.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



‘(Luck and Relational) Egalitarians of the World, Unite!’

109

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Wolf, J. 1998. ‘Fairness, Respect, and the Egalitarian Ethos.’ Philosophy & Public Afairs 27: 97–122. Wolf, J. 2010. ‘Fairness, Respect and the Egalitarian Ethos Revisited.’ Journal of Ethics 14 (3–4): 335–50. Young, I. M. 1990. Justice and the Politics of Diference. Princeton, NJ: Princeton University Press.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

5

Varieties of Relational Egalitarianism Zoltan Miklosi

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1.  DISTRIBUTIVE VERSUS RELATIONAL CONCEPTIONS OF JUSTICE In recent decades, relational egalitarianism has emerged as an infuential rival of distributive conceptions of justice.1 Te main thrust of the relational critique of distributive conceptions is that the latter, with their focus on  individual shares of some relevant goods (be they welfare, resources, capabilities, etc.), neglect the importance of social relations, as distinct from distributions of non-relational goods, in the ideal of a just society. However, diferent versions of the relational critique present diferent accounts of the importance of social relations, and as a consequence imply diferent views about the suggested shortcomings of distributive conceptions. My aim in  this chapter is threefold: frst, to distinguish diferent versions of the relational critique; second, to clarify the extent to which the diferent versions represent a fundamental challenge to the distributive conceptions; and third, to assess the success of those versions of the theory that represent a fundamental challenge. By “fundamental challenge” I mean a challenge that cannot be accommodated simply by extending the theory to incorporate additional dimensions, or by revising some particular aspects of it without altering the basic orientation of the view. I will argue that some critics who identify themselves as relational egalitarians do not pose a fundamental challenge, understood in this way, to distributive conceptions of egalitarianism. I will also argue that the views that represent the clearest challenge to the distributive conception are also the least plausible ones and therefore the least likely to succeed. 1   Representative expositions of relational egalitarianism include Elizabeth Anderson (1999, 2010), Samuel Schefer (2003, 2005, 2015) and Jonathan Wolf (1998), among others.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

111

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Before discussing diferent versions of the relational critique, however, it is necessary to make a few clarifcations.2 Many of the early formulations of  the relational critique focused not on distributive conceptions of justice in general, but on specifc versions of distributive egalitarianism. In particular, the early focus was on luck egalitarianism or even specifc versions of luck egalitarianism only.3 For instance, Elizabeth Anderson’s claims that luck egalitarianism is too harsh on the victims of bad option luck, or that it is disrespectful towards the victims of bad brute luck (1999, pp. 295–307) address such features that are not shared by many other distributive views.4 Tese criticisms, if valid, could be avoided simply by dropping or revising those features of a distributive conception that are responsible for generating the putatively damning implications. Terefore, it is important to stress that my focus here is on the relational critique of distributive egalitarianism in general, whether of the responsibility-sensitive kind or not. I will focus on those aspects of the relational critique that seek  to undermine this general orientation (i.e. the contention that the distribution of some goods has such signifcance from the point of view of justice that is independent of its impact on social relations) rather than particular specifcations of it. By the same token, my focus is the relational critique of distributive conceptions in general as distributive conceptions, and not only of those that endorse an egalitarian distributive principle rather than, say, a prioritarian or sufcientarian one. Te basic issue of contention is not the particular pattern of distribution that should be favored by egalitarians, but whether egalitarians should be concerned with distributions (of resources, welfare, etc.)5 independently of their impact on social relations in the frst place. 2   As a general note, I am using the term “justice” to refer to that particular set of moral concerns that involves duties that we owe to each other, i.e. interpersonal wrongs as distinct from impersonal wrongs. I also take it that duties of justice are pro tanto enforceable, but I don’t treat this as part of the concept of justice. 3   Elizabeth Anderson’s very infuential article that is widely regarded as giving rise to relational egalitarianism as a theory (though Anderson herself takes Rawls to be a relational egalitarian) is mostly a critique of luck egalitarianism (Anderson  1999). Schefer’s work (Schefer 2003, 2005) also focuses on luck egalitarianism. 4   In particular, Anderson’s claim that the luck egalitarian compensation for victims of bad brute luck is motivated by the judgment that their lives are “less worth living” (Anderson 1999, p. 305) applies, if at all, only to welfarist versions of luck egalitarianism. Luck egalitarians need not be committed to any such judgment; they may be motivated by the thought, not that the lives of the victims of bad luck are subjectively miserable, but that they are objectively disadvantaged. 5   It is important to emphasize that this general orientation is neutral on the so-called currency debate in the theory of justice, i.e. on what it is the distribution of which should be the central concern of justice. Distributive egalitarians of all stripes accept that the distribution of whatever it is that they take to be the correct currency of distribution has

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

112

Zoltan Miklosi

Next, certain conceptions of justice hold that distributive inequalities are unjust only if they occur between persons who stand in certain social, economic, or political relations with each other. Te relevant relations may be those of economic cooperation, shared social structures, or shared political institutions.6 Tese views do not count as relational in the sense used in this chapter. Te “associational” views defend a claim about the ground of the duties of justice, and are consistent with the claim that distributions have relation-independent moral signifcance whenever the relevant associative relations obtain. By contrast, relational views are typically understood as making claims about the content of the requirements of justice. To be sure, these two views are not mutually exclusive, and the thought that the content of the requirements of justice is primarily about the quality of social relations may motivate an associative restriction on the scope of justice. Be that as it may, my concern in this chapter is only the relational view as specifed here. Finally, some contractualist authors claim that distributions are not what is ultimately at stake from the point of view of justice. Tis is so, they argue, because distributive patterns “cannot be owed” to persons given the constraints on owing that are imposed by the contractualist framework—in particular, by its claim that the reasons that are eligible candidates for warranting the rejection of moral principles must be personal ones, and distributional patterns are supported, if at all, by impersonal reasons (James 2011). What can be owed, on this view, is at most one’s respective share of society’s resources (James 2011, p. 276). I take it that this claim, as far as it goes, is not hostile to the core of the distributive view as I understand it, because it is consistent with individuals having personal, relation-independent claims to a certain share of certain goods, and for rejecting distributive principles that deny it to them. Te distributive view need not be committed to the impersonal value of distributive patterns; it is committed only to the relation-independent signifcance of individual distributive shares. Terefore, this particular controversy is outside the focus of this chapter. Given this focus, however, it will be necessary to give a rough account of the normative core of the distributive conceptions, i.e. that aspect of the theory that must be undermined if the fundamental relational challenge is to be successful. I suggest tentatively that this normative core is the following simple claim: the distribution of nonrelational goods among members of society has signifcance from the point of view of justice that is independent signifcance from the point of view of justice that is independent of the distribution’s impact on social relations. I am grateful to an anonymous referee for pressing me to clarify this point. 6   Examples include Tomas Nagel (2005).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

113

of the efects, if any, that distributions have on the quality of social relations, or of their being the outcome of unequal social relations. Tis is the thesis of the relation-independent signifcance of distributions. I deliberately avoid talking about the “basic” or “fundamental” or “intrinsic” signifcance of distributions, because I want to leave it open that the signifcance of distributions is ultimately grounded in some more basic idea, such as the equal worth or moral status of persons. Te claim needed by the distributive conceptions is simply that distributions have such moral signifcance that is independent of the moral signifcance of the quality of social relations.7 It is this claim of relation-independent signifcance that a fundamental relational challenge has to reject. Let’s call this the Core Distributive Tesis.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Core Distributive Tesis: the distribution of non-relational goods has relation-independent signifcance from the point of view of justice.8

Now, as I discuss later in this section, almost all distributive egalitarians actually endorse a much stronger claim than this minimal one concerning distributions having some relation-independent signifcance. Tey usually hold that this signifcance is fairly central for the ideal of justice. Nonetheless, it is important to identify that form of the relational critique that poses the most fundamental challenge to the distributive view. I shall regard a relational criticism as posing a fundamental challenge to the distributive conception only if it rejects the core distributive thesis.9 I will refer to such views as instances of relational exclusivism. In the last section of this chapter I will discuss separately the problem of the relative weight of distributions within an overall egalitarian conception, and thus the “moderate” relational challenge that even though distributions can have relation-independent moral signifcance, this signifcance is relatively minor. For now, the focus is on exploring the fundamental challenge of relational exclusivism. Te thesis of relation-independent signifcance is a positive claim. I take it that the corresponding negative claim that social relations have no distribution-independent signifcance from the point of view of justice is not part of the core distributive view. Let us call this negative claim the 7   “Quality” is understood broadly to include a concern with the way social relations respect the equal social status of persons, for instance. 8   I note that welfarist distributive egalitarians can endorse this claim, too, since in their view the signifcance of the distribution of resources is grounded in their importance for welfare, and not (primarily) for social relations. 9   Tis is obviously not to say that only such a relational critique can be important or serious that is fundamental in this sense. But nonfundamental challenges are better understood as seeking to modify, revise, or supplement existing distributive conceptions, rather than rejecting them altogether.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

114

Zoltan Miklosi

thesis of distributive exclusivism.10 Tere are multiple reasons for not regarding this negative claim as part of the core of the distributive view. First, this thesis is not relevant for the assessment of the relational critique. Tat critique poses a fundamental challenge to the distributive view only if it denies the relation-independent signifcance of distributions, and it is irrelevant in this respect whether or not the theory under attack happens to accept the distribution-independent signifcance of social relations. If the relational critique amounted only to asserting the distribution-independent signifcance of social relations, then it would insist merely on supplementing rather than challenging the distributive view. However, this is not how most relational egalitarians see their position; they criticize distributive theories not simply for being incomplete but for being misguided in being primarily oriented towards distributions. Second, distributive exclusivism does not refect the expressed views of most distributive egalitarians. Most theorists who are criticized under that description have no difculty accepting that the quality of social relations is independently important from the point of view of justice,11 or at any rate there is little or nothing in their expressed views that would rule out such an acceptance. I will make the intuitive suggestion that individuals can have at least some such goals, plans, and pursuits, for instance the pursuit of excellence or knowledge and understanding, of excitement and adventure, or of aesthetic experience, that are not plausibly understood as being relational projects and which give rise to interests that are nonrelational at a fundamental level. (Tis is not to say that developing and sustaining meaningful relationships cannot fgure among the most important interests of people—it almost always does.) Te distributive conceptions claim that people’s morally relevant nonrelational interests fgure prominently in the justice-based evaluation of 10  In her original statement of the contrast between distributive and relational conceptions, Anderson construes the distributive position in this exclusivist manner (1999, p. 313). Tis may be true of only a few distributive egalitarians, such as Richard Arneson, who suggests that the quality of social relations matters only instrumentally, as a dimension of individual well-being (Arneson, 2000, p. 342). Kasper Lippert-Rasmussen’s chapter in this volume develops a similar “guardedly reductionist” account. 11   See, for instance, G. A. Cohen’s discussion of community and solidarity (2009). Ronald Dworkin, too, professes to acknowledge this importance, e.g. in his reply to Schefer (Dworkin 2003, pp. 195–6). However, in his discussion of political equality he insists that inequality of political infuence is objectionable only when it is a consequence of (or is likely to reinforce) unjust economic inequality, which seems to suggest that egalitarian political relations have no independent normative signifcance (Dworkin 2000, pp. 195–6). At the same time, he does admit that political participation has independent signifcance and rejects only the idea that this signifcance calls for an equal division of infuence. Tus, he is most naturally interpreted as acknowledging the distributionindependent signifcance of the quality of (political) relations and endorsing an adequacy (rather than equality) standard of political infuence.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

115

distributions, and they so fgure not only because the frustration of these interests may undermine the desired quality of social relations, but rather they fgure in such evaluations in their own right. Tat is to say, in the distributive view they fgure in evaluations in a relation-independent manner. Te relational egalitarian need not deny the existence of nonrelational interests, but she must reject that they fgure in the justice-based evaluation of distributions and social arrangements in their own right, independently of how individuals’ opportunity to pursue them may afect the egalitarian character of social relations. Now, it may be suggested that this way of characterizing the basic structure of the disagreement between distributive and relational conceptions of justice gives undue advantage to the defenders of the distributive conceptions, because they need only to defend the fairly minimal claim that distributions have some (no matter how tiny) relation-independent signifcance for justice. By contrast, in my characterization it is not sufcient for a successful relational criticism to defend a similarly minimal relational thesis (i.e. that egalitarian relations have some—no matter how tiny—distributionindependent signifcance for justice). Tis appears to give a huge advantage to the distributive view. My response to this suggestion is twofold. First, as far as relational exclusivism is concerned, it seems to me that the distributive view is in fact entitled to this advantage, because relational exclusivism denies even the minimal distributive claim, and therefore it is sufcient to dislodge that criticism to establish the relation-independent signifcance (no matter how tiny) of distributions.12 By contrast, the distributive view as I understand it is not committed to denying the parallel minimal relational claim; therefore, it is not threatened by the latter, either. And since some of the most infuential relational critiques are best construed as exclusivists’ (or so I will argue), establishing the plausibility of the minimal distributive claim is highly relevant for the relational-distributive debate. Second, I would like to acknowledge the intuitive force of this point by recognizing a somewhat weaker version of the relational critique. Tis “moderate” version holds, roughly, that even though distribution may have some relation-independent signifcance for justice, this signifcance is really minor compared to the signifcance of egalitarian social relations. Of course, this criticism has some bite only if the distributive theorist endorses more than just the minimal claim noted above and asserts that the relation-independent signifcance of distribution is fairly large (even if not exclusive) for justice. Since this is typically the case, the “moderate” relational critique will also pose a serious challenge to the distributive view. I will discuss this issue briefy in the last 12   See the discussion of Anderson in §4, on relational exclusivism, for support that she in fact makes this strong claim.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

116

Zoltan Miklosi

section of this chapter, where I will argue that the relation-independent signifcance of socioeconomic resources is likely to be fairly central for a plausible egalitarian conception of justice.13 I now turn to various diferent presentations of the relational critique, to see if they represent a fundamental challenge in just this sense.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

2.  DISTRIBUTION IS NOT ENOUGH In discussing diferent versions of the relational critique, I will proceed from what I take to be the less demanding, weaker claims toward the strongest ones. Tus, I frst take up the position that holds that the concerns in the focus of the distributive views cannot account for every aspect of injustice. Christian Schemmel illustrates this objection through the following example (I leave out many details): imagine that in a society access to some vital good such as an important nutrient is very unequally distributed as a consequence simply of large economic inequalities. Now imagine that the exact same distribution of the vital good prevails in the society, but now as a result of ofcial, state-mandated discrimination or ofcially permitted private discrimination. We have a very clear sense that the latter scenarios are more severely unjust (and that ofcial discrimination is more unjust than ofcially tolerated private discrimination) than the frst one. How can we account for this sense of diferential injustice? Te distributive profles of the two societies are the same, and therefore the extent of inequality or of absolute deprivation cannot explain the more damning judgment regarding the latter cases. Schemmel argues that distributive views cannot capture the additional injustice present in cases where ofcial intention is involved in producing an unjust inequality. He suggests that the additional injustice is explicable in terms of the attitudes the state expresses, and that this expressive dimension is not reducible to distributive outcomes (Schemmel  2012, pp. 123–48). Te expression of hostility, inferiority, or neglect is an unjust, inegalitarian way of relating to some group of citizens, and the injustice of this relation is not exhausted by the unjust distributive patterns that result from them. It seems clear that Schemmel’s criticism of the distributive view does not amount to a fundamental challenge in the sense that I defned this earlier in  this chapter. In particular, it does not presuppose the rejection of the core  distributive thesis, the thesis that the distribution of nonrelational goods has relation-independent signifcance from the point of view of 13   I am grateful to the editors of this volume for bringing the issues discussed in the paragraph to my attention.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

117

justice. His arguments do not show—or even purport to show—that unequal access to the vital nutrient is not independently unjust, and therefore they are compatible with the relation-independent signifcance of distribution. He merely asserts the distribution-independent signifcance of the quality of social relations—in this case the distinct injustice of ofcially mandated or tolerated discrimination and the accompanying attitudes and expressions. Distributive egalitarians need not deny this, and many of them would readily agree.14 Now, Schemmel claims that distributive views cannot accommodate the  additional sense of injustice resulting from wrongful expressive acts (2012, pp. 137–41). But this claim is ambiguous. It may simply mean that distributive views qua distributive views cannot explain the wrongness of wrongful expression, in the sense that that part of these theories that explains the signifcance of distribution cannot account for this wrongness. Tis may be so, but that leaves it open whether other aspects of the theories can do the job. Or it might mean that any theory that accepts the relation-independent signifcance of distributions is barred, just in virtue of this fact, from being able to recognize the injustice of wrongful ofcial expression. But this is highly implausible and not supported by any of the arguments produced. (Furthermore, it would undermine Schemmel’s own account, too, given its acceptance of the independent signifcance of distributions.)15 A broadly similar view is outlined by Tomas Scanlon, who examines the  grounds of our objections to economic inequality (Scanlon 2003, pp. 202–18). He suggests that most of our moral unease with economic inequality is best explained with reference to either the noncomparative phenomenon of absolute deprivation or the way economic inequality threatens such independent, relational ideals as nondomination, equal status, procedural fairness, or nondiscrimination. Scanlon does acknowledge the independent signifcance of access to material, nonrelational goods such as health, education, food, and shelter. But he believes that this signifcance supports only a noncomparative standard of adequate or sufcient access, rather than the comparative standard of equal distribution. However, he contends that any concern with material inequality above the level necessary to make sure that no one sufers absolute deprivation of vital resources is 14  It might be suggested that the more damning judgment regarding ofcially mandated or tolerated discrimination is simply related to the moral signifcance of the distinction between doing and allowing. I will not explore this issue here. 15   It seems to me that the same is true regarding Jonathan Wolf’s relational view. Wolf writes that “there is more to a society of equals than a just scheme of distribution of material goods. Tere may also be goods that depend on the attitude people have toward each another” (Wolf, 1998, p. 104). As the quote itself suggests, Wolf does not question that the distribution of material goods is also a concern of justice in its own right.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

118

Zoltan Miklosi

best accounted for in terms of the relational concerns just mentioned. Once again, it seems clear that this position does not pose a fundamental challenge to the distributive view. It does not undermine the core claim of relationindependent signifcance, but merely asserts the distribution-independent signifcance of sociopolitical relations. In addition, it challenges a particular aspect of certain egalitarian views, i.e. their choice of distributive principle: it endorses adequacy rather than equality as the correct principle. But that debate is internal to the distributive conception, and not an external criticism of it.16 One might suggest that this analysis misrepresents Scanlon’s view (or views like his), because for him the relation-independent signifcance of avoiding absolute deprivation is a matter of humanitarian concern rather than of justice. On this proposal, justice is an exclusively relational ideal and distributions have no relation-independent signifcance for justice, though they have such humanitarian signifcance. If so, then his view still represents a fundamental challenge to the distributive conception.17 It seems to me that this way of understanding the link between justice and distribution creates a potential dilemma rooted in the ambiguity of the adequacy standard. To recall, the suggestion is that adequate access to vital goods has relation-independent moral signifcance, but this signifcance is grounded in duties of humanity rather than of justice. Inequalities in these goods above the adequacy level are objects of moral concern only to the extent that they have a corrosive efect on the egalitarian quality of social relations, which constitute the sole concern of justice. Te dilemma is the following. On the one hand, if the adequacy standard is understood, narrowly, to include the provision of subsistence levels only, then it is very implausible that access to these goods above the subsistence level has no relation-independent moral importance. On the other hand, if the adequacy standard is understood, more generously, to include access to goods that are necessary to a decent, socially acceptable life, then it is more plausible to hold that inequalities above that level do not matter from a nonrelational perspective, but then the very distinction between humanitarian and justice-based duties begins to look unmotivated. One can, of course, use 16   A set of recent works arguing that the currency of distributive justice should include relational as well as nonrelational goods does not constitute a fundamental challenge to the distributive conception, either. Tus, Zofa Stemplowska describes relational (or “social”) egalitarianism as follows: “social egalitarianism concerns itself exclusively with the distribution of relational goods and, specifcally, the relational good of social status” (Stemplowska 2011, p. 116). Similar proposals are found in Anca Gheaus (2009, 2016) and Cordelli (2015). Tese theories call for a revised or expanded understanding of the equalisandum of distributive justice; this is a controversy internal to the distributive conception, and therefore does not challenge its basis. 17   I am indebted to Chiara Cordelli for suggesting this response.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

119

the term “justice” to refer to exclusively relational concerns, but that amounts to attempting to undermine the distributive conception by conceptual stipulation. Tis move would make it appear that the debate between the distributive and relational conception is merely terminological, rather than substantive. It may be countered that there is a normatively motivated, nonarbitrary distinction between duties of justice and duties of humanity, with the former being based in mutual advantage or reciprocity more generally, and the latter being grounded simply in the reason-giving force of the badness of sufering and deprivation, even in situations not involving reciprocity. In response to this suggestion, it seems to me that, even if accurate, it cannot be of much help to the relational egalitarian.18 Tis is because the reciprocity view does not overlap with the relational claim that distributive shares have no relation-independent moral signifcance; instead, it is a claim about the basis of distributive claims. It suggests a necessary (and perhaps sufcient) condition of someone’s having a justifed claim to a certain share of some goods, without asserting anything about the moral signifcance of persons having access to those goods. Terefore, the thesis that distributive claims are based in reciprocity is consistent with distributive shares having relationindependent signifcance for justice.19 I therefore conclude that this version of the relational critique, whether or not it is right in the specifc points that it raises, does not represent a genuine alternative to the distributive conception.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.  EQUALITY AS A PRACTICE I now turn to what I take to be a more ambitious challenge to the distributive egalitarian approach. Tis criticism has been developed over many years and in numerous articles by Samuel Schefer. When contrasting the relational and distributive conceptions of egalitarianism, it is not so much the focus on distributions that is singled out for criticism, but the manner in which distributive theorists seek to answer the problem of distribution. Schefer sees distributive egalitarian theorists as addressing “themselves directly to questions of distribution.” Te key term in this characterization is “directly.” Distributive egalitarians seek to arrive at the correct formula or principle on the basis of abstract philosophical considerations, without any “attempt to 18   I am skeptical that duties of justice are grounded in reciprocity, for some of the same reasons that are developed by Allen Buchanan (Buchanan  1990). However, I have no space to engage with that debate. 19   I am grateful to an anonymous referee for bringing this suggestion to my attention.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

120

Zoltan Miklosi

situate the distributive principles they favor within a broader conception of the nature of egalitarian relationships” (Schefer 2005, p. 20). But what is meant exactly by “situating” distributive principles in such a broader conception? One thing that it might mean is just that the correct distributive principles are part of a broader ideal that takes social relationships as also, or even primarily, signifcant, and that the correct principles must be sensitive to the importance of social relations. Tere are many places in the text that support this reading.20 But this would simply reassert the points discussed in §§1 and 2; it would call only for supplementing the distributive focus rather than challenging it. At other places, however, Schefer implies something both more specifc and more ambitious by “situating” distributive principles in social relationships. He claims that the basic egalitarian ideal is that of a society of equals: of individuals committed to living together on equal terms and working out together, from a position of equality, shared answers to the problems created by their divergent goals and corresponding interests that frequently cannot be satisfed fully or equally at the same time. Te frst order of business for the egalitarian theorist is to identify the attitudes and practices constituting egalitarian relations. Only once we have a clearer sense of this idea can we try to answer the following question:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

[W]hat system for the allocation of economic resources is appropriate to a society of equals and what bases for the assignment of benefts and burdens such a society would recognize. In my view, an egalitarian scheme of distributive justice is best understood as one that tries to provide answers to these questions. (Schefer 2005, p. 20, emphasis added)

To put the point bluntly, the mistake made by distributive theorists is that of trying to identify the objectively correct distributive principles through direct theorizing. By contrast, the relational view now under consideration asks instead: what principles are likely to arise from the egalitarian practices of an idealized society of equals in which everyone is committed to treat others as equals? Te distributive principles should be selected or constructed in egalitarian social practices and not in philosophical reasoning.21 20   E.g. Schefer claims that “the basic reason [equality as a social value] matters to us is because we believe that there is something valuable about human relationships that are, in certain crucial respects at least, unstructured by diferences of rank, power, or status” (Schefer 2005, p. 17). 21   Tis idea is diferent from the thought, raised earlier, that the distributive principles might ultimately be grounded in the idea of the equal moral worth or status of persons. Te latter thought is consistent with the philosophical project of identifying the correct distributive principles through “direct” theorizing by using the idea of equal moral worth as one of its premises. By contrast, Schefer’s suggestion is that the principles of distribution ought to be adopted in egalitarian social practices rather than discovered in abstract reasoning (I thank an anonymous referee for pressing me to clarify this point).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

121

In a recent paper, this thought is presented by Schefer in deliberative terms. An egalitarian relationship is one that is characterized by a set of “egalitarian deliberative constraints,” chief among them being the requirement that “the comparably important interests of each party should play a comparably signifcant role in infuencing decisions made within the context of their relationship” (Schefer 2015, p. 2). And the right question to ask is what principles are likely to be accepted in a properly egalitarian deliberation. Schefer presents various considerations as to why luck egalitarianism would be an unlikely outcome of the egalitarian deliberation, but this is largely beside the point for present purposes. Te main point is that in his relational view the core of equality as a social value is the practice of equality, of relating to others as equals and of egalitarian deliberation, and the distributive principles are the outcome of this practice, rather than external standards applicable to it. To “situate” distributive principles in  egalitarian relationships is thus to construct them in an egalitarian deliberative practice. It is not quite accurate to say that this view denies the relation-independent signifcance of distributions. To the contrary, it asserts the importance of individual, nonrelational interests in specifying the governing deliberative constraint as attributing comparable importance to the comparable interests of the parties of deliberation. To this extent, the importance of egalitarian deliberative practices is directly linked to the importance of the nonrelational interests that are considered in the process of deliberation. Tus, the signifcance of distribution is not being denied. It is rather that this relational conception suggests a particular manner of settling the signifcance of distributions: their signifcance consists in what parties to the egalitarian deliberation are likely to agree to. At points, Schefer appears to deny altogether that distributions have such signifcance. Tus, he writes: “distributive equality matters, [relational egalitarians] claim, only because and insofar as it is necessary in order to achieve a society of equals” (Schefer 2015, p. 2).22 Tis would suggest that an equal distribution has only instrumental (relation-­ dependent) signifcance to the extent that it is a necessary condition of egalitarian deliberation. But the full view that emerges is more complex: it holds that distribution matters independently, but the precise implications of this importance should be identifed not by abstract philosophical reasoning, but in the actual social deliberation of equals. It is for this reason that the primary focus of egalitarian theorizing should be, in his view, the specifcation of the requirements of egalitarian deliberation and not a direct answer to the question of distribution. 22   Schefer suggests this as a possible understanding of the core relational view, yet it is not entirely clear from the context whether he endorses this claim.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

122

Zoltan Miklosi

Schefer’s critique thus focuses on direct distributive theorizing that searches for philosophical grounds to identify correct distributive principles. He also makes it clear that the egalitarian deliberation is unlikely to come down decisively in favor of any particular distributive principle: the egalitarian deliberative constraints may underdetermine the choice of distributive principle, although he suggests that they are very likely to rule out some conceptions, such as utilitarianism or luck egalitarianism (p. 19). His description of what is wrong with direct theorizing about distribution leaves some important questions open. One way to understand his account is as claiming that there is no deliberation-independent test of the correctness of candidate distributive principles: the correct principles just are the ones that are selected in the egalitarian deliberation. Egalitarian deliberation is constitutive of the correctness of the outcome. Tis would be a purely procedural conception of distributive justice.23 Such a conception does indeed pose a fundamental challenge to “direct” distributive theorizing, since if there is no deliberation-independent standard of correctness to be applied to distributive principles, then developing substantive theories of just distribution cannot contribute anything to our understanding of the demands of equality. But there is also reason to believe that this is not Schefer’s view. In many of his writings, including the ones that present the view of equality as a social practice, he develops substantive philosophical arguments against particular distributive principles (most notably, luck egalitarianism) that rely on criteria, such as the rejection of moralism or the  metaphysics of choice, which are quite independent of deliberation (Schefer  2003,  2005). However, if the purely deliberative-procedural understanding is the correct one, it faces some obvious difculties. If we understand the conception to be deliberative all the way down, in the sense that the participants’ egalitarian deliberative attitudes do all the work of selecting the principles, entirely independently of their substantive views of what counts as a fair distribution, then the deliberation is unlikely to get of the ground in the frst place. Te egalitarian deliberators could all agree to treat each other respectfully and impartially, to consider the reasons ofered by others honestly and fully, to be committed to reaching a principled agreement, etc., but there would be no substantive considerations ofered in the frst place.24   Te notion of pure procedural justice is introduced by Rawls (1999, p. 74).   It might be objected that the egalitarian deliberative constraint is just a rendering of the principle of Equal Consideration of Interests (ECI), and as such it is capable of yielding determinate outcomes. However, it seems to me that ECI can yield determinate outcomes only if it includes a specifc account of what it is to give equal consideration to interests (such as equality of welfare, equality of resources, etc.), in which case it is itself a substantive distributive principle rather than merely a constraint on deliberation. By 23 24

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

123

By contrast, if the conception is not purely deliberative, and the egalitarian deliberators’ substantive views about distribution can be assessed as objectively more or less correct according to a deliberation-independent standard, then it is not clear why this conception of equality as an egalitarian deliberative practice should be seen as a rival of the distributive conception, or as challenging the point of “direct distributive theorizing.” If it is the case that some views about distribution are objectively better than others, then it is a  perfectly reasonable philosophical project to try to identify the correct views. In this case, “direct” distributive theories can be seen as contributing to the egalitarian deliberation by providing material to it and sharpening the conceptual tools of the deliberators. Tere is nothing fundamentally misguided about this enterprise if some candidate distributive principles are objectively more correct than others.25 To be sure, someone persuaded by Schefer might say that the view of equality as an egalitarian deliberative practice still has bite. Te thought is this. While the correctness of the distributive principles that govern our social institutions matters, it is of great importance that the social practices in which we seek to identify the correct principles be egalitarian ones. One might even add that the egalitarian character of deliberative practices is more important than the objective correctness of the principles selected. Admittedly, it may look fetishistic to insist that the attitude of seeking to treat others justly is more important than whether we manage to actually treat them justly. Nevertheless, the point can still be appreciated. Understood in this way, the criticism of direct theorizing about distribution can be taken to suggest not so much misguidedness but as having the wrong priorities. It  is more important, normatively, to identify the criteria of egalitarian deliberative practices than to identify the criteria of just distribution. contrast, if ECI does not endorse any specifc account of what it is to consider interests equally, then it will not be able, by itself, to select among the many possible accounts (I am grateful to Tomas Christiano for pressing me on this point). 25   It seems to me that similar things can be said about the contractualist view that James attributes to Scanlon. In this rendering, “the signifcance of distribution depends on independently valuable relations among people.” Te right kind of relations obtain when we are “acting only in ways we could justify to them, because this sustains a valuable ‘relation of mutual recognition’ with them” (James 2011, p. 277). Given that the right kind of relation is defned with reference to what can be justifed to each person, and the relevant type of justifcation proceeds through the personal reasons of each person who is asked to accept the ofered principles, and given further that their legitimate personal reasons may well include non-relational ones, this account does not show that the distribution of material goods does not have relation-independent signifcance. To the contrary, such a signifcance is built into the conception of the relation of “mutual recognition.”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

124

Zoltan Miklosi

It is not necessary here to assess this criticism on the merits. Te position seems open to doubt to me, but let us put that aside. Even so, this still does not amount to a fundamental challenge of the distributive view. Te view of equality as primarily a deliberative practice criticizes distributive theorists not qua distributive theorists but qua egalitarian theorists more generally, for focusing on an aspect of equality that is, on the current proposal, of relatively lesser importance. Tere is nothing in this view that casts doubts on the independent signifcance of distributions. It holds only that their independent signifcance is relatively minor compared to the importance of  egalitarian deliberative practices (I will return to the issue of relative importance in §6). Terefore, Schefer’s arguments fail to show the distributive views to be misguided. One might counter that the view of equality as a deliberative practice does challenge, despite the arguments presented in this section, the value of  philosophical theorizing about distribution. Te argument may go as follows: when a particular view about fair distribution emerges from the egalitarian deliberative practice as the favored view, the ultimate justifcation of this view—the decisive moral reasons that we have for adopting it as a society—will not be linked to the substantive reasons that are ofered in its support during the deliberation, but rather to the fact that it is the view that emerged as victorious from the deliberation. Te substantive reasons favoring it play no role in showing that the view should be adopted and enforced in our society and to this extent philosophical theorizing that focuses on these substantive reasons is misguided.26 It seems to me that this strategy to establish the misguidedness of direct distributive theorizing is itself misguided, at least to the extent that it seeks to provide a conception of egalitarian justice. Te idea that the ultimate justifcation for society’s adopting and implementing certain views about fair distribution rests on the fact that these views emerged from a process of egalitarian social deliberation belongs to the problem of legitimacy, not to that of substantive justice. It provides a crucial part of  the answer to the question concerning when it is permissible for a community to implement a particular view of justice; it does not seek to answer questions concerning the objective correctness of particular views about justice. One may adopt this view about legitimacy and still consider “direct” theorizing about distribution a perfectly sensible and appropriate project. Terefore, I conclude that the view of equality as an egalitarian practice of deliberation does not pose a fundamental challenge to the distributive view of justice.   Tis interpretation of Schefer’s view was suggested to me by Chiara Cordelli.

26

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

125

4.  RELATIONAL EXCLUSIVISM

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Last, I take up the strongest version of the relational critique. To recall, I previously identifed the core distributive thesis as asserting the relationindependent signifcance of distributions from the point of view of justice. I argued that any relational view that is consistent with that thesis does not pose a fundamental challenge to the distributive conception as a whole, but either seeks to supplement it, or criticizes particular versions of the distributive conception, or suggests that distribution is not the most important aspect of the egalitarian ideal. All of that is consistent with seeing distributive theorizing of the familiar form as having philosophical interest in its own right and does nothing, by itself, to impugn the pursuit of identifying correct distributive principles. I suggested that with the partial exception of Schefer’s view, all of the previously discussed relational views fall in one of these categories and therefore they are not properly seen as alternatives to the distributive project as such, but rather as supplementing it. Tis is not the case, I believe, with the most infuential version of the critique developed by Elizabeth Anderson. It is worth quoting her text at some length. When describing the contrast between the distributive conception and her own relational view in abstract terms, Anderson writes that the distributive view conceives of equality as a pattern of distribution. Tus, [it] regards two people as equal so long as they enjoy equal amounts of some distributable good—income, resources, opportunities for welfare, and so forth. Social relationships are largely seen as instrumental to generating such patterns of distribution. By contrast, democratic equality27 regards two people as equal when each accepts the obligation to justify their actions by principles acceptable to the other . . . Certain patterns in the distribution of goods may be instrumental to securing such relationships, follow from them, or even be constitutive of them. But democratic egalitarians are fundamentally concerned with the relationships within which goods are distributed, not only with the distribution of goods themselves.  (Anderson 1999, pp. 313–14)

Te last clause of this quote appears to admit the possibility that distributions can have independent signifcance, even though distribution is not the only or most important concern for relational egalitarians. However, read as a whole, the passage suggests that distributions matter only as instruments of egalitarian relations or as constituents of such relations, not independently of them.28 Tis reading is reinforced by a more recent formulation in which 27   “Democratic equality” is Anderson’s term for her favored conception of relational equality. 28  To be sure, if certain distributions are constitutive of egalitarian social relations, then their importance is not derivative, although it is still relation-dependent. However, if

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

126

Zoltan Miklosi

Anderson writes that relational egalitarians see equality “as a kind of social relation between persons—an equality of authority, status, or standing” and distributive inequality is “unjust . . . when it refects, embodies, or causes inequality of authority, status, or standing” (Anderson 2010, pp. 1–2). I take it that the last part of this quote states a necessary and sufcient (and not simply a sufcient) condition for a distributive inequality to count as unjust.29 Tus, distributive inequalities constitute a matter of justice only if they contribute to, or derive from, inequalities along these other lines. Tis is the view that I will call relational exclusivism, since it holds that inequalities in nonrelational goods matter from the point of view of justice only when and only because they are linked to relational inequalities.30 Te picture becomes more complicated when Anderson suggests a further way of understanding the contrast between the distributive and relational views. She suggests that on the distributive view, one frst identifes the features that make states of afairs just or unjust, and the “justice of agents is derived from an independent standard of the justice of states of the world.” In the relational view, justice of the “disposition” of agents is primary, and

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

[d]istributions of socially allocated goods are just if they are the result of everyone acting in accord with such principles [of social equality]. Te justice of distributions is derived from an independent standard of the justice of agents. (Anderson 2010, p. 2)31

In this version of the relational view, what is just nonderivatively is the disposition of agents to treat others as equals, and any state of afairs that results from such treatment is just. Conversely, there is no independent standard of the justness of distributions: a distribution is unjust if and only if it is a result of agents acting on the basis of unjust dispositions.32 certain distributions are constitutive of egalitarian social relations, then distributive theorizing aiming at identifying these distributions is a sensible and important philosophical project (I thank an anonymous referee for this point). 29   If we construe it as stating a sufcient condition, then the proposed view is again supplementary to, and not a rival of, the distributive conception. 30   I include the “only because” formula since the distributive view may admit that distributive inequality matters “only when” it is linked to relational inequalities. Tis may be so if by sheer coincidence or due to some strong causal link non-relational inequality always occurs together with relational inequality. Tis is consistent with both having independent normative signifcance. 31  It may be objected that Anderson’s target here is luck egalitarianism only, not distributive egalitarianism in general. However, the feature of luck egalitarianism that she singles out as the basis of the “fundamental” disagreement (i.e. that there is an “independent standard of justice of states of the world” that is not derived “justice of agents”) is not specifc to luck egalitarianism but shared by all distributive conceptions as understood in this chapter. 32   For the emphasis on dispositions, see Anderson’s defnition of the virtue of justice as the “disposition to treat individuals in accordance with the principles that express”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

127

It is worth noticing that this way of contrasting the relational and distributive conceptions is in signifcant tension with the contrast outlined earlier in this section. Te frst shift that occurs compared to the earlier characterization is that while the former takes relations of authority etc. as  primary, the second takes the dispositions of agents as normatively fundamental. Now, it is possible to reconcile the two characterizations by suggesting that relations are unequal just in case the agents do not display and act from the right dispositions in their treatment of others. But this would be a rather implausible conception of egalitarian relations and it is  in  tension with the reference to authority and status in the earlier characterization: inequality of authority and status is compatible with everyone acting from the right dispositions, and equality of authority and status is compatible with not acting from the right dispositions. However, the more important shift is the following. In her frst characterization of the contrast, the signifcance of distributions is presented as instrumental: they matter to the extent that they help to sustain or achieve egalitarian social relations. Tis is a mostly forward-looking view of the importance of distribution. Te latter account, by contrast, holds that distributions are just if they emerge from agents acting from just dispositions. Tis suggests a backward-looking test for evaluating the justness of distributions. Te two can agree that there is no independent standard of justice of distributions, but they diverge on the nature of the correct relation-dependent standard. On the latter view, distributive inequalities are not morally problematic at all but may merely signal the independent injustice of some agents acting from unjust dispositions, i.e. on the basis of principles that reject treating the other as one’s social equal. And insofar as the second account is taken as a necessary and sufcient condition of the justness of distributions, then the two accounts are incompatible with each other. A charitable reading of the various claims that Anderson makes suggests that the instrumental and backward-looking accounts should both be seen as sufcient conditions for an inequality to count as unjust. On this construal, a distributive inequality is unjust if and only if one or both of the following is true: (a) it tends to undermine egalitarian social relations, and/or (b) it is the result of some agent(s), including institutional agents, acting on the basis of unjust attitudes. What is needed at this point is exploring the reasons for denying the relation-independent moral signifcance

social equality. I am construing Anderson’s “just if ” formula in the last quote as “just if and only if ” to convey a necessary and sufcient condition. If it is read as a sufcient condition only, then the intended contrast with the distributive view disappears. Tis construal is supported by the claim that there is no independent standard of the justness of distributions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

128

Zoltan Miklosi

of distributions. Anderson’s stated theoretical reasons are summed up in the following three claims that she subsequently makes:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. Tere can be no injustice without an agent who is (or was) substantively responsible for it—someone obligated to avoid, correct, or bear the costs of the injustice. 2. Tere can be no injustice where all agents continuously and successfully comply with all the demands that everyone can reasonably make of them. 3. Justice as an evaluation applied to states of afairs is entirely derivative of justice as an appraisal of the conduct of agents. (Anderson 2010, p. 5, emphasis in the original)

Tese theoretical claims provide support for relational exclusivism because distributions are states of afairs, therefore if the claims are true then justice as an evaluative term has no nonderivative (relation-independent) application to distributions. It is noteworthy that these theses, when taken together, have wide-ranging implications beyond the rejection of the distributive conception. Tey rule out any conception of justice that endorses principles for the evaluation of states of afairs that are not derived from principles applying to agents’ conduct, and not merely those conceptions that endorse specifcally distributive principles for such evaluation. Furthermore, it appears to me that these three theses, when taken together, create a dilemma for Anderson’s position. On one horn of the dilemma, if we understand the claims in a narrow sense, we get an objectionably restricted account of social equality with implications that Anderson herself appears to reject. On the other horn, if we take a broader view of them, then they do not rule out distributive conceptions of justice after all. Te three claims together state that justice as an evaluative standard is applicable nonderivatively only to the conduct of agents of various types; it is applicable to states of afairs only derivatively, if they are the result of failures of justice of the conduct of agents. However, they are silent on the  question whether other moral predicates, such as “morally bad,” are applicable to states of afairs nonderivatively. It is consistent with the view that only agents’ conduct can be evaluated nonderivatively as just or unjust that states of afairs can be nonderivatively morally bad, and as such they can provide reasons of justice for agents to avoid or correct them. Anderson’s own formulation appears to admit this, and this is also the view of e.g. Rawls, who famously claimed that even though the natural distribution of talents is not in itself unjust, the way social institutions treat these natural facts can be evaluated as just or unjust (Rawls 1999, p. 102). But of course the way social institutions treat these facts can be just or unjust only if these states of afairs provide moral reasons for institutions to treat them in certain ways, i.e. if they can be evaluated as morally bad (or good or neutral)

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

129

nonderivatively.33 Tis is important because if states of afairs can provide moral reasons for agents (nonderivatively), and if their failure to act on these reasons can constitute an injustice, then Anderson’s three claims stating the theoretical grounds of her rejection of distributive conceptions are in fact consistent with distributive principles committed to the nonderivative moral signifcance of distributions. And that is enough. Te distributive theorist need not hold that distributions can be unjust nonderivatively; she needs only the weaker claim that distributions can nonderivatively provide reasons such that failure to act on them constitutes an injustice.34 Terefore, if Anderson’s three claims are to be understood as admitting that states of afairs can be nonderivatively morally bad and as such provide reasons of justice for agents, then they are consistent with distributive conceptions and do not support relational exclusivism. On the other hand, if the three claims are understood as not admitting of the possibility that states of afairs can be nonderivatively morally bad and provide reasons of justice, then they have very implausible implications that are also incompatible with some of Anderson’s stated positions. In discussing an anticipated objection to her view, Anderson invokes the following example:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

[o]rphans sufer an injustice if society has not organized an agency to take care of them . . . Were an epidemic to kill everyone in a position to help the orphans, they would not be sufering an injustice. Tey would be sufering a calamity. (Anderson 2010, p. 5)

Let us suppose that in this example the orphans’ parents’ death was not related to any injustice: they have died of natural causes that no one could have prevented. Te situation where orphans are not taken care of is very bad, but it is not unjust unless there is someone in a position to take care of them. If there is no one who could help, there is no injustice. Fair enough. Now suppose there are agents in a position to take care of the orphans. Anderson claims that they have a duty of justice to take care of them and that if they fail to do so they contribute to an injustice. But note that the very bad situation that they are called to correct has arisen through no unjust action at all. Te actions they are required by justice to perform do not consist in preventing (or correcting the consequences of ) unjust actions. 33   To be sure, I do not suggest that moral reasons are exhausted by evaluative facts as distinct from deontic ones. Te claim here is only that evaluative facts about states of afairs can also provide reasons for actions such that the failure of suitably positioned agents to act on them constitutes an injustice (I am grateful to an anonymous referee for bringing this point to my attention). 34   I am grateful to Janos Kis for a very helpful discussion of this and related points discussed in this section.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

130

Zoltan Miklosi

Terefore, the only way in which Anderson can justify her claim just quoted is by accepting that the state of afairs in question—orphans in need of being taken care of—is nonderivatively morally bad and provides reasons for action such that a failure to act on them by suitably positioned agents is unjust. Were she to reject the possibility that states of afairs can provide such reasons of justice, she would not be in a position to claim that “orphans sufer an injustice if society has not organized an agency to take care of them.” And since it would be absurd to deny the injustice of society’s failure to organize orphanages, it would be equally absurd to reject the theoretical premise.35 But if she accepts it, then her three claims do not rule out distributive conceptions. Furthermore, in the passage just quoted above, Anderson acknowledges not merely that states of afairs can be nonderivatively morally bad and provide reasons of justice to agents, but also that even if there is no currently existing agent that could act on the relevant reasons, states of afairs can nonetheless provide reasons of justice to create or organize the appropriate collective institutional agents when this is necessary. Tis entails that collectives of people or society have a collective duty of justice to organize themselves in such ways to enable themselves to eliminate certain morally bad states of afairs. Terefore, on this view, the fact that no currently existing agent is in a position to prevent or correct a morally bad state of afairs does not signal the absence of reasons of justice to correct the state of afairs by  creating the necessary collective agent. All this seems to be obviously correct to me. But then, nothing that Anderson claims rules out that certain distributions in particular could be among the states of afairs that can provide such reasons. Now, to be sure, it is equally true that nothing that has been argued in this section provides a positive argument to the efect that some distributions are among the states of afairs that provide reasons of justice for agents such that failure to act on them constitutes an injustice. It is still open to Anderson to argue to the contrary. Te mere fact that the three theoretical claims she makes are consistent with the relation-independent signifcance of distributions does not imply that distributions have in fact such signifcance from the point of view of justice. But if the arguments in this section are sound, then one supposedly important theoretical motivation for thinking that they do has been removed. It is incumbent on Anderson to provide substantive arguments for her claim. 35   It might be suggested that the duty to organize orphanages is a humanitarian one, and not one of justice, if we take duties of justice as involving the possibility of reciprocity. I am skeptical of this claim (see fn 19). Note also that Anderson herself identifes this duty as one of justice.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

131

In §6 I provide some considerations suggesting that distributions should be among the features of states of afairs that can provide reasons of justice for action.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

5.  RELATIONS AND DISTRIBUTIONS If we put aside the theoretical arguments for relational exclusivism discussed in §4, then relational exclusivism must provide a substantive account of the moral signifcance of egalitarian relationships. Furthermore, that account must show not only that relations matter from the point of view of justice but also that nonrelational interests do not (independently) matter from that point of view. Relational egalitarians are often vague on the nature of the signifcance of egalitarian relations.36 Generally speaking, egalitarian relationships may be morally signifcant because they have personal value, i.e. they are good for the people who participate in them, making their lives better in certain ways (cf. Gheaus 2016). Egalitarian relationships may also be thought to have impersonal value, such that their existence is good independently of their contribution (if any) to the goodness of people’s lives, in a similar way to the way many take the existence of the Grand Canyon or great works of art to be valuable (Tomlin 2015, pp. 160–1). Both of these views suggest that there are at least pro tanto reasons for people to create and sustain egalitarian relationships where they are absent. A third view of the signifcance of egalitarian relations does not necessarily have this implication: it does not see their signifcance in their contribution to personal or impersonal value, but in their being the ftting response to the equal moral status of each person as a being with a certain kind of moral authority. On this view, equality in relationships is signifcant not (or not only) as something to be promoted, but as a constraint, something that is to be respected. Te signifcance of status is distinct from that of well-being, of  what is valuable for persons and is related to the value or worth of persons.37 On this third view, we owe it to others not to relate to them in inegalitarian ways, as that would be wrong in light of their equal status and equal moral authority, but there is no implication that we owe it to others to establish egalitarian relationships with them where they are absent, just on account of the signifcance of such relations. To put it diferently, on this view we owe it to others that if we relate to them at all, we do so on egalitarian terms, but we do not owe it to them to establish egalitarian social relationships with them in the frst place. On the third view, that is, it is not 36

  For an illuminating discussion, see Patrick Tomlin (Tomlin 2015).   See e.g. Tomas Nagel’s point (Nagel 2008).

37

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

132

Zoltan Miklosi

necessarily better (personally or impersonally) if more egalitarian relating occurs, but it is wrong to relate to others in inegalitarian ways.38,39 Needless to say, these three views of the signifcance of egalitarian relationships are fully compatible with one another. In fact, it seems obvious that many egalitarian relationships have the potential to make the participants’ lives better, and that equality is a constraint on relationships (I confess that I fnd the idea that egalitarian relationships have impersonal value harder to grasp, but certainly not absurd). Furthermore, refecting on the plurality of ways in which equality in relationships may be morally signifcant highlights an additional complexity that is often obscured in relational egalitarian views: they tend to discuss intimate, personal relationships and broader socialpolitical ones together, suggesting that equality plays the same normative role(s) in all of them. Yet it seems plausible to me that in diferent kinds of relationships equality has diferent kinds of signifcance. In close, personal relationships between particular persons who are identifable to each other and play a signifcant role in each other’s life, it is the contribution to personal value that seems to dominate, even though equality is still signifcant as a constraint.40 In anonymous social relations among large numbers of persons, which are mediated by social institutions, equality is primarily signifcant as a constraint and may make a smaller contribution to the goodness of persons’ lives. Relations of the latter type are not established as part of our conception of the good life, but rather we necessarily fnd ourselves in their midst, and as long as we are part of them we ought to strive to make them egalitarian. We owe this to others not primarily as a matter of concern for their well-being, but in acknowledgment of their equal moral status, involving a certain kind of authority over what others may do to them, including how others may relate to them (Nagel 2008, p. 111, Anderson 2010, p. 3). Tese remarks leave many details to be worked out. However, my main concern here is not a precise characterization of the signifcance of egalitarian relations, but rather the following question: what do these accounts imply about the signifcance of nonrelational interests from the point of view of 38   Tat is, the third account of the signifcance of egalitarian relations is at root deontic rather than axiological: its objection to inegalitarian relationships is not that egalitarian ones contribute to better outcomes. See e.g. Schemmel (Schemmel 2011, p. 366). 39   Tis view is consistent with the idea that we are morally required to establish social relations when they are necessary for discharging some of our independent duties, and once those relations exist we ought to make sure that they are egalitarian. A prominent example of such an independent duty could be Rawls’ natural duty to create and support just political institutions (Rawls 1999, pp. 114–17). 40   I leave it open whether only egalitarian relationships can contribute to well-being or  equality only adds to the value of personal relationships—i.e. whether its role is constitutive or additive for personal value.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Varieties of Relational Egalitarianism

133

justice? Do they provide support for relational exclusivism, or even for the more moderate view that the relation-independent signifcance of nonrelational interests is relatively minor? It strikes me that the answer depends on more general issues about the nature of justice as an ideal, and in particular whether a concern for personal well-being is part of that ideal. If it is, and if  the signifcance of egalitarian relationships consists primarily in their contribution to the goodness of people’s lives, then it is hard to see why egalitarians should be concerned only with this aspect of personal well-being, with complete disregard of the contribution that nonrelational projects make to it (Tomlin 2015, pp. 172–3).41 Such a restriction is arbitrary, given the widely accepted importance of nonrelational goods and projects—such as the appreciation of esthetic value, the pursuit of adventure and excitement, the search for scientifc knowledge or philosophical understanding, the performance of challenging and complex tasks, etc.—as parts of plausible conceptions of the good life.42 Without in the least denying the great importance most people attach to relational goods of romantic love, family, friendship, and so on, it would be absurd to deny the centrality of the nonrelational goods just mentioned in many reasonable conceptions of the good. It is difcult to see a principled rationale as to why justice should be concerned with access to the former but not with access to the latter types of goods, if it is concerned with well-being at all. And if that is so, then this provides the answer to a question raised in the introduction of this chapter. To recall, I entertained the objection that relational egalitarianism can pose a serious challenge to the distributive view not only if it takes the form of relational exclusivism, but also if it insists that the relation-independent signifcance of distributions is rather minor compared to the independent signifcance of egalitarian relations. But if the primary signifcance of egalitarian relations is in their contribution to the goodness of persons’ lives, then it is difcult to see how this kind of challenge can be successful. Te contribution of nonrelational goods to people’s lives appears to me in every respect on a par with that of relational goods. 41   I want to stress that taking individual well-being as a concern for justice is neutral between welfare-based and resource-based views regarding the “currency” of justice: resource-based views also recognize that the signifcance of resources consists in their potential to improve people’s lives. Teir reasons for insisting on resource as the proper currency are independent of this point. See e.g. Dworkin’s remarks to this efect in his classic defense of resource equality (Dworkin 2000, p. 14). 42   To be sure, these non-relational goods can be pursued in cooperation with others, and being pursued in that way may add to their value. However, it is not a necessary condition of their being valuable that they are pursued in this manner. My understanding of non-relational goods is consistent with Stemplowska’s, who writes that “nonrelational goods are understood . . . to be goods that people can have independently of how others relate to them, in the specifc sense that how others relate to them is not part of the defnition of having the good” (Stemplowska 2011, p. 116).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

134

Zoltan Miklosi

Terefore, if this is the primary signifcance of egalitarian relations for justice, then I cannot see why the similar signifcance of nonrelational goods should not also have similar weight for justice. If non-relational goods have any signifcance for justice on account of their contribution to well-being, then it is unlikely to be only a minor one. Tis leaves us with the alternative suggestion that the concern with wellbeing is not a matter of justice. Ten, the relational egalitarian might claim that egalitarian relations have some further moral signifcance, unrelated to their efects on well-being, that nonrelational goods lack, and that that signifcance is a matter of justice. Tis thought points towards the third view noted above, which claims that equality in relationships is signifcant (for justice) as a constraint, as a proper response to people’s moral status, involving a certain kind of authority, as distinct from their well-being. Now I turn to the question whether this view can plausibly give support to relational exclusivism. Te pertinent question in this regard is whether this sort of concern constitutes the exclusive focus of justice. Tis is of course a much broader issue than could be properly addressed here. I will suggest only one reason to think that the answer is unlikely to be afrmative. Te reason is this: one of the central domains in which persons exercise their rational agency and authority is in developing and pursuing their conception of the good life. If people’s equal status as rational agents with moral authority commands respect in structuring social relations so that each is treated as a social equal, then it seems similarly compelling that we take it as a matter of great importance whether they have fair opportunities to exercise this authority in shaping their own lives. It appears to me unmotivated to attribute exclusive signifcance to persons’ being treated with the respect that is owed to them as social equals with moral authority, and at the same time to regard it as a matter of indiference (from the point of view of justice) whether they have a fair chance of exercising that same authority in their personal goals and projects. It would seem that acknowledging the importance of their nature as rational agents with moral authority requires both treating them as social equals and showing proper concern for their opportunities for shaping their own lives. Terefore, the signifcance of equality in relationships as a constraint is best seen as only one aspect of justice among others and as such it is unlikely to provide support for relational exclusivism.43 43   To be sure, what having a fair chance amounts to will be controversial. Te point here is simply that the signifcance of having a fair share of realizing one’s conception of the good is not derived from the signifcance of egalitarian social relations, or of equal social status. A point of clarifcation: equality of social status is a fundamentally relational ideal. By contrast, the more basic idea of equal moral status invoked in this section is non-relational and is grounded in the possession of rational capacities that endow one with a certain moral authority. Terefore, the suggestion that people are owed, on the

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

135

I do not expect these remarks to be compelling. Te more important point that emerges from this brief discussion of the signifcance of status is the following. For an account of the signifcance of equality in relationships to support relational exclusivism, it has to insist not merely that egalitarian relationships contribute to well-being, nor even that moral status, as distinct from well-being, is morally signifcant, but that status is the only thing that matters from the point of view of justice. Tis is so because if the account focuses on the contribution that egalitarian relationships make to well-being, then it must admit the similar signifcance of nonrelational goods, which would validate the relation-independent signifcance of the distribution of socioeconomic resources. Te only way to avoid this conclusion is to resist the idea that well-being matters from the point of view of justice. But this move strikes me as very implausible.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

6.  CONCLUDING THOUGHTS It seems to me that the important motivating insight of relational egalitarianism is the thought that our status as moral equals commands a kind of concern towards each that is at least partly independent of a concern of how well our lives as individuals go, and which imposes genuine constraints on the ways we may permissibly relate to each other, both in our direct, personal interactions and in social relations that are structured by institutions. However, taking this insight on board does not lend support to  relational exclusivism. It is consistent with acknowledging the vast importance, from the point of view of justice, of each person’s having fair opportunities to pursue their conception of the good, and therefore of the relation-independent signifcance of the distribution of socioeconomic resources. As I see it, the intuitive appeal of this thought suggests that the burden of proof is on the relational exclusivist to show otherwise. A more attractive view is that relational concerns constitute only one aspect of justice, supplementing direct distributive concerns. In my view, an attractive egalitarian conception of justice can incorporate relational equality as a constraint on permissible social (and interpersonal) relations, while at the same time showing proper concern for each person’s fair opportunity for pursuing both relational and nonrelational goods. Tus, the signifcance of  egalitarian relations may fgure in multiple ways in such a view: as components of reasonable conceptions of the good life, and as constraints basis of their equal moral status, a fair chance of realizing their conception of the good, is not committed to relational egalitarianism (I thank an anonymous referee for pressing me on this point).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

136

Zoltan Miklosi

on relations in response to our equal moral status. Yet these two aspects of the signifcance of egalitarian relationships still do not exhaust all the justice-related signifcance of distributions. Furthermore, a full appreciation of the diferent ways in which equality in relationships may be signifcant for justice does not imply the lesser signifcance of nonrelational goods from the point of view of justice.44

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Elizabeth Anderson, “What Is the Point of Equality?” Ethics 109: 2 (1999): pp. 287–337. Elizabeth Anderson, “Te Fundamental Disagreement between Luck Egalitarians and Relational Egalitarians,” Canadian Journal of Philosophy 40: suppl. (2010): pp. 1–23. Richard Arneson, “Luck Egalitarianism and Prioritarianism,” Ethics 110: 2 (2000): pp. 339–49. Allen Buchanan, “Justice as Reciprocity versus Subject-Centered Justice,” Philosophy and Public Afairs 19: 3 (1990): pp. 227–52. G.A. Cohen, Why Not Socialism? (Princeton, NJ: Princeton University Press, 2009). Chiara Cordelli, “Justice as Fairness and Relational Resources,” Journal of Political Philosophy 23: 1 (2015): pp. 86–110. Ronald Dworkin, Sovereign Virtue (Cambridge, MA: Harvard University Press, 2000). Ronald Dworkin, “Equality, Luck and Hierarchy,” Philosophy & Public Afairs 31: 2 (2003): pp. 190–8. Anca Gheaus, “How Much of What Matters can We Redistribute? Love, Justice, and Luck,” Hypatia 24: 4 (2009): pp. 63–83. Anca Gheaus, “Hikers in Flip-fops: Luck Egalitarianism, Democratic Equality, and the Distribuenda of Justice,” Journal of Applied Philosophy, doi: 10.1111/japp.12198 (2016). Aaron James, “Te Signifcance of Distribution,” in R. Jay Wallace, Rahul Kumar, and Samuel Freeman (eds.), Reasons and Recognition: Essays on the Philosophy of TM Scanlon (Oxford: Oxford University Press, 2011), pp. 276–304. Tomas Nagel, “Te Problem of Global Justice,” Philosophy and Public Afairs 33: 2 (2005): pp. 113–47. Tomas Nagel, “Te Value of Inviolability,” in Paul Bloomfeld (ed.), Morality and Self-Interest (Oxford: Oxford University Press, 2008), pp. 102–14. 44   I received valuable comments on an early draft of this chapter from Charles Beitz, Chiara Cordelli, and Lucas Stanczyk. A later draft was presented at the Oxford Studies Workshop in Political Philosophy in Barcelona, in June 2016, where useful feedback was given by several members of the audience, including in particular by Ian Carter, Tomas Christiano, David Estlund, Serena Olsaretti, and Alex Voerhoove. Special thanks are due to Tom Parr, my commenter at the workshop, for his detailed commentary. I am grateful to Janos Kis for very illuminating discussions on several parts of the chapter, and to the editors of the volume and two anonymous referees for several important suggestions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Varieties of Relational Egalitarianism

137

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

John Rawls, A Teory of Justice, revised edition (Cambridge, MA: Harvard University Press, 1999). Samuel Schefer, “What is Egalitarianism?” Philosophy and Public Afairs 31: 1 (2003): pp. 5–39. Samuel Schefer, “Choice, Circumstance, and the Value of Equality,” Politics, Philosophy and Economics 4: 1 (2005): pp. 5–28. Samuel Schefer, “Te Practice of Equality,” in Carina Fourie, Fabian Schuppert, and Ivo Wallimann-Helmer (eds.), Social Equality: On What It Means to Be Equals (Oxford: Oxford University Press, 2015), pp. 1–21. Christian Schemmel, “Why Relational Egalitarians Should Care about Distributions,” Social Teory and Practice 37: 3 (2011): pp. 365–90. Christian Schemmel, “Distributive and Relational Equality,” Politics, Philosophy & Economics 11: 2 (2012): pp. 123–48. Zofa Stemplowska, “Responsibility and Respect: Reconciling Two Egalitarian Visions,” in Carl Knight and Zofa Stemplowska (eds.), Responsibility and Distributive Justice (Oxford: Oxford University Press, 2011). Patrick Tomlin, “What Is the Point of Egalitarian Social Relationships?” in Alexander Kaufman (ed.), Distributive Justice and Access to Advantage: G.A. Cohen’s Egalitarianism (Cambridge: Cambridge University Press, 2015), pp. 151–79. Jonathan Wolf, “Fairness, Respect, and the Egalitarian Ethos,” Philosophy and Public Afairs 27: 2 (1998): pp. 97–122.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Pa rt I I I

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

L I B E RT Y A N D C O E RC I O N

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved. Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

6

Moralizing Liberty Ralf M. Bader

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. Introduction Tere are some things we have a right to do.1 Tese things we are free to do. Tere are other things that we do not have a right to do. Tese things we are not free to do. Tis connection between rights and freedom2 lies at the basis of moralized or rights-based conceptions of freedom, which form a subfamily of negative conceptions of liberty.3 Moralized accounts distinguish liberty from license and do not indiscriminately consider all interferences as being inimical to liberty. In particular, liberty is not infringed when people are prevented from performing actions that they do not have a right to do, since they are not free to do those things in the frst place. Nonmoralized conceptions, in contrast, consider liberty to be a purely descriptive concept, the analysis of which does not make reference to rights or other moral notions, and hence do not diferentiate interferences on the basis of whether or not the person being interfered with has a right to perform the action in question.4 1  Rights, which are throughout to be understood as moral rights, include not only claim rights but also liberty rights (Hohfeldian privileges). 2   ‘Liberty’ and ‘freedom’ will be treated as synonyms. 3  Tis chapter is concerned with negative conceptions that consider liberty as an opportunity-concept and that are to be distinguished from positive conceptions that treat it as an exercise-concept. Moralized accounts are not the only negative conceptions of liberty that invoke moral or evaluation notions. Responsibility-based views (such as those espoused by Miller: 1983 and Kristjánsson: 1996) appeal to the notion of moral responsibility in specifying which interferences can be classifed as constraints on freedom (cf. Bader: 2016), whilst accounts (such as Kramer’s) bring in evaluative notions when it comes to determining the extent of an agent’s overall freedom. 4   Tis does not imply that nonmoralized conceptions have to consider all interferences with all actions as infringements of liberty. Restrictions can be imposed both on the range of interferences that classify as constraints, e.g. interferences can be restricted to those that result from intentional human agency, and on the kinds of actions that agents can be free

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

142

Ralf M. Bader

Whilst the rights-based approach used to have some prominent proponents, having been endorsed and developed by Locke, Kant, Nozick, and Dworkin amongst others, it has recently come under sustained criticism and has largely fallen out of favor. Tis chapter will argue in support of moralizing liberty on the grounds that distinguishing between liberty and license allows us to develop a theoretically fruitful notion of freedom that is intrinsically normatively signifcant and that can play a substantive role in political philosophy. Section 2 argues that the contrast between liberty and license is to be understood in terms of a moralization of the z-parameter, whereby the domain of this parameter consists of permissible courses of action. Section 3 defuses the prisoner objection, which is frequently taken to be one of the primary reasons for rejecting moralized accounts. Section 4 argues that only moralized conceptions of liberty can underwrite the presumption of liberty by providing us with a notion of freedom that is intrinsically normatively signifcant. 2.  Liberty Versus License Attributions of liberty can be understood in terms of the following triadic schema (cf. MacCallum: 1967):

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

agent x is free from constraints y to do/be/become z Difering conceptions of freedom disagree about the specifcation of the  domains of these parameters, i.e. to which subjects freedom is to be attributed (= x), which obstacles classify as constraints on freedom (= y), and which actions an agent can be free or unfree to perform (= z).5 Using this schema, we can distinguish between an agent being free, being unfree, and being not-free to perform action ϕ.6 – x is free to ϕ if ϕ is in z and nothing in y is preventing x from ϕ-ing. – x is unfree to ϕ if ϕ is in z and something in y is preventing x from ϕ-ing. – x is not-free to ϕ if ϕ is not in z.7 to perform, e.g. a nonmoralized trivalent conception such as an ability-dependent theory, will restrict the domain of the z-parameter to those actions that the agent in question is able to perform. (For arguments to the efect that the justifcation and precise specifcation of the requisite restrictions of the y-parameter cannot proceed without bringing in moral notions cf. Miller: 1983; Miller: 1985; Shnayderman: 2013, §§1–3; and Bader: 2016.) 5   Since we are concerned with negative conceptions of liberty, cases where z involves being/becoming can be set aside. Te domain of the z-parameter accordingly is a domain of actions. 6  An agent being free simpliciter can be understood in terms of being free to ϕ for all ϕ in z. 7   Te locution ‘x is not-free to ϕ’ is not equivalent to ‘it is not the case that x is free to ϕ’. As long as ϕ is restricted to actions, it is equivalent to ‘x is neither free nor unfree to ϕ’.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Moralizing Liberty

143

Te domain of the z-parameter specifes the possible actions that an agent can be free or unfree to perform. Any action that is not in this domain is not in the relevant sense a possible action in the frst place and is hence not one to which freedom is applicable, which ensures that an agent can be neither free nor unfree to perform such an action. Te distinction between freedom and unfreedom is applicable only within the domain of the z-parameter, i.e. amongst ‘possible’ actions. Rights-based conceptions consider only rights-violating interferences as infringements of liberty. A rights-violating interference is an interference with an action that (i) an agent has a right to perform, and that (ii) violates this right.8 Tis implies that such accounts involve moralized characterizations of both the y- and the z-parameter, giving rise to restrictions with respect to the source of the obstacle as well as with respect to the type of action that is being interfered with.9 In particular, they distinguish liberty from license when characterizing the domain of the z-parameter (cf. Locke: 1689, §6), insofar as they consider this domain to consist only of legitimate courses of actions, i.e. one can only be free to do things one has a right to do. Te relevant notion of possibility that specifes which actions an agent can be free or unfree to perform is thus normative possibility. Actions that are not normatively possible, i.e. impermissible actions, are not ones that one is free to perform (where one is not unfree to perform them either, but where one is rather neither free nor unfree to perform such actions). As such, they are not in the domain of the z-parameter and are not part of liberty, but instead belong to mere license. Te restriction to normatively possible courses of action has the efect that certain obstacles do not infringe liberty. In particular, any obstacle that interferes with a normatively impossible action will not be a constraint on freedom. Tis is because actions that are not normatively possible are ones to which freedom is not applicable and hence ones which one cannot be Whilst one can understand not-free in terms of neither free nor unfree, this does not imply that these notions are conceptually prior, nor that not-free is essentially negative. Since free, unfree, and not-free exhaust the domain of actions, any two can be used to defne the third, e.g. one can defne ‘free to ϕ’ as ‘neither unfree nor not-free to ϕ’. In fact, not-free can be characterized positively by the moralized theorist insofar as it is to be identifed with license. 8   Whilst both claim rights and liberty rights satisfy condition (i), only interferences with claim rights can satisfy condition (ii). Tis is because there is no such thing as a violation of a liberty right. Tis means that, though mere privileges belong to the domain of liberty as opposed to license, there are no rights-violating interferences when it comes to privileges. Tis, in turn, implies that it is not possible to be rendered unfree to perform an action that one is merely at liberty to do and which is not protected by claim rights. Interferences with mere privileges, accordingly, do not result in unfreedom but only in inability. 9   Tis chapter focuses only on the latter. For an account of the former cf. Bader: 2016.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

144

Ralf M. Bader

rendered unfree to perform. Put diferently, if the agent is prevented from doing something that he is not free to do in the frst place, owing to his not having a right to do that thing, then the agent’s freedom has not been infringed. Accordingly, only rights-violating interferences infringe liberty, whereas preventing someone from doing something that person does not have a right to do does not infringe freedom. Instead of these kinds of interferences infringing liberty, they merely infringe license.

2.1  Justifed Infringements

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Moralized accounts are frequently characterized as classifying only unjustifed interferences as infringements of liberty. Tis construal locates the moralization in the y-parameter, distinguishing justifed from unjustifed obstacles and considering only the latter, but not the former, to be constraints on freedom. Whilst rights-based views do involve a moralization of the y-parameter, the contrast between liberty and license that is at issue here, and which gives rise to the distinctive commitment that liberty is not infringed when preventing someone from doing something that he does not have a right to do, is to be understood in terms of the z-parameter. Te construal in terms of justifed v. unjustifed interferences is problematic for four reasons. 1. Te y-parameter characterization is not extensionally adequate if there can be justifed rights violations, since this ensures that there can be justifed infringements of liberty. If there is no absolute prohibition on rights violations, then x can be permissibly prevented from ϕ-ing despite x having a right to ϕ. Tis requires that rights are not considered as absolute side-constraints that render interferences inadmissible and that one rejects the equivalence between preventing x from ϕ-ing being illegitimate and ϕ-ing being something that x has a right to do.10 If such an understanding of rights is accepted, it becomes possible for there to be cases where x has a right to ϕ, but where y can nonetheless be justifed in preventing x from ϕ-ing, such that x can be rendered unfree in a permissible manner. Tis means that there can then be justifed unfreedom. Allowing for justifed infringements of liberty does not in any way afect the core commitment of rights-based accounts that if x lacks a right to ϕ, then x is not-free to ϕ and preventing x from ϕ-ing does not constitute an 10   Cf. “Rightful actions—actions that are exercises of rights—are thus both permissible and inviolable. Actions interfering with or preventing the performance of rightful actions are themselves impermissible” (Steiner: 1977, p. 767).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Moralizing Liberty

145

infringement of freedom. However, if there can be interferences that are justifed despite x having a right to ϕ, it will no longer be the case that justifably preventing x from ϕ-ing never constitutes an infringement of liberty. Instead, preventing x from ϕ-ing constitutes an infringement of x’s liberty if doing so involves a violation of x’s rights (independently of whether or not this is justifed). Tis means that, rather than liberty being infringed only by unjustifed interferences, liberty is infringed by interferences with what one has a right to do. Te operative distinction, accordingly, is that between interferences that violate rights and those that do not, rather than that between unjustifed and justifed interferences. Even those who endorse a strong connection between (moralized) liberty and (moralized) voluntariness, claiming that voluntariness is not undermined and liberty is not infringed as long as people act within their rights (cf. Nozick: 1974, p. 262), can retain their core commitments and only need to slightly restate their position.11 Although allowing for justifed infringements makes it possible that y, whilst acting within his rights, infringes x’s freedom and makes x’s action nonvoluntary (for instance, if x has the right to ϕ or ψ and y is justifed in preventing x from ψ-ing in order to avoid a catastrophic moral horror, then x’s ϕ-ing may well end up being nonvoluntary as a result of y’s interference), this problem turns out to be superfcial (contra Cohen: 1995, p. 63). Tis is because one can allow for justifed coercion, i.e. y can be justifed in acting in a way that undermines x’s voluntariness, in the same way as one can allow for justifed unfreedom. Accordingly, the claim can be restated in a straightforward manner insofar as liberty is not infringed, and voluntariness is not undermined as long as no rights are being violated. Tere is thus a simple restatement in terms of behavior that does not involve any rights violations rather than in terms of justifed behavior. 2. Te fact that it is conceptually possible for the y-parameter characterization to be extensionally inadequate implies that this way of understanding moralized conceptions is not perspicacious and does not carve nature at its joints, even in those cases where the characterization in terms of justifed v. unjustifed interferences is extensionally adequate (because unjustifed interferences are treated as being equivalent to interferences that violate the agent’s rights). Tis is because it mistakenly suggests that, by their very nature, moralized conceptions cannot allow for justifed infringements of liberty, which, however, is not the case, but only follows given certain additional commitments that are not integral to a moralized theory. Te fundamental characterization of rights-based accounts, accordingly, has to focus on the agent’s rights rather than the 11

  For criticisms of this understanding of voluntariness cf. Olsaretti: 1998.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

146

Ralf M. Bader

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

legitimacy of the interference and is, accordingly, to be understood in terms of the z-parameter, i.e. in terms of interferences with actions that the agent has a right to perform versus actions that she lacks a right to perform. 3. Te y-parameter characterization will misclassify Hohfeldian privileges as belonging to license. If the operative distinction is between justifed and unjustifed interferences, and only the latter are considered to be infringements of liberty whereas the former are classifed as infringements of mere license, then privileges will mistakenly be classifed as belonging to mere license. Tis is because, unlike claim rights, mere privileges do not have correlative duties. As a result, agents can justifably be prevented from exercising them. Yet they do not belong to license but are part of liberty. Tis is achieved by the z-parameter characterization, since it understands the domain of liberty as the domain of permissible courses of actions, which includes mere privileges alongside actions protected by claim rights. 4. Moralizing the y-parameter and focusing on the legitimacy of interferences misleadingly suggests that proponents of moralized conceptions are concerned only with unfreedom resulting from unjustifed interferences. Tis misses an important dimension of the concern for liberty, insofar as it focuses only on infringements of liberty and ignores unjustifed restrictions of liberty. Tere are two ways in which something can be inimical to liberty, namely insofar as an agent can be rendered unfree to ϕ or not-free to ϕ. infringement of liberty: x’s liberty is infringed if x is rendered unfree to ϕ. restriction of liberty: x’s liberty is restricted if x is rendered not-free to ϕ.

Te latter way of being inimical to liberty has nothing to do with unjustifed interferences that prevent an agent from ϕ-ing, yet nevertheless constitutes an important issue of concern. If x is not prevented from ϕ-ing but ϕ-ing is illegitimately rendered normatively ineligible, such that there is a switch from x being free to ϕ to being not-free to ϕ, then x’s liberty is impermissibly restricted, owing to this impermissible shrinking of the domain of the z-parameter (cf. Bader: manuscript).

2.2  Legitimate Courses of Action Specifying the domain of the z-parameter amounts to specifying the possible courses of action. Once they have been specifed, one can then determine which of them are prevented, i.e. those that one is unfree to do, and which

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

147

are not prevented, i.e. those that one is free to do.12 Te ratio of those that are not prevented to the set of possible actions yields the degree of freedom. – x is free to ϕ if ϕ ∈ z and nothing in y is preventing x from ϕ-ing – x is free to degree n if |ϕ: x is free to ϕ|/|ϕ: ϕ ∈ z| = n – x is free simpliciter if x is free to ϕ for all ϕ ∈ z, i.e. x is free to degree = 1

An important problem arises for trivalent conceptions of freedom (that countenance ‘not-free’ in addition to ‘free’ and ‘unfree’) if the notion of possibility that is used in specifying the domain of the z-parameter gives rise to a variable domain.13 If the domain can be modifed, then there is the risk that it can be manipulated. Tis has the problematic implication that one can become more free as a result of a restriction of the domain of the z-parameter.14 Te restriction can be internal, i.e. on the part of the agent, as well as external, i.e. on the part of other people.

Desires Te problem is particularly pronounced when the domain is construed in terms of desires, i.e. free if not prevented from doing what one wants to  do.15 Desires can be externally restricted through brainwashing and indoctrination, as well as internally restricted if an agent can extinguish his desires (what Berlin: 1958 calls the ‘retreat to the inner citadel’). Tis leads to the problem of desire-dependence.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

desire-dependence: a person who is prevented from ϕ-ing can give up the desire to ϕ and thereby make himself more free/can be brainwashed into no longer desiring to ϕ and thereby be made more free.

It is worth noting that it is not the case that the agent becomes free to ϕ as a result of the change in desires, i.e. we do not have a switch from unfree to ϕ to free to ϕ. Instead, there is a switch from unfree to ϕ to not-free to ϕ, since ϕ-ing is no longer in the domain of the z-parameter. Te agent

12   If inability does not classify as a constraint on freedom, then the set of actions one is able to do can be a subset of the set of unprevented actions, i.e. one can be free to ϕ yet not able to ϕ. On a moralized account, it is also possible that one is able to ϕ but that ϕ-ing is not a possible (= legitimate) action, i.e. one can be able to ϕ without being free to ϕ. Likewise, on a desire-dependent account, one can be able to ϕ yet not desire to ϕ and hence not be free to ϕ. 13   Not all trivalent conceptions give rise to a variable domain. For instance, an account in terms of nomological possibility generates a trivalent conception with a fxed domain. 14  Restricting the domain to increase freedom classifes as a form of ‘moral avoidance’ (cf. Nozick: 1981, pp. 460–2). 15   Berlin initially held this view, but abandoned it in response to Wollheim’s critique.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

148

Ralf M. Bader

accordingly becomes more free, since the ratio of actions that the agent is free to perform relative to the total number of actions in z increases.

Abilities While many people reject accounts in terms of desires for precisely this reason, this problem is usually considered to be relatively superfcial and it is thought that it can be dealt with by switching from desires to abilities, i.e. free if not prevented from doing what one is capable of doing.16 Te problem, however, runs deeper. In the same way that desire-dependence yields objectionable results, insofar as freedom can be increased, not by removing obstacles or interferences, but by modifying desires, so abilitydependence yields problematic results, insofar as freedom can again be increased, not by the removal of obstacles or interferences, but by modifying abilities.17 In particular, abilities can be restricted, both internally and externally.18 ability-dependence: a person who is prevented from ϕ-ing can make himself incapable of ϕ-ing and thereby make himself more free/can be rendered incapable of ϕ-ing and thereby be made more free.

Cohen has suggested that every theory involves a form of ability-dependence since

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

interference is never sufcient for unfreedom because lack of means is always necessary for it. Tat is so, because, absolutely generally, an interference restricts my freedom to do x only if I lack the means to overcome that interference, and, therefore, the ability to do x despite that interference.  (Cohen: 2011, p. 195)19

First, this variety of ability-dependence is unproblematic and to be sharply distinguished from that involved in characterizing the z-parameter in terms of abilities. Tis is because Cohen’s version only allows for rendering oneself unfree, insofar as changing one’s abilities can ensure that a certain interference does make an action impossible. However, it does not make room for rendering oneself more free, on the grounds that if one can acquire 16  Cf. “A person’s abilities and inabilities, rather than her preferences, form the boundaries of her freedom and unfreedom” (Kramer: 2003, p. 34). 17  Additionally, problems arise if one wants to distinguish between ability and freedom, something that is not possible if the z-parameter is characterized in terms of abilities. 18   Kramer does not straightforwardly identify the domain of the z-parameter in terms of abilities, since he wants to treat inabilities that are self-inficted or brought about by nature diferently from those that are attributable to other agents. As a result, his account does not allow for external restrictions, since he considers them to result in unfreedom rather than mere inability (cf. Kramer: 2003, p. 42). 19   He attributes this point to Arnold Zubof.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

149

an ability, then the relevant action is already possible for the agent (i.e. possibility claims are governed by a collapse principle such that if x is at t in a position to acquire at t′ the ability to overcome the interference to ϕ-ing, then it is already possible at t for x to ϕ at t′, such that the interference does not render ϕ-ing at t′ impossible).20 Even if this were possible, there would not be any problems, since this would render the interference inefective; this difers from the problematic situation whereby the agent is classifed as being free despite the interference still being present and despite the agent still not being in a position to perform the action. It is this latter situation that involves moral avoidance, since the morally problematic interference is  still present but is not classifed as such because the domain of the z-parameter has been shrunk. Second, Cohen’s claim is problematic insofar as ϕ-ing will (at least in ordinary cases) involve diferent actions when it requires overcoming the interference than ϕ-ing does in the absence of the interference. Even if the same outcome can be brought about, this will be achieved by diferent means. Tis is because one needs to do something diferent and/or do something in addition in order to overcome the interference. Tis means that, although interference with ϕ-ing is not sufcient to ensure unfreedom to ϕ, it is nevertheless sufcient to ensure some unfreedom, whether this is unfreedom to ϕ in a particular way or unfreedom to perform the conjunctive action of ϕ-ing without ψ-ing.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Rights According to moralized conceptions of freedom, the z-parameter ranges over all legitimate courses of actions. Given that normative possibilities can change, insofar as people can waive, alienate, or forfeit rights, it would appear that the moralized account does not avoid the problem at hand. Likewise, others can change what rights an agent has, at least when they possess the relevant normative powers to do so. Te problem of rights-dependence would seem to arise in an analogous manner. For instance, one can make oneself more free by alienating rights that are being infringed. rights-dependence: a person who is prevented from ϕ-ing can waive, alienate, or forfeit the right to ϕ and thereby make himself more free/can have his right to ϕ taken away and thereby be made more free.

Rights-dependence difers crucially from the other forms of dependence. What is problematic about desire- and ability-dependence is that what is wrong, namely the interference, is still in place. As a result of the change in 20   Te only thing that can happen is that other people increase one’s abilities and make it possible for one to overcome various interferences.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

150

Ralf M. Bader

desires/abilities it simply is not classifed as such any more. Tis is what makes these construals of the domain of the z-parameter problematic, i.e. they allow that the objectionable situation still persists but is no longer classifed as involving an infringement of liberty.21 For instance, being in chains is problematic even when one does not want to move around or does not have the ability to do so. Such an obstacle is objectionable and in need of justifcation even when the agent lacks the relevant desires/abilities. In particular, it seems intuitive to criticize the obstacle on the basis of its being an undue limitation of the agent’s freedom. Tis kind of interference is something that one should be concerned about when one cares about freedom and, accordingly, should be classifed as an infringement of liberty. Tis kind of problematic situation, however, cannot arise in the case of rights. If a right is waived, alienated, or forfeited, then there is no longer any objectionable interference. Although the relevant obstacle can still be present and although it can, correspondingly, still be impossible for the agent to ϕ, this obstacle will no longer be objectionable, since the agent will no longer have a right to ϕ.22 In fact, since ϕ-ing will now be part of license, preventing the agent from ϕ-ing may be precisely what is called for. Accordingly, we can see that the problematic form of dependence, whereby the objectionable interference or obstacle is still in place but the agent is nevertheless classifed as free, cannot arise when dealing with a rights-based characterization of the z-parameter.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.  Freedom, Prison, And Poverty Although rights-dependence does not have objectionable implications analogous to those involved in desire- and ability-dependence, it might 21  Tat this is what is problematic about these cases can be brought out by considering cases involving contingent causal connections where an interference will be removed if the agent’s desires or abilities change. For instance, if someone is in chains, but the chains will be removed as soon as that person no longer has the desire to move around, then that agent can render himself free by changing desires without ending up with a problematic retreat to the inner citadel. (Pettit’s argument that liberty as noninterference is unstable and should be replaced by non-domination mistakenly treats the causal and the domainrestriction cases alike (cf. Pettit:  2011, p. 704). One needs to sharply distinguish the contingent removal of an obstacle from a situation in which an obstacle is still present but no longer classifes as a constraint.) 22  It is of course possible that there is something objectionable about the manner in which the right is changed and in which the objectionableness of the interference is removed. Yet once the right has been waived/alienated/forfeited, there is no longer anything objectionable about the relevant interference. Tat is, even though the process by means of which this situation was brought about may well be objectionable, the resulting situation does not involve an objectionable interference.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

151

nevertheless be deemed to be problematic on the grounds of generating a notion of liberty that does not cohere well with ordinary usage.23 Since the extent to which one is free is a function of the extent to which one is prevented from doing the things one has a right to do, it is possible to be completely free despite being subject to numerous interferences as long as one does not have a right not to be interfered with in these ways. Tis problem is nicely illustrated by the well-known prisoner objection. As Cohen has argued, the rights-based conception will not classify a prisoner who is justly imprisoned as being unfree. Given that the restrictions imposed on such a person are justifed and do not violate the prisoner’s rights, they do not count as infringements of liberty (cf. Cohen: 1995, p. 60). Tis, however, is highly counterintuitive, since being imprisoned seems to be a paradigmatic case of unfreedom. Tis unpalatable consequence is one of the main reasons that is usually cited when rejecting rights-based conceptions.24 Tis section will show that moralized accounts have the requisite resources to defuse this objection and provide a satisfactory explanation of the prisoner case.

3.1  Justifed Imprisonment

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

According to the prisoner objection, moralized conceptions have the counterintuitive commitment that they do not classify justifed imprisonment as a constraint on freedom. Whether this commitment does in fact follow depends on the reason for which imprisonment is deemed to be justifed. Tere are three options in this regard:

3.1.1 Unjustifed One can deny that imprisonment can be justifed. Te idea that imprisonment can be justifed is denied by a number of theories, such as restorative as well  as restitutive approaches (even a retributivist need not think that imprisonment is the correct form of punishment). In that case the prisoner objection does not get of the ground.

3.1.2  Justifed rights violation One can allow that imprisonment can be justifed, e.g. for deterrence reasons, but consider it to involve rights violations. If justifed imprisonment is 23  Bentham already complained that defenders of moralized conceptions “pervert language; they refuse to employ the word liberty in its common acceptation; they speak a tongue peculiar to themselves” (quoted in Waldron: 2006, p. 168). 24  Te other main reason, namely, that the moralized notion of liberty cannot play a fundamental justifcatory role, will be discussed in §4.2.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

152

Ralf M. Bader

considered to amount to a justifed rights violation, then the prisoner objection can be addressed straightforwardly. By allowing for justifed infringements of rights, one makes room for justifed unfreedom and hence can claim that the prisoner has a right to, say, move about freely but that this right can be justifably infringed, such that imprisonment is justifed whilst nevertheless violating rights and infringing liberty. A moralized account thus yields the result that the justly imprisoned prisoner is unfree to leave as long as imprisonment is classifed as a justifed rights violation.

3.1.3  Justifed without rights violations One can allow that imprisonment can be justifed and that it does not involve any rights violations. Tis can be due to various mechanisms.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. First, it can be due to rights forfeiture. Someone who is justly imprisoned has forfeited various rights and hence does not have a right to leave the prison. Preventing that person from leaving, consequently, does not involve any rights violations and does not infringe liberty. 2. Second, it can be due to the prisoner’s right being taken away by someone who has the Hohfeldian power to do so. Someone can be imprisoned without this involving any rights violations once a suitable authority has divested the person of the relevant rights. 3. Tird, one can bring in conditional rights, saying that the right to move about freely is conditional in nature, e.g. it is a right to move about freely unless justly imprisoned. Given that it is conditional upon not being justly imprisoned, this right is not infringed by justifed imprisonment. Tis is the most challenging scenario for moralized accounts (though not necessarily the most plausible understanding of punishment). Since no rights are violated, liberty is not being infringed. A justly imprisoned prisoner, accordingly, would not seem to be lacking in freedom. It is important to note that, although the prisoner is not considered to be unfree to leave, this does not imply the even more troubling consequence that the prisoner is deemed to be free to leave despite being prevented from doing so. Carter, for instance, suggests that the moralized account goes against our linguistic intuitions insofar as it will answer the following questions in the negative: “If I am physically prevented from crossing your justly acquired land, do I not lack the freedom to do so?”, and “Does the justly imprisoned thief in the high-security prison not lack the freedom to leave?” (Carter: 1999, p. 71). Tis, however, is incorrect. Tese questions can be answered in the afrmative, since the agents in question will not be  free to perform the relevant actions. Te moralized account does not consider the prisoner to be free to leave. Rather the prisoner is judged to be

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

153

free despite lacking the freedom to leave, because leaving is no longer something that is in the domain of the z-parameter and hence is not something with respect to which the agent can be unfree. A prisoner can, accordingly, be free (simpliciter) since it is possible that, despite being imprisoned, he is not prevented from doing anything he has a right to do. Te issue of contention regards, on the one hand, what makes it the case that these agents are not free to perform the actions in question. Te moralized account understands the lack of freedom in terms of the agent’s being not-free to leave and explains it on the basis that the person lacks the relevant rights, whereas the nonmoralized account understands the lack of freedom in terms of the agent’s being unfree to leave, which is explained in terms of the physical obstacles involved in being imprisoned. Tis diference implies that situations in which there is no physical prevention will be evaluated diferently. In such cases, the nonmoralized account will classify the agent who has forfeited the relevant rights as being free to perform the actions in question, whereas the moralized account will consider the agent to lack this freedom. On the other hand, the views difer as to whether this lack of freedom can be classifed as an infringement of liberty. According to the moralized conception, the prisoner will be free despite lacking the freedom to leave, whereas on the nonmoralized account lacking this freedom amounts to being unfree.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.2  Defusing the Prisoner Objection In order to address the prisoner objection, one needs to identify a salient and normatively signifcant sense in which the prisoner is less free than someone who is not imprisoned.25 Te problem now is that if justifed imprisonment is considered not to involve any rights violations, then imprisonment does not classify as an infringement of freedom. Te moralized view, however, can nevertheless account for the fact that a justly imprisoned prisoner is lacking freedom. Tis is because there are three ways of understanding the extent of liberty on the moralized view: 25   Mack has suggested that “individual I ’s liberty consists in the absence of those interferences that violate his rights or would violate his rights had those rights not been waived or forfeited by I. Te justly incarcerated I does sufer a loss of liberty even though no right of his is violated because he is subjected to interferences that would violate his rights had he not forfeited them” (Mack:  1995, p. 57 footnote 1). Tis suggestion is problematic, since it disconnects the loss of liberty from rights violations, thereby depriving freedom of its intrinsic signifcance. Given that the loss of liberty is construed counterfactually, insofar as it consists in the fact that the interferences in question would have violated rights had they not been forfeited, it becomes unclear in what way it is meant to be normatively signifcant.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

154

Ralf M. Bader

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. the extent of actions one has a right to do. 2. the extent of unprevented actions one has a right to do. 3. the extent to which one is being prevented from doing the things one has a right to do. Te frst construal is concerned with normative freedom. It is concerned with the size of the z-parameter, i.e. with all the actions one has a right to perform independently of whether or not they are infringed, and thus measures the range of permissible actions. Te second and third are concerned with physical freedom, i.e. with the actions that are in z and that are not prevented by anything in y. Whereas the second is an absolute notion that is concerned with the range of unprevented actions that one has a right to do, the last construal is a relative notion that is concerned with the degree to which one’s liberty is infringed, which is given by the proportion of unprevented actions to the total number of actions one has a right to perform, i.e. it is concerned with the extent of type-2 relative to type-1.26 Tis third notion is based on the core commitment of the moralized conception that one is free to ϕ if one has the right to ϕ and is not being prevented from ϕ-ing in a rights-violating manner, whereas one fails to be free to ϕ if one either lacks the right to ϕ or is prevented from ϕ-ing. Given that it is a relative notion, one can be entirely free independently of the number of rights one has (at least given that the number is greater than zero). Te extent to which an agent is free, according to this construal, is thus not to be understood in terms of the range of options available to the agent, but in terms of the extent to which the agent is interfered with, such that an agent is completely free if he is not subject to any interference, even if there might not be many options that are open to him. Tis distinction between the absolute notions and the relative notion allows us to understand the justly imprisoned prisoner as someone whose freedom has been reduced by having had certain freedoms taken away from him. Tis means that such a prisoner is someone who sufers a reduction in terms of type-1 and type-2 freedoms. Tough the range of things he is free to do is restricted, he is not being prevented from doing any of the things he has a right to do, which means that his freedom is not infringed. Accordingly his type-3 freedom is unafected.27 Te fact that imprisonment is not classifed 26  Although the degree of infringement is a relative notion that is defned in terms of the size of the domain of the z-parameter, the badness of infringement is not relative in this way. 27   Tis presupposes that none of the prisoner’s rights are violated, such that the degree of freedom = 1 both before and after he has had various freedoms taken away from him. Otherwise, there will also be a reduction in type-3 freedom, since a shrinking of the

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Moralizing Liberty

155

as a constraint on freedom thus only implies that it will not constitute an infringement of liberty. Tis, however, still allows for the prisoner to be lacking in freedom owing to a restriction of freedom. In short, we need to distinguish a restriction of freedom from an infringement of freedom. Te prisoner sufers the former but not the latter.28 Moralized conceptions thus have the resources to explain how someone who is justly imprisoned is lacking in freedom in terms of the fact that the range of actions that are permitted to a prisoner is vastly reduced. Since there are very few things that a prisoner has a right to do, there is not much that he is free to do. Tis means that, even when no one is infringing his freedom, he is free to do fewer things than other people are free to do, in the sense that the domain of the z-parameter contains far fewer courses of action for him than for non-prisoners. In short, prisoners are less free because they have fewer freedoms. Additionally, the distinction between type-1 and type-2 freedom allows us to account for the diferences between justly and unjustly imprisoned prisoners. Such people do not difer in terms of type-2 freedom, i.e. every uninfringed right had by the one is also had by the other. Tey do, however, difer in terms of the rights that they have and that are being infringed, which ensures that they difer in terms of type-1 and hence also in terms of type-3 freedom. Tere is thus both a sense in which they are equally free and a sense in which their freedom difers. An imprisoned and a non-imprisoned criminal, by contrast, will not difer in terms of liberty but only in terms of license, i.e. they do not difer in terms of what they are free to do, but in terms of which actions belonging to license they are prevented from doing.29 Tis means that the physical obstacles involved in imprisonment are inimical only to freedom in the case  of unjustly imprisoned prisoners. Te justly imprisoned prisoner z-parameter implies that the proportion of prevented actions in z increases if the number of infringements is held fxed. 28  As we will see in §4.1, both type-2 and type-3 are normatively signifcant notions of freedom, though they difer in important respects. In particular, whilst the signifcance of type-2 is to be understood axiologically, that of type-3 is to be construed deontically. 29   Tis only holds as long as they have the very same rights. Whether this is in fact the case depends on the particular mechanism that ensures that imprisonment does not involve rights violations. For instance, they have the same rights if the action that warrants imprisonment also ensures rights forfeiture. By contrast, if the relevant rights are taken away by someone with the relevant Hohfeldian power, then diferences can emerge. When concerned with conditional rights, there can be diferences in terms of detached rights, e.g. if the agent has the conditional right to move about freely unless justly imprisoned, since the condition will then only be satisfed in the case of the non-imprisoned criminal. Tis diference, however, disappears if the condition corresponds to that involved in rights forfeiture, e.g. if the agent has the conditional right to move about freely unless he has performed an action that warrants imprisonment.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

156

Ralf M. Bader

lacks freedom owing not to being imprisoned but to having had various freedoms taken away from him.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.3  Freedom and Poverty By diferentiating action-tokens from action-types, we are, moreover, able to distinguish the lack of freedom of a poor person from the lack of freedom of a prisoner. Whereas certain action-types are ruled out for the prisoner, a person lacking resources has the same rights regarding action-types as a  person having plenty of resources and difers only in terms of actiontokens.30 When comparing a poor person with a rich person, there is accordingly again a diference in freedom (this time merely a diference regarding action-tokens and not also action-types), without there being a diference in terms of infringements of freedom, i.e. x can be less free than y without being more unfree.31 Tis implies that a moralized conception allows one to make sense of the intuitive belief that both poverty and prison in some sense compromise liberty, even when the person in question is not unfree in the sense of having his liberty infringed. On the one hand, one can account for the fact that there is an important sense in which a person lacking resources is less free, namely insofar as there are fewer action-tokens such a person is free to  do. On the other hand, one can account for the fact that there is an important sense in which a prisoner is less free, namely insofar as there are fewer action-types (as well as action-tokens) such a person is free to do. Moreover, we can make sense of the idea that the restriction of freedom in the case of a prisoner is more drastic, given that entire classes of actions are ruled out, i.e. a prisoner is someone who has had certain types of freedoms taken away.32 30  As Cohen has argued, lacking money or resources is very likely to lead to interference if one should attempt to do those things that one could do if one had the requisite resources. (Te precise characterization of the connection between freedom and money is rather complicated, owing to the fact that money is neither necessary nor sufcient for avoiding being liable to interference but is only an inus condition, cf. Cohen: 2011, pp. 177–8.) 31   Te individuation and classifcation of action-types is a complicated matter. Te intuitive verdicts regarding prisoners and the poor are underwritten by an intuitive individuation of action-types. 32   Te concern that certain classes of actions are ruled out also seems to be underlying Waldron’s discussion of the troublesome nature of the lack of freedom resulting from homelessness, insofar as the particular prohibitions afecting the homeless efectively amount to a general prohibition, i.e. by preventing someone from performing all the tokens of a given type, one ends up preventing the action-type altogether (cf. Waldron: 1991, pp. 308–11). In this way homelessness would appear to be more akin to imprisonment than to poverty.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

157

In a free society, everyone is equally free in the sense that their liberty is not being infringed. No one is prevented from taking whatever legitimate courses of actions are available to them. Yet, they can difer in terms of the range of types as well as of tokens of legitimate courses of actions that are open to them, i.e. they can difer in terms of what they have a right to do and hence are free to do. Te poor and the rich are equally free in terms of action-types, yet they difer in terms of the number of action-tokens they are free to perform. Prisoners and non-prisoners, by contrast, difer in terms of both action-types and action-tokens. 4.  Intrinsic Normative Significance

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Once we distinguish the extent to which liberty is infringed, i.e. the extent to which there are interferences with legitimate courses of actions, from the number of actions that one is free to do, i.e. the number of legitimate courses of action that are open to an agent, the question arises as to which of these we should primarily be concerned with. Should we be concerned with ensuring that people are not interfered with, that their liberty is not infringed? Or is the goal to provide people with a wide range of options and maximize what they are free to do (weighted by the value of the courses of action)? In short, is a free society to be understood in terms of type-2 or type-3 freedom? In order to address this question, one needs to identify the normative signifcance of freedom.

4.1  Te Wrongness of Interference As soon as the freedom to torture, to rape, and to murder classify as genuine freedoms (as they do on a nonmoralized account), it is evident that considerations of freedom do not as such constitute pro tanto reasons.33 Tere will be some freedoms that are good and not to be infringed, as well as some that are bad and to be restricted, whilst yet others are neutral. Te diferences between these freedoms are not explicable in terms of the notion of freedom, but in terms of other values or principles. As a result, freedom loses intrinsic normative signifcance, i.e. it will not be the case that there is something that speaks in favor of freedom, as such, but only that there are various things that speak in favor of specifc freedoms. By contrast, the moralized account has intrinsic normative signifcance built into the concept, which guarantees that infringements of freedom are 33   Cf. “no sane person can expect that his primitive freedom merely as such should be protected” (Williams: 2001, p. 12).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

158

Ralf M. Bader

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

morally problematic. Te moralized account in this way underwrites the presumption of liberty, insofar as we have at least pro tanto reason to protect liberty and avoid infringements thereof.34 In particular, the presumption of liberty is not merely an epistemic default, in that we consider interference to be problematic unless we are provided with evidence to the contrary, but a normative commitment insofar as it amounts to recognizing a pro tanto reason against interference. Tis ensures that the moralized notion of liberty is suitable for playing a justifcatory role in political theorizing.35 It might be objected that the moralized notion is not particularly meaningful, insofar as being free is compatible, on the one hand, with having very few opportunities, and, on the other hand, with the opportunities being merely formal as opposed to real opportunities. Since negative freedom is an opportunity-concept, it is natural to consider the value of freedom to be a function of the value of the opportunities that one is free to pursue. Given that only real and not merely formal opportunities can be actualized, only they can be said to be valuable on this way of understanding the value of freedom. Insofar as, according to a moralized conception of liberty, being free as such does not guarantee any real but at most only formal opportunities, it would seem that this kind of freedom is not in itself signifcant and meaningful but merely formal. Accordingly, it is reasonable to ask why one should care about liberty understood in the moralized sense. Even though, given certain background understandings of rights,36 being free is compatible with hardly having any real opportunities, one can still 34   Cf. “If one regards coercion as a neutral concept, like ‘speaking’ or ‘walking’, it is something of a mystery why there should be a presumption against coercion, as many philosophers have insisted. Te link with individual rights shows how this presumption (like the presumption against theft) is, as it were, built into the concept itself ” (Ryan: 1980, pp. 494–5). 35   Intrinsic normative signifcance is compatible with the possibility of justifed infringements of liberty, since this amounts to allowing for the possibility that the signifcance of freedom can in certain cases be outweighed or overridden, thereby rendering interference justifed. Accordingly, one can still hold that all interferences with rightful conduct are pro tanto unjustifed in virtue of constituting rights violations (i.e. there is something that speaks against any such interference), but then accept that an interference may be all-things-considered justifed in certain conditions (i.e. what speaks in favor of the interference may be more weighty than what speaks against it). 36   Te precise characteristics of a particular rights-based conception depend on the particular rights that are part of the background moral theory. If people have a right to a certain level of resources, then being free will require having (at a minimum) opportunities corresponding to this level of resources, which would ensure that the account of freedom would not be merely formal. Te (minimal) extent of opportunities bound up with being free accordingly depends on the particular account of rights that is adopted when providing a particular conception of freedom belonging to the family of moralized conceptions. Even in the case of an entitlement theory, the proviso on acquisition, together with its historical shadow, ensures that the opportunities cannot be

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Moralizing Liberty

159

consider this kind of liberty to be meaningful and normatively signifcant. In particular, this merely ‘formal’ liberty has intrinsic signifcance that is not reducible to the value of the actions that one is free to do. Infringements of liberty are not only morally problematic because they preclude the value of  the actions being prevented, but also because they are intrinsically objectionable. Preventing someone from doing something that person has a right to do is objectionable on the grounds that it does not respect that person’s moral status.37 Liberty, understood in this sense, is not something that is to be maximized or promoted, but something that is to be respected. Respecting the dignity and moral status of persons requires not preventing them from doing things they have a right to do, even if they lack the ability or resources to do those things. Given that rights are deontic notions, the best way to think of the normative signifcance of moralized freedom is not axiologically in terms of the goodness of the resulting opportunities, but deontically in terms of the wrongness of interference. Te reason why being free to ϕ matters is not because one thereby has the opportunity to ϕ, but rather because one’s right to ϕ is being respected and because one is not illegitimately being prevented from ϕ-ing. Any illegitimate interference with what one has a right to do is intrinsically unjust and disrespectful of one’s moral status. By contrast, the opportunity to ϕ is at best derivatively good, and moreover its being good presupposes that the opportunity is a real as opposed to a merely formal opportunity. Illegitimate interference is thus intrinsically objectionable, and not merely bad in virtue of precluding valuable opportunities. Tis implies that this kind of interference is wrong independently of whether one is concerned with the liberty to do something signifcant or something trivial, independently of whether or not the person is able to perform the action in question or lacks the relevant ability and would thus only have had a formal opportunity that could not have been realized had the interference not taken place, as well as independently of whether the person cares about the action and would have done it had it not been for the interference. Te wrongness is thus not a function of the desires and abilities of the agent nor of the value of the action. Tis kind of wrongness cannot be explained in terms of preventing the agent from realizing the relevant opportunity entirely formal. (Tis should be sufcient to defuse Waldron’s objection based on the case of homelessness (cf. Waldron: 1991).) 37  As Cohen notes with respect to legal permissibility, “it is an insult to the status of persons when certain acts are [legally] forbidden to them, whether or not the [legal] permissibility of those acts would generate a corresponding freedom [= real opportunity] for those persons” (Cohen: 2011, p. 191).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

160

Ralf M. Bader

but instead has to be accounted for in terms of the interference being intrinsically wrong. Instead of saying that it is good to be free to ϕ, one should thus say that it is wrong to be rendered unfree to ϕ. Focusing on the wrongness of interference, accordingly, involves a twofold departure from the traditional approach that is concerned with the value of freedom. First, the signifcance of freedom is understood not axiologically but deontically. Second, unfreedom is deemed to be objectionable and freedom is only an object of comparative pro-attitudes insofar as it consists in the absence of something wrong, which contrasts with the traditional approach that considers freedom to involve the presence of something good and treats unfreedom to be only comparatively bad insofar as it consists in the absence of a good thing. Tis diference ensures that more freedom is not preferable as such on the moralized account, since this does not involve more of a good thing. Accordingly, freedom is not to be promoted but to be respected, i.e. what is preferable is less unfreedom rather than more freedom. Tus, the core issue when it comes to the signifcance of freedom per se is whether freedom is infringed, not whether the number of action-types or action-tokens is greater or lesser. What primarily matters is thus that one is more free in the sense that one is being less interfered with, not in the sense of being free to do more things, i.e. having a greater range of options available. It is the former sense of freedom that has intrinsic signifcance. Tat is, the problem lies in the frst place with the interference itself, and not with the resulting lack of opportunities. Considerations of the extent of action-types and action-tokens available to the agent are important issues, but they are not issues that pertain to the intrinsic signifcance of freedom, and are not pertinent when it comes to respecting and protecting liberty. Instead, other considerations relating to the value of the relevant opportunities need to be appealed to when it comes to explaining their signifcance. A lack of opportunities only matters derivatively to the extent that the opportunities are signifcant, such that the importance of promoting the extent of the courses of action that the agent is free to do derives from the extrinsic value of being free to perform these actions.38 In other words, when concerned with the intrinsic signifcance of freedom one is not concerned with the range of options that are open to the agent. Te intrinsic signifcance of freedom is to be understood deontically in terms of the wrongness of interference, rather than axiologically in terms of the value of the things one is free to do. Accordingly, the primary sense 38  Tis means that the way in which the justly imprisoned prisoner is lacking in freedom is not intrinsically objectionable but is problematic (i.e. is bad for that person and hence is classifed as a form of punishment) for extrinsic reasons.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

161

in which one is to care about the extent of freedom is the sense concerned with the extent to which an agent is prevented from doing the things he has a right to do. It is this sense of freedom that matters intrinsically.39

4.2  Liberty as Nonfundamental

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Although liberty has intrinsic normative signifcance according to the moralized account, there is a concern that this notion cannot play any fundamental justifcatory role if liberty is defned in terms of rights. In particular, it will not be possible to appeal to the notion of liberty in justifying an account of what rights we have. Waldron, for instance, claims that the moralized notion “excludes the concept of freedom altogether from the debate about the justifcation of property rights” (Waldron: 1991, p. 308). Tis is because rights will be more fundamental, ensuring that issues regarding rights need to be settled frst before freedom can be addressed. Accordingly, liberty will not be fundamental and will not belong to the set of basic normative notions.40 Tis problem can be illustrated by the well-known fact that Nozick’s claim that liberty upsets patterns is dialectically inefective. Nozick tries to argue for a certain account of property rights on the basis of considerations regarding freedom, even though he holds a moralized conception of liberty. Tis kind of argument, however, cannot succeed.41 If one uses a moralized notion, then it already needs to be settled what rights people have, if we are to know which interferences infringe their liberty. Put diferently, claims 39  As mentioned in §2.1, liberty can be compromised not only if it is infringed but also if it is illegitimately restricted as a result of an immunity violation. Te latter also has intrinsic normative signifcance, yet can be set aside when concerned with the freedom of prisoners and the poor. 40   Liberty will be downstream, not only in terms of justifcation but also epistemically. Tis implies that we might be radically mistaken about how free people are, as well as about which interferences infringe their liberty. Tis is because we need to know what is legitimate and what is illegitimate, what people have a right to do and what people do not have a right to do. Otherwise, we will not know how free they are and in what ways their liberty is being infringed. Cf. Bentham: “According to this defnition, I can never know whether I have the liberty to do an action, until I have examined all its consequences” (quoted in Waldron: 2006, p. 168). Since there can be a mismatch between the objective facts about morality and our subjective conception thereof, it may turn out that the interferences that really do infringe liberty do not match up with what we ordinarily take to be infringements of liberty. 41   Tis was already pointed out in an early review of Anarchy, State, and Utopia: “It follows, then, that Nozick cannot appeal to the value of liberty in support of the idea of justice as rights of entitlement, since the former logically presupposes the latter . . . It deprives him of a logically independent concept of freedom in terms of which his theory of justice might be vindicated” (Smith: 1976, p. 89). Also cf. Ryan: 1977; Cohen: 1995, pp. 60–2; Waldron: 1991, p. 308.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

162

Ralf M. Bader

about infringements of liberty already presuppose that which is to be established, namely an account of property rights, thereby rendering them question-begging and dialectically inefective. Although the moralized notion requires one to treat facts about liberty as  being derivative, which excludes the notion of liberty from playing a fundamental justifcatory role, it has the advantage of having intrinsic normative signifcance. As a result of having (derivative) intrinsic signifcance it can feature in normative explanations and justifcations, though not in ultimate explanations. By contrast, the nonmoralized account lacks intrinsic normative signifcance altogether. Liberty, as such, will not have any value or signifcance. A nonmoralized account will only end up with a “fat description that carries, in itself, no suggestion of endorsement or complaint” (Dworkin:  2000, p. 125). It is only particular liberties that can be (dis)valuable, deriving their signifcance from the value of the specifc opportunities that are available to  the agent.42 As a result of lacking intrinsic normative signifcance, the nonmoralized notion cannot play any justifcatory role, not even a derivative justifcatory role, in particular it cannot be appealed to in establishing substantive conclusions about rights. Tis means that anyone wanting to assign a central role in political theorizing to the notion of liberty faces a dilemma. Either one adopts a  moralized account, which ensures that liberty has intrinsic normative signifcance but this signifcance is not fundamental—in this case, liberty cannot play a fundamental justifcatory role, since all justifcation will ultimately proceed via the more basic notions in terms of which the moralized notion is defned; or one accepts a nonmoralized account, in which case liberty lacks intrinsic normative signifcance, which means that it is unable to play a substantive justifcatory role and that, similarly, all justifcation will ultimately proceed via the normative notions that account for the extrinsic value that nonmoralized liberties can have. Either way, the notion of liberty is unable to perform fundamental justifcatory work.43 42   Carter has attempted to argue that freedom has nonspecifc value (cf. Carter: 1999). However, his arguments at best establish that freedom has extrinsic value (primarily instrumental value) but not intrinsic value, which is brought out by the fact that the supposed nonspecifc value is contingent and dependent on the particular empirical circumstances. 43   Tis type of dilemma, whereby a notion either is too thin and lacks intrinsic normative signifcance or is too thick and ends up being only derivatively signifcant, arises in the case of other concepts as well. For instance, the concept of harm can either be moralized by means of a normative baseline, or it can be construed in a nonmoralized manner by means of an actual, counterfactual, or statistical baseline. Neither account will classify the notion of harm as a basic normative notion, which means that this concept will not play any role in fundamental theorizing. (On the upside, the widespread phenomenon of

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

163

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4.3  A Consistent System of Liberty Conficts of liberty and hence infringements of liberty will inevitably arise on the nonmoralized approach. Tis implies that a nonmoralized account does not allow for a consistent system of liberty and does not validate the idea that liberty should not be in confict with itself. It will in principle not be possible to arrive at a situation in which everyone is free. At most, one can end up with everyone being equally (un)free.44 However, complete freedom (whereby everyone is free to ϕ for all ϕ in z) will be impossible. By contrast, the moralized account can avoid conficts of liberty as long as rights are compossible. In this way, it allows for a consistent system of liberty and makes room for a situation in which everyone is (completely) free. Tis suggests that the claim that the moralized concept of freedom is excluded altogether from the justifcation of rights is actually too strong. Whilst considerations of freedom will not have a foundational role, they can nonetheless play a limited justifcatory role. In particular, the accounts of freedom and of rights can be mutually reinforcing. One can support a particular conception of rights on the basis that it allows for a consistent system of liberty. Since a consistent system of liberty requires a consistent system of rights, we have an additional reason for accepting a compossibility requirement on the theory of rights, and hence adopting a theory that recognizes only negative but not positive rights. More generally, structural requirements that need to be satisfed by a theory of freedom, such as a compossibility requirement, need to be satisfed by the underlying structure of rights. Whereas particular claims about what people are free or not free to do depend on what rights they have, structural claims about liberty are not dependent in this way and can accordingly function as constraints on the theory of rights. Moreover, whilst rights-based accounts presuppose rights that are logically prior and independently specifed, an alternative Kantian moralization in terms of a universalizability requirement simultaneously generates both a moralized notion of freedom and a specifcation of rights. For instance, Waldron has suggested that “we might attach normative signifcance in Kantian fashion only to assignments of liberty that are compatible with the moralization might be taken to suggest that, rather than there being a heap of independent normative notions, there is hope for a unifed approach based on a supreme principle underlying the various moralizations. Te diferent moralized concepts would then represent diferent facets of this unifed underlying structure.) 44  When clashes of freedom arise, one will have to bring in substantive moral commitments that are separate from considerations of freedom to determine which side is to be privileged. Freedom by itself will not settle the confict.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

164

Ralf M. Bader

assignment of equal liberty (even equal mundane liberty) to everyone else, and denigrate as license claims to freedom that are not universalizable in this way” (Waldron: 2006, p. 170). Freedom, on this account, will be moralized and one will be free to do only what one has a right to do, yet freedom will not be derivative from rights but on a par with rights.45 Te Kantian approach starts with the nonmoralized notion of freedom, i.e. nonmoralized liberty is the input that is subjected to the universalization constraint to result in a partition into universalizable moralized liberty and nonuniversalizable license. Here one might wonder why we are universalizing freedom as opposed to something else. In particular, it might be thought that nonmoralized freedom has to matter in order to explain why moralized freedom (= universalizable nonmoralized freedom) matters. Yet, the reason we are interested in nonmoralized freedom is not because it has normative signifcance but because it constitutes the relevant domain. Tis is because external actions are the objects of juridical evaluation and thus function as the external counterparts of maxims. In the same way that one can care about universalizability of maxims without having to think that maxims as  such are of value or the like, one can care about the universalizability of  nonmoralized freedom without having to think that this notion of freedom matters. Tis means that, although one starts with nonmoralized freedom, what is doing all the normative work and gives rise to the normative signifcance of the resulting moralized notion is universalizability.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

5. Conclusion Tus, we have seen that moralized conceptions of liberty have signifcant theoretical advantages and that the objections that have been raised against them can be defused. In particular, we have to be operating with a moralized notion when we say that freedom matters as such. Te moralization ensures that liberty is intrinsically normatively signifcant, whereas nonmoralized conceptions are unable to attribute intrinsic value to freedom but are able to locate value only in the actions that people are free to do rather than in freedom itself. Moreover, a rights-based defnition allows us to provide a fne-grained account that distinguishes the extent of infringement of liberty from the extent of action-types and action-tokens one is free to do. Tis enables us to capture the sense in which a justly imprisoned prisoner is lacking freedom, insofar as there are fewer action-types such a person is free 45  Whether this can be made to work depends on whether the universalizability requirement is sufcient for symmetry-breaking and can exclude not only asymmetrical distributions but also rule out problematic symmetrical distributions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Moralizing Liberty

165

to do, as well as the sense in which poverty is inimical to freedom, insofar as there are fewer action-tokens a poor person is free to do, while retaining the idea that, as long as they are not subject to rights-violating interferences, none of their liberties are being infringed. Maybe political philosophy (as well as political discourse) would be served best if the concept of freedom was abandoned in favor of more fundamental concepts that have nonderivative signifcance. But it is clear that if this concept is not relinquished, then it is to be retained in the form of a moralized conception.46

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Bader, R. M. “Moralised conceptions of liberty.” In Oxford Handbook of Freedom, D.  Schmidtz, Ed. Oxford University Press, 2016. Online at http://www. oxfordhandbooks.com/view/10.1093/oxfordhb/9780199989423.001.0001/ oxfordhb-9780199989423-e-4 Bader, R. M. “Liberty, threats, and ineligibility.” Manuscript. Berlin, I. Two Concepts of Liberty. Oxford University Press, 1958. Carter, I. A Measure of Freedom. Oxford University Press, 1999. Cohen, G. A. Self-Ownership, Freedom and Equality. Cambridge University Press, 1995. Cohen, G. A. On the Currency of Egalitarian Justice, and Other Essays in Political Philosophy. Princeton University Press, 2011. Dworkin, R. Sovereign Virtue—Te Teory and Practice of Equality. Harvard University Press, 2000. Kramer, M. Te Quality of Freedom. Oxford University Press, 2003. Kristjánsson, K. Social Freedom—Te Responsibility View. Cambridge University Press, 1996. Locke, J. Second Treatise of Government. Awnsham Churchill, 1689. MacCallum, G. “Negative and positive freedom.” Te Philosophical Review 76, 3 (1967), 312–34. Mack, E. “Moral individualism and libertarian theory.” In Liberty for the 21st Century—Contemporary Libertarian Tought, T. Machan and D. Rasmussen, Eds. Rowman & Littlefeld Publishers, 1995, pp. 41–58. Miller, D. “Constraints on freedom.” Ethics 94, 1 (1983), 66–86. Miller, D. “Reply to Oppenheim.” Ethics 95, 2 (1985), 310–14. Nozick, R. Anarchy, State, and Utopia. Basic Books, 1974. Nozick, R. Philosophical Explanations. Harvard University Press, 1981.

46   For helpful comments, I would like to thank audiences in Tucson, in Prague, and at the OSPP conference in Barcelona, as well as Joseph Carlsmith and Dan Waxman. Special thanks to Ian Carter for detailed comments on an earlier draft of this chapter. Tanks also to two anonymous referees.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

166

Ralf M. Bader

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Olsaretti, S. “Freedom, force and choice: against the rights-based defnition of voluntariness.” Te Journal of Political Philosophy 6, 1 (1998), 53–76. Pettit, P. “Te instability of freedom as noninterference: the case of Isaiah Berlin.” Ethics 121, 4 (2011), 693–716. Ryan, C. C. “Yours, mine, and ours: property rights and individual liberty.” Ethics 87, 2 (1977), 126–41. Ryan, C. C. “Te normative concept of coercion.” Mind 89, 356 (1980), 481–98. Shnayderman, R. “Social freedom, moral responsibility, actions and omissions.” Philosophical Quarterly 63, 253 (2013), 716–39. Smith, G. “Review of ‘Anarchy, State, and Utopia.’ ” Philosophical Books 17, 2 (1976), 87–90. Steiner, H. “Te structure of a set of compossible rights.” Te Journal of Philosophy 74, 12 (1977), 767–75. Waldron, J. “Homelessness and the issue of freedom.” UCLA Law Review 39 (1991), 295–324. Waldron, J. “Mr. Morgan’s Yacht.” In Te Egalitarian Conscience—Essays in Honour of G. A. Cohen, C. Sypnowich, Ed. Oxford University Press, 2006, pp. 154–76. Williams, B. “From freedom to liberty: the construction of a political value.” Philosophy and Public Afairs 30, 1 (2001), 3–26.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

7

May a Government Mandate More Comprehensive Health Insurance than Citizens Want for Temselves? Alex Voorhoeve

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

What do you say to Mark and Lucinda in my district who had a plan, they liked it, it was afordable, but it is being terminated [because it is not sufciently comprehensive to meet the requirements of the Afordable Care Act]? It was what they wanted and I will remind you: some people like to drive a Ford, not a Ferrari. Some people like to drink out of a Red Solo cup, not out of crystal. You are taking away their choice. Representative Marsha Blackburn (Republican) to U.S. Secretary of Health Kathleen Sebelius, Hearing on the Afordable Care Act, 30 October 2013

1. Introduction In most developed and many developing countries, governments require citizens to be insured for a minimum set of health-care interventions. Tey typically do so through some combination of the following: (i) the provision of tax-fnanced care (such as, for example, Medicare provides for the elderly in the U.S.); and (ii) requiring the purchase of at least a basic package of insurance from a social fund or a private provider (as the U.S. Afordable Care Act does for the non-poor and non-elderly) (Carrin and James 2005; Cotlear et al. 2015). For simplicity, in what follows, I shall refer to both tax-fnanced care and required purchase as “mandated insurance.” In this chapter, I shall address two questions about such a mandate. First: what, if anything, justifes the limitation it places on the freedom to spend one’s fair share of resources as one sees ft? Consider, for example, the typical

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

168

Alex Voorhoeve

citizens invoked by Representative Blackburn. Suppose they have average incomes and modest current health risks, are prudent and adequately informed. Why shouldn’t they be able to spend the money they are required by the Afordable Care Act to spend on health insurance in whatever way they prefer, for example by purchasing cheaper, less comprehensive insurance and using the remaining funds for other purposes? Te second question is this: if mandatory insurance is justifed, how should governments determine the content of the mandated minimum package? Various answers to these questions have been proposed, in part because mandated coverage serves a variety of ends (see Daniels 2013 for a review). One such end is protection against the consequences of imprudent risk-taking (Dworkin 2000, p. 492 n6; Bou-Habib 2006; Moncrief 2013). Another is to fairly and efciently cover the costs of the care that we are obligated to provide on humanitarian grounds, such as basic care for those in urgent need who could not aford to pay for such care out of pocket (Buchanan 1984; Bou-Habib 2006; Segall 2010, ch. 5; Menzel 2012; Rulli and Wendler 2016). In this chapter, I shall not focus on these reasons for requiring coverage, important though they may be. Instead, I shall critically analyze the following well-known pair of liberal egalitarian answers for the case of prudent, adequately informed adults (Dworkin 2000, chap. 8, 2012a,b; Braun 2012; Menzel 2012):1

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

(i) To keep premiums afordable for those at unfairly high risk, those at low risk must be forced to insure themselves on relatively equal terms. Fairness towards those who, owing to bad brute luck, have comparatively large anticipated health-care needs requires that they can insure themselves on terms that are similar to the terms faced by those with smaller anticipated needs. But insurance on such equal terms will be most attractive to those with the greatest anticipated health-care needs and unattractive to those with the smallest anticipated needs. If such insurance was voluntary, therefore, a disproportionate share of enrollees would be those with relatively high health-care costs. Coverage for this group would be expensive, and there would be little of the desired sharing of risks and costs between the lucky and the unlucky. Tis outcome can be most efciently prevented by forcing those with low expected health-care costs to insure themselves on terms that are similar to the terms for those with high expected costs.

(ii) Te mandated minimum package of services should be determined by the Representative Prudent Individual Test (or RPIT, for short). A government should mandate the insurance package that a typical prudent and adequately informed individual would wish to purchase for themselves if they were placed in fair conditions of choice. 1

  Gibbard (1984) makes a similar argument on utilitarian grounds.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

169

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

In this chapter, I will argue that answer (i) is incomplete. A further reason to mandate insurance is that this contributes to important social egalitarian public goods. I will also argue that answer (ii) is insufciently egalitarian. My argument proceeds as follows. In §2, I ofer a fuller explanation of  the common liberal egalitarian justifcations for a mandate and the RPIT. In § 3, I consider circumstances in which individuals face diferential risks. I argue that under such circumstances, mandating the insurance that a representative prudent individual would purchase wrongly treats interpersonal trade-ofs as if they were intrapersonal trade-ofs and thereby ignores unfair inequalities. In § 4, I consider circumstances in which all individuals are at  equal risk and have equal purchasing power and explain why, under such circumstances, the common liberal egalitarian account holds that there is no reason for a mandate. In § 5, I argue that this account fails to consider the social egalitarian public goods to which a well-designed mandatory health insurance package contributes. Tese public goods are underprovided when individuals are free to make their insurance decisions independently, and this, I argue, gives us a reason to mandate insurance even when people are equally situated. I conclude that governments should design a mandatory package by appealing to a pluralistic egalitarian view, which cares about improving people’s well-being, reducing unfair inequality, and maintaining egalitarian social relationships. 2. The Argument For Mandating The Package Desired By A Representative Prudent Individual Leading thinkers have argued that a mandated minimum of insurance is justifed as the best solution to two problems that would occur in a free market for health insurance (Gibbard 1984; Dworkin 1994, 2000, ch. 8, 2012a,b; Braun 2012; Menzel 2012; WHO 2014). Te frst is that people with low expected expenditures on needed health care in a given period would be charged relatively little for insurance in that period, while those with high expected expenditures would be charged much more. Assume, as is plausible, both that those with higher expected expenditures tend to be those with worse health and that a large proportion of diferential health is not the result of free, adequately informed choice under fair circumstances but is rather due to bad brute luck. It follows that a free market in health insurance would add fnancial burdens to the unfair disadvantage of poor health. Such compounding of burdens would be especially problematic because those with worse health are already more likely to have lower incomes and lower social status (Deaton 2013; Sreenivasan 2014).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

170

Alex Voorhoeve

One standard way of avoiding such unfair premium diferentiation is to require that insurers take anyone who applies at a premium that is relatively independent of the applicant’s personal risk profle. (Tis is known as “community rating.”) However, such community rating contributes to a second problem. Suppose that frms were to set the uniform premium at a  price that would cover the average person’s health expenditure. At this premium, insurance would be least attractive to people who believe themselves to have lower-than-average expected expenditures on covered health services and most attractive to people with higher-than-average expected expenditures. Since many people are aware of their (approximate) risk profle, those who would enroll would tend to have disproportionately high health expenditures, a phenomenon known as “adverse selection” (Arrow 1963). To cover the costs of this disproportionately sick, enrolled population, frms would therefore have to raise the premium above the initial rate. But doing so would, at the margin, tend to cause those with the lowest anticipated health costs to drop out of the pool, since they would then regard insurance as too expensive relative to their personal risk profle. Tis would further worsen the risk profle of the enrolled population, meaning that the premium would have to be raised again, leading to a so-called “death spiral” of an escalating premium and an insurance pool consisting of people with ever-greater health risks. Without intervention, this adverse selection mechanism may severely impair the functioning of the insurance market (Akerlof 1970; for a real-world example, see Barro 2010). In theory, these problems might be addressed by pairing community rating with extensive subsidies for voluntary enrollment. However, the experience of countries at all levels of development suggests that this is not a promising route (Carrin and James 2005; Lagomarsino et al. 2012; Cotlear et al.  2015). By contrast, together, community rating and mandatory enrollment achieve both fair cost-sharing between those at low and high health risk and a solution to the adverse selection problem. Requiring enrollment of those who would rather not now enroll because they are currently at low risk is not merely a case of forced transfers to others who are less well-of. Presumably, many people who are currently at low risk want health insurance available to them at reasonable cost when their risks become high (such as when they develop a need for expensive care). Owing to the aforementioned death spiral, insurance might not then be available to them at afordable cost if enrollment was voluntary. Many may therefore prefer, on their own behalf, that everyone is required to be insured when they are at low risk in order to guarantee that there is a well-functioning insurance market with reasonably priced coverage when their expected health-care costs become high. Mandatory payments may be justifed to these people as an efcient way to get something they would be unable to

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

171

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

get if everyone was free to make insurance decisions on their own (Braun 2012; Menzel 2012). Naturally, to avoid unfairly burdening the poor, contributions (through taxes and/or individual premiums net of subsidies) should be incomedependent. A mandate of this kind redistributes resources from rich to poor and from the healthy to the infrm. Like many solidaristic arrangements, it has a mixed justifcation. In part, it appeals to an egalitarian principle of justice that cares about reducing unfair inequality. In part, it appeals to each person’s self-interest in establishing an institution that serves this interest better than uncoordinated individual choices would. I now turn to the question of how one ought to draw the boundaries of the package of mandated care. To many, the following line of reasoning has seemed compelling (Gibbard 1984; Dworkin 1994, sec. 2; 2000, ch. 8; Menzel 2012, p. 596 n28; Kurtulmus 2012). Suppose that each member of our society were prudentially rational, had a fair share of resources, and had the average health risks of people in our society. Suppose further that they had to choose a lifetime health insurance policy with full pertinent information about the prospective benefts and costs to them of each feasible package of health services. In such circumstances, there would be no need to require community rating, since everyone’s health risks (and their beliefs about these risks) would be identical at the moment the insurance contract was signed. For the same reason, there would be no need to prevent adverse selection. Moreover, there would be no other reasons of justice to interfere with individuals’ choices regarding health insurance—the resulting pattern of insurance would be fully just. We should use this idealization of our society as a reference point for mandatory insurance in our actual, unjust circumstances. In the idealized circumstances, individuals with diferent preferences would purchase diferent insurance plans. We cannot replicate this diversity of individually-tailored insurance plans in our society for many reasons, including the adverse selection problem outlined. But we should approximate it by determining the content of the required insurance package as follows. Te Representative Prudent Individual Test (RPIT): We should mandate the lifetime package of insurance that a representative prudent individual with a fair share of resources would want for themselves if: (a) their self-regarding preferences and risk attitudes were those of a typical person in our society (once corrected for irrationalities); (b) they believed that they faced the average lifetime health risks in our society; (c) they knew the potential costs and benefts to them of the package; and (d) their risks, costs, and benefts were calculated under the assumption that the package selected would become the required minimum for all.2 2   While it difers from their individual proposals in some details, this formulation of and motivation for the RPIT picks out the relevant common elements in proposals by Gibbard, Dworkin, Menzel, and Kurtulmus. Te formulation employed avoids some of the problems with Dworkin’s (1994, 2000) version of the RPIT highlighted in Macleod (1998, pp. 92–6) and Kurtulmus (2012).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

172

Alex Voorhoeve

In the remainder of this chapter, I shall raise objections to this view and propose revisions to it.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

3.  Objections To The RPIT I shall begin by arguing that the RPIT needs to be discarded because it is insufciently egalitarian. Te following Diferential Prospects Case highlights its problems. Suppose that a three-person society of Unhealthy, Healthy, and Avy is about to implement a new mandatory package. Unhealthy has recently been diagnosed with a serious health condition and her lifetime need for health care is certain to be great. Healthy, by contrast, is certain to possess excellent lifetime health and have little need for health care.3 Avy has average prospects: she is in good health now, but will end up in either the same situation as Unhealthy or the same situation as Healthy, with each possibility being equally likely. Te three feasible policies are: Small (a low-cost package with limited coverage), Medium (a  moderately costly package of middling coverage), and Large (a more costly, but also more comprehensive package). So that we have a concrete measure of prudential interests in terms of which we can assess these packages, I shall assume that prudential interest is identical to well-being and is measured as follows, in line with orthodox decision theory.4 A frst alternative has higher expected well-being for a person than a second alternative just in case the frst would be strictly preferred for this person’s sake after rational, calm deliberation with all pertinent information while considering their self-interest only. Two alternatives yield equal expected well-being just in case such deliberation would yield indiference between them.5 Te assumed impacts of these three insurance packages on each person’s well-being are represented in Table 1. 3   Of course, in reality, no adult enjoys such certainty, but we grant Healthy knowledge of her unusual fortune for the sake of a simplifying idealization of the situation of individuals who are known to have better than average health prospects. 4   In the Appendix, I argue that one ought to reject RPIT even if one pairs it with a well-known rival decision theory. 5  I follow Gibbard (1984) in assuming this measure, which draws on idealized preferences which respect the von Neumann–Morgenstern axioms. Tis measure does not presuppose any particular view on what well-being consists in. One might maintain both that well-being consists in something other than preference satisfaction and that the specifed idealized preferences fully track the magnitude of this other thing (Otsuka and Voorhoeve  2009, pp. 172–3 n3). I also assume this measure is fully interpersonally comparable. Te assumed measure can be seen as an idealization of common healthrelated well-being measures based on the so-called “Standard Gamble” (Bognar and Hirose 2014, ch. 2). 

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

173

Table 1  Lifetime well-being for three insurance packages in the Diferential Prospects Case Individual Policy Small Medium Large

Unhealthy

Avy

Healthy

35 40 44

59.5 60 59.5

84 80 75

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Note: Unhealthy’s and Healthy’s well-being is certain under each policy; the well-being listed for Avy is her expected well-being.

In this case, by construction, Avy’s prospects are equally good under Small and Large. Te RPIT is therefore indiferent between them. Moreover, her prospects are most advanced by the choice of Medium. Since Avy is the representative individual, the RPIT therefore selects this package. Te central objection to this test is that it treats what is a trade-of between the interests of two separate people, Unhealthy and Healthy, in precisely the same manner as it treats a trade-of between two potential futures of a single individual, Avy. In doing so, it fails in two ways to recognize reasons for favoring the worse of that arise in interpersonal trade-ofs (Gauthier 1963, pp. 121–7; Nagel 1970, pp. 132–42; Rawls 1999, secs. 5 and 39; Roemer 2002). First, the RPIT is insensitive to the fact that some will be better of than others. When one evaluates a policy solely on the basis of Avy’s expected well-being, one pays attention to how she will fare under that policy in each of her potential futures, but one does not consider it as in itself bad that in one potential future she would be better of than she would be in another potential future. Small and Large are equally good prospects for Avy because they yield the same expected well-being, even though the gap between her worst and best potential outcome under Small is greater than the gap between her worst and best potential outcome under Large. Te RPIT takes the same attitude to the fates of coexisting, separate people as it takes to the equally likely potential futures of a single person. It is therefore indiferent to the inequality between two coexisting, separate people (Unhealthy and Healthy), just as it is indiferent to the inequality between two mutually exclusive possible futures of the same person (Avy). For the RPIT, Small and Large are equally good policies, even though both policies generate the  same expected total well-being and the former contains more interpersonal inequality than the latter. Tis lack of concern for how some fare compared to others is not merely substantively mistaken; it is also inconsistent with the aversion to unfair inequality that motivates the common liberal egalitarian account.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

174

Alex Voorhoeve

Second, the RPIT fails to recognize that when diferent people’s interests confict, the strength of a person’s claim to a beneft of a given size depends in part on how badly of this person is compared to others.6 By way of illustration, consider again Small and Large. As just noted, the RPIT holds that these two policies are equally good, so that, if one faced a choice between these two alone, it would be permissible to choose either one. Unhealthy could challenge this judgment with the following rhetorical question (Otsuka and Voorhoeve 2009, pp. 183–4): How could you choose to forgo a very substantial beneft to a worse of person to instead provide an equally large beneft to someone else who will, in any case, be much better of?

Next, consider the choice between Small, Medium, and Large. Te following complaint on Unhealthy’s behalf against a choice of Medium rather than Large would be almost as forceful, and, I submit, equally unanswerable:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

How could you choose to forgo a very substantial beneft to a worse of person to instead provide only: (i) a somewhat larger beneft to a diferent, much better of person and (ii) a comparatively small improvement in the expectations of another person, who already has far superior prospects?

No such questions arise when we compare these policies solely in terms of their impact on Avy’s prospects—one could not complain, on behalf of the unlucky possible future of Avy, that by choosing Medium (rather than Large) one would be failing to beneft them in order to provide a somewhat larger beneft to a diferent person (Otsuka and Voorhoeve 2009, p. 184). Tese problems are avoided if we abandon the RPIT and determine the content of the mandatory package with reference to a pluralist egalitarian view, which cares about reducing unfair inequality in both prospects and outcomes and about improving total well-being.7 On such a view, Large will  obviously be regarded as better than Small, since Large contains less inequality than, but the same total expected well-being as, Small. Properly specifed, this pluralist egalitarian view will also rank Large over Medium, because the former very substantially reduces inequality at only a minor cost in terms of total expected well-being. Moreover, because this view is averse to inequality, it recognizes that Unhealthy’s claim to Large is strengthened by the fact that she is less well-of than others and that Healthy’s claim 6   While this second objection is related to the frst, it is, as Otsuka (2012) shows, logically distinct. See Lange (2017) for further discussion of the way in which being worse of than others afects the strength of a person’s claim. 7   See Voorhoeve and Fleurbaey (2016) and Otsuka and Voorhoeve (2017) for defenses of a pluralist egalitarian view. See Ottersen et al. (2016) and Voorhoeve et al. (2016, 2017) for the use of a view of this kind to make concrete health policy decisions.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Government Mandate Comprehensive Health Insurance . . . ?

175

against Large is weakened by the fact that she is better of than others. Finally, using this form of pluralist egalitarianism to design the mandatory package also makes for a consistent rationale, because it means that the very same values that motivate the mandate—reducing unfair disadvantage and improving the general welfare—also determine what is mandated. It is worth emphasizing that while, in the context of this simple example, such pluralist egalitarianism will recommend Large because it improves the situation of the worst of at an acceptable cost, it will not turn health care into a “bottomless pit.” One reason is that, plausibly specifed, this pluralist view will rule that a small improvement in the situation of the worst of is not worth a very large diminution of others’ prospects. To illustrate, suppose that in our simple case, the government could also mandate an Extra Large insurance package, which would improve Unhealthy’s lot  (and, if she were unlucky, Avy’s as well) to a well-being level of 45, but at the cost of leaving Healthy (and, if she were lucky, Avy) with a well-being level of 65. A reasonable pluralist egalitarian view will, I submit, prefer Large to Extra Large, because the reduction in total expected wellbeing that the latter would involve would be too great compared to the improvement in the prospects of the least well-of. Moreover, the real world is more complex than our example, in which more comprehensive coverage always leads to greater transfers from the healthier and wealthier to the sicker and poorer. Tis example is useful, because it demonstrates the RPIT’s failings and because such transfers are an important part of well-designed, real-world mandatory insurance packages. And insofar as, in the real world, an expansion of coverage leads to more such transfers, the proposed egalitarian criterion will indeed require more comprehensive coverage than the RPIT. But in reality, there  may also be interventions that would modestly improve average well-being but also disproportionately beneft the better of and impose a substantial burden on the worse of, either in fnancial terms or in terms  of other improvements to their lives that could be made with the resources in question. While the RPIT would recommend including these interventions, the pluralist egalitarian view would not. Te pluralist egalitarian package will therefore likely include both more coverage for some conditions (e.g. those that primarily afict the worse of) and less coverage for others (e.g. those that primarily afect the better of). More generally, since it requires that fnite resources be spent in the manner that best advances the aims of improving the general welfare and reducing inequality, the view proposed here will motivate cost-control measures in health care, such as those that reduce the frequency of very expensive interventions with low expected beneft (Menzel 2012, pp. 591–4). It will also direct us not to spend money on health when we could do more

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

176

Alex Voorhoeve

to improve people’s lives and reduce inequality by spending it on education, environmental quality, personal consumption, and so on. In sum, the proposed pluralist brute luck egalitarian view can avoid the objections I have raised against the RPIT while taking account of the need to limit expenditure on health. As we shall see in §§4 and 5, however, the outlined view is incomplete, since it needs to be supplemented by a principle that captures distinctive social egalitarian concerns.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

4. The Case Of Identical Risks In this section and §5, I shall argue that the grounds for a mandate extend beyond the need to solve the problems created by unfair diferential risks and adverse selection. I shall therefore focus on a scenario in which these problems are absent. Consider the following Identical Prospects Case. Suppose a large society consists only of adult Avys, each of whom, as shown in §4, faces a 50–50 gamble between a serious illness and a long, healthy life. Also suppose that it is known that precisely half of them will end up in either position. As in our earlier discussion, everyone is prudent and informed of their risks. Te government can either leave everyone free to choose between the aforementioned three packages, or instead require everyone to purchase Large. Table 2 represents, for each insurance package, the prospects of each person and the distribution of fnal well-being in the population that would result if everyone had this package of insurance. Given our assumptions, if each person had a free choice between these insurance packages and acted independently on their own behalf, each person would purchase Medium, because this is what maximizes their expected well-being. Tis will lead to substantially more outcome inequality Table 2  Prospects and outcomes for three packages in the Identical Prospects Case Package

Each person’s expected well-being

Distribution of fnal well-being if everyone had the package

Small

59.5

Half at 35 Half at 84

Medium

60

Half at 40 Half at 80

Large

59.5

Half at 44 Half at 75

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

177

than would arise if everyone were instead required to purchase Large. Tis inequality notwithstanding, in a case such as this, the account I have been criticizing holds that we have no reason to mandate insurance. In Ronald Dworkin’s words:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

whatever this . . . community actually spends on health care [as the result of people’s free choices from a position of equality] is the morally appropriate amount for it to spend: it could not be criticized, on grounds of justice, for spending either too much or too little.  (2000, pp. 312–13)

Tere appear to be several good reasons for Dworkin’s conclusion. First, leaving people the choice between these three packages respects their freedom to use an equal share of resources as they see ft. By contrast, mandating Large in order to lessen outcome inequality would appear to infringe this freedom. Second, given our assumptions, each will use this freedom to make what is, given the risks they face, the prudentially best decision. Concern for the well-being of each person, considered in isolation, should therefore motivate us to endorse each person’s choice of Medium; it also counts against mandating Large. Tird, it appears that brute luck egalitarianism does not give us grounds to object to the diference in outcome inequality between Medium and Large. Brute luck egalitarianism objects to unchosen inequalities because they are unfair. It also holds that inequalities that are due to competent, free, and informed choice against an equal background need not be unfair (see, e.g., Cohen 1989; Arneson  1989; Temkin  2001). In the Identical Prospects Case, the increase in outcome inequality between Medium and Large would be entirely due to such choices. Tis increase in inequality would, of course, be due to luck, but this would not be brute luck, but rather what Dworkin (2000, p. 73) defnes as “option luck”: “a matter of how deliberate and calculated gambles turn out—whether someone gains or loses through accepting an isolated risk he or she should have anticipated and might have declined.”8 In other words, both Medium and Large would contain the same amount of objectionable brute luck inequality. Still, it does not follow that a brute luck egalitarian must accept the additional outcome inequality that results from people’s choices in this case. 8   Tere are well-known difculties in adequately spelling out the distinction between brute luck and option luck (Vallentyne  2002, pp. 531–4). Tere are also powerful arguments that even inequalities due to diferential option luck are objectionable because people cannot reasonably be held substantively responsible for them (Lippert-Rasmussen 2001). I shall not engage with these challenges, since I argue in §4 that even an egalitarian who is generally willing to accept inequalities due to diferential option luck should object to such inequalities in this case.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

178

Alex Voorhoeve

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

By itself, an aversion to brute luck inequality does not tell us what kinds of  choice-based inequalities one should permit among people who face equal opportunities. Brute luck egalitarians must therefore rely on distinct principles to decide which choice-based inequalities to countenance (Olsaretti 2009; Stemplowska 2009; Segall 2010, ch. 4). One such principle, advanced by Dworkin, is that, in the absence of market failures, the government should allow equally situated, informed, and rational people to freely contract and let them bear the consequences of these contracts (Dworkin 2000, ch. 2). On the assumption that, with adverse selection and its attendant problems ruled out, there are no market failures in the Identical Prospects Case, it follows that people should be left free to choose and that the inequality generated by everyone’s choice of Medium should be accepted. Another such principle, advanced by Peter Vallentyne, focuses on the value of people’s opportunity sets. Vallentyne (2002, pp. 551–2) proposes that when, under conditions of equal opportunity, we must decide whether individuals should face a set of choices that will generate inequality, we should ask whether it would be prudentially valuable for each person to face these choices. Since we are assuming that individuals choose what has higher prudential value for them, this principle also favors leaving everyone a free choice in the Identical Prospects Case. We have seen that respect for freedom and concern for people’s wellbeing appear to favor leaving people a choice in this case, while the resultant increase in inequality (compared to mandating Large) can be put down to acceptable diferential option luck. Despite its appeal, I shall argue in §5 that this line of reasoning is mistaken. 5.  Social Egalitarian Public Goods As an initial step towards this critique, we should note that an individual in the Identical Prospects Case might, while preferring Medium for themselves on prudential grounds, also object to the inequality that results from everyone choosing Medium. Indeed, they might, when considering both their personal prospects and the overall degree of inequality in society, prefer a society in which everyone has Large over a society in which everyone has Medium.9 Here, I shall explore the reasons for such a preference that derive from a social egalitarian ideal. Tis ideal is fully realized when all adult members 9   My argument that a focus on only the individual impacts of a transaction omits consideration of its distributional and relational impacts has parallels to Cohen’s (1977) critique of Nozick’s (1974) “Wilt Chamberlain argument.”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Government Mandate Comprehensive Health Insurance . . . ?

179

of society have the ability to live as free citizens10 who can participate in social life and contribute to public decision-making on equal terms. More specifcally, this view involves the attainment of four interrelated goods, each of which is partly constitutive of a society of equals (Anderson 1999; Rawls 2001, sec. 39; O’Neill 2008; Satz 2010; Fourie et al. 2015). Te frst is the absence of domination and exploitation in private life. Health insurance contributes to this good because it improves people’s ability to access needed care without falling into destitution. It thereby prevents the forms of subordination and unfair advantage-taking that occur when the less well-of become highly dependent for needed care or fnancial assistance on more powerful others. Te second is the absence of domination in political life. To realize this good, society must prevent situations in which those who are less well-of do not have their interests efectively represented. Health insurance helps do so when, through transfers from the healthy and wealthy to the unhealthy and less well-of, it blunts two of the forces that lead people to become socially and politically marginalized, to wit, ill health and fnancial disadvantage (Mattila et al. 2013; Smets and van Ham 2013). Te third category encompasses what Rawls (1999) called the social bases of self-respect. Among these is the ability to participate in public life with dignity—to be able to join in work, play, and general social interaction without shame or a sense of inferiority based on characteristics irrelevant to the status of citizen, including low income or ill health. Te care to which health insurance gives access can ensure that fewer people have conditions that generate a debilitating sense of shame or inferiority. Moreover, insofar as it redistributes from the well-to-do and healthy to the poor and ill, it limits the type of inequalities that give rise to norms of appearance and functioning that the less well-of struggle to meet. Te care to which it gives access can also bolster other sources of self-worth, including the ability to carry out everyday tasks, as well as to support oneself and one’s family. Te fnal social egalitarian good consists in the attitudes required for social cooperation, which include a willingness to work with others to mutual beneft, to enter into reasoned discussion with them about fair social arrangements, and to abide by fair arrangements if others do so too. Large inequalities in health and income threaten these attitudes in various ways. For one, they can generate a sense of inferiority among the less fortunate and arrogance among the fortunate (Tawney  1964, pp. 37–8). Te less well-of may also become resentful of others’ advantages, leading 10   Since the view focuses on citizens, a question arises of what it requires with respect to the treatment of resident noncitizens. For simplicity, I set aside this issue.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

180

Alex Voorhoeve

them to prefer a situation in which the lives of the better of are diminished, even if this does not improve their own lot. In response, the better of may be inclined to jealously guard their relative advantage at a cost to themselves and others (Rawls 1999, pp. 467–8). One further possible efect is described by Adam Smith. Te better of tend to avoid interacting with those who are very badly of in order to prevent the painful sympathetic feelings that would be a natural consequence of being in their company. For their part, an uncomfortable awareness of others’ desire to turn away from them causes the badly of to withdraw from society (Smith 1982, I.iii.1; see also Chase and Bantebya-Kyomuhendo  2015). Insofar as it improves the fate of the least fortunate and lessens the gap between them and the rest of society, more extensive health insurance plays a part in preventing the spread of such problematic attitudes. Te ways in which health insurance contributes to these goods can provide grounds for a citizen to prefer a society in which everyone has Large over one in which everyone has Medium. But they may also make one wonder whether it is coherent to assume that Medium is prudentially superior. After all, each individual’s self-interested prospects should take into account the impact on their well-being of these social efects of an  insurance package. Tese efects might indeed be substantial. Being dominated, politically marginalized, having one’s self-respect threatened, feeling envious, or withdrawing from social interaction out of shame all depress one’s well-being. A prudent chooser will, therefore, consider the ways in which purchasing Medium (rather than Large) raises the risk that these evils will befall them. Moreover, insofar as we consider the impact of everyone choosing Medium, we should note that an increase in social inequality could depress well-being by lessening social cohesion. Nonetheless, social egalitarian goods (or the lack thereof ) are but one component of wellbeing. It is therefore perfectly possible that from the perspective of each person’s self-interested prospects, these potential downsides of Medium are outweighed by its possible upside, namely, that one is much better of if one  is healthy. Furthermore, the importance of social equality cannot be reduced to its prudential value. Independently of one’s own well-being, one has reason to pursue a society in which people can relate to one another as equal citizens, because such a society involves relationships and attitudes which are in themselves good, and because it limits the evils of domination, indignity, lack of self-respect, and unsociable attitudes. Te assumption that Large is prudentially costly but superior in social egalitarian terms allows us to keep the possible trade-of between prudence and social equality clearly in view. Let us now consider the signifcance of these social goods and evils for the question of mandating health insurance. Suppose that, considering both

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

181

their prospects and the degree of social equality, all citizens prefer a society in which everyone has Large over one in which everyone has Medium. Wouldn’t it follow that if given the independent choice between Small, Medium, and Large, each would voluntarily purchase Large? No. For these social egalitarian goods possess two characteristics of public goods.11 Te frst is nonrivalry: one person’s enjoyment of an egalitarian environment does not impair others’ access to it. Te second is substantial nonexcludability: if, for example, everyone else were to purchase Large and thereby help secure a society of equals, then whether or not one personally purchased such insurance, one would partake in many of the relevant social goods (e.g. one would live in a society in which domination and exploitation are less common, and in which people are more likely to have the attitudes required for social cooperation). Together, nonrivalry and nonexcludability create a familiar problem encountered in generating the good of an egalitarian environment through independent action. To illustrate this point, consider the simplifed representation in Table  3 of the situation of a citizen who must choose between insurance packages independently. (It is simplifed by leaving out Small, which is inferior on both prudential and egalitarian grounds, and by considering only two patterns of choice exerted by all other citizens: everyone else chooses Medium and everyone else chooses Large.) Each cell Table 3  Relevant outcomes for a citizen choosing between Medium and Large for themselves alone

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

All other citizens One citizen

Medium

Large

Medium

Expected well-being = 60 Low degree of social equality

Expected well-being ≥ 60 Degree of social equality is fractionally below ‘high’

Large

Expected well-being ≤ 59.5 Degree of social equality is fractionally above ‘low’

Expected well-being = 59.5 High degree of social equality

11   It may be useful to compare my proposal with the view that mandatory basic health insurance contributes to the public good of living in a society that meets its humanitarian obligations towards its members (Buchanan 1984; Segall 2010, pp. 78–80; Menzel 2012). It is, plausibly, a precondition for a society of equals that, out of appreciation of the value of each person’s life, society is committed to meeting citizens’ basic needs when this can be done at reasonable cost (cf. Rawls 1999, p. 298). However, such a commitment is not sufcient for social equality. Te social egalitarian public goods that I discuss here are therefore more encompassing than this humanitarian public good.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

182

Alex Voorhoeve

describes the two things this citizen cares about: their expected well-being and the degree of social equality. By assumption, this citizen prefers the southeast cell (everyone has Large) to the northwest cell (everyone has Medium). But they cannot choose between these two cells, because they control only their own choice of Medium or Large. To choose rationally, this citizen will therefore consider their best reply to what others might do. Now, if all others were to choose Medium, then it is consistent with our assumptions about this citizen’s values that they would also choose Medium. For if they were to choose Large, they would only fractionally improve social equality, but they would sufer a signifcant loss in expected wellbeing, not merely because of the cost of the insurance package, but also because they would, in any case, sufer the negative efects of the inegalitarian environment created by others’ choices. (Te latter explains why their expected well-being if they alone choose Large is at best equal to their expected wellbeing if everyone were to choose Large.) If, instead, all others were to choose Large, then it is again consistent with our assumptions that this citizen would choose Medium. For their choice of Medium would only fractionally lower the degree of social equality. But it would signifcantly improve their personal prospects. For they would not have to pay the cost of Large and would beneft from the positive efects on their well-being of living in the more egalitarian society created by others’ choice of Large. (Te latter are why their expected well-being if they alone choose Medium is at least as great as their expected well-being if everyone were to choose Medium.) One might, at this point, object that a citizen who adequately values social equality would be willing to do their bit by purchasing Large if they were confdent that enough others would do so too. But we should note that even such a conditional cooperator has every reason to choose Medium if they believe that many others are likely to do so as well. In sum, even citizens who care enough about social equality to give up some prospective well-being for its sake may fnd themselves facing a collective action problem, because each may rationally purchase Medium when they face the decision on their own, even though they would prefer a  situation in which everyone has Large to one in which everyone has Medium.12 12  Depending on citizens’ attitudes, this collective action problem may therefore take one of two forms. One is a multi-person, one-shot Prisoner’s Dilemma. Tis occurs if, in the situation outlined in Table 3, each citizen rationally chooses Medium, no matter what others do. But citizens may also be conditional cooperators. In Table 3, such a citizen will prefer Medium if he believes others will choose Medium, and Large if he believes others will choose Large. If all citizens are conditional cooperators, then their interaction has the structure of an Assurance Game. Te purpose of a mandate is diferent in each case. If the

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

183

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

As with collective action problems generally, one solution is to restrict choice, and mandate Large. Especially in a numerous society, such a mandate may be the most efcient solution and may therefore be favored by each member of the population over less restrictive alternatives, such as legally permitting choice but establishing strong social norms in favor of purchasing Large. It is noteworthy that when the mandate is endorsed by all, we do not have a confict of values, with on one side, the reasons for permitting choice enumerated in §3 and, on the other, social egalitarian concerns. For each is then, by assumption, happy to sufer a small loss in their expected well-being and a limitation of their freedom for the sake of acting collectively to improve the degree of social equality. Moreover, under these circumstances, the two arguments presented in §3 for the acceptability of the inequality generated by free choice are not compelling. Recall that the frst of these arguments, derived from Dworkin (2000, chs. 2 and 8), relied on the principle that, in the absence of market failures, equally situated individuals should be left free to contract, along with the assumption that there are indeed no such market failures in the Identical Prospects Case. We have seen, however, that the latter assumption is false, because the market undersupplies social egalitarian public goods. Dworkin’s principle therefore does not imply that we should countenance the inequalities that would result from unconstrained choice.13 Te second argument relied on the principle that inequalities are acceptable when people face equal and problem is a Prisoner’s Dilemma, then it is to force cooperation on a social optimum; if it is an Assurance Game, then it is to generate coordination on a social optimum by providing assurance that others will contribute. I am grateful to Joshua Cohen for pressing me to clarify this point. 13  Indeed, Dworkin (2000, pp. 155–8) endorses a “principle of correction” that requires that the state prevent such market failures. It follows that, if Dworkin had accepted my argument that the social egalitarian goods produced by health insurance are public goods, then he would have had to withdraw his claim (quoted at the start of §4) that a society of Avys who were all left free to choose Medium “could not be criticized on grounds of justice.” I should note that Dworkin (2002, p. 114) appeals to his principle of correction as a justifcation for mandatory “personal accident” insurance on the grounds that an individual’s decision not to purchase such insurance imposes costs on “the rest of the community, including employers and dependants.” Dworkin does not clarify the costs he has in mind, but perhaps he was thinking of the emergency medical care that hospitals are, out of humanity, obligated to provide in many countries (e.g., under the Emergency Medical Treatment and Labor Act in the U.S.), as well as public welfare support for the person’s dependants. If so, then this argument of Dworkin’s for mandatory personal accident insurance would appeal to the requirement to fairly share the costs of meeting humanitarian obligations. As mentioned in n. 11, meeting these obligations is but one part of establishing social equality. Dworkin therefore seems to overlook the contribution of health insurance to the full panoply of social egalitarian public goods that I discuss here.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

184

Alex Voorhoeve

prudentially valuable opportunities (Vallentyne 2002). But this principle is questionable in this context, because each citizen will deny that a choice between Small, Medium, and Large is, all things considered, the most valuable one they could face. Considering both prudential and social egalitarian value, they prefer not to face this choice (so long as this choice is removed from everyone). And there does not seem to be a strong case for permitting inequalities that result from unwanted choices. We can conclude that, when all agree to a mandate, there is decisive reason for mandating Large. However, the case for a mandate is less straightforward when some (but not all) citizens would object to it, because they judge that the gain in social equality is not worth the loss of liberty and the cost to their expected well-being. Tese dissenters need not be purely self-interested; they might merely make a diferent judgment about the balance of values involved. In these circumstances, we face a confict between, on one side, the liberty and prudential interests of those who oppose a mandate and, on the other side, egalitarian public goods that cannot be generated through unconstrained individual choice. Here, I shall not add to extant discussions about how democratic societies can, in the face of such conficts of value and opinion, determine a right level of public goods provision (see Claassen 2013 for a review). Instead, I merely note that the arguments made in this section explain why those who favor more comprehensive insurance because of its contribution to social equality cannot be expected by dissenters to simply advance this end through voluntary, private action. Tat an egalitarian environment is  a public good gives us a reason, albeit a defeasible one, to secure it through a mandate, when this is the most efcient way of overcoming the collective action problems that arise in producing and protecting such an environment. Te ways in which health insurance can further social equality are relevant not merely to the justifcation of a mandate, but also to its content, including in circumstances in which citizens have diferential prospects. For these contributions give us reason to judge this package by three criteria: improving total expected well-being, reducing brute luck inequality, and securing social equality. Under some circumstances, the two egalitarian criteria will align. Insofar as the benefts of an increase in the size of the mandated package come to the least well-of and its costs are principally borne by the better of, both brute luck egalitarianism and social egalitarianism will tend to favor this expansion. We should note, however, that the two egalitarian views can also diverge. For social egalitarianism assigns particular importance to the impact of health on an individual’s ability to perform the role of a citizen. Te conditions that threaten this ability, such as disfgurement or

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

185

severe impairments in mobility, communication, or mental health, may therefore be given greater weight than their well-being impact alone would warrant (Hausman  2015, ch. 13). Social egalitarianism can therefore be expected to prompt a shift in the mandatory package of insurance towards interventions that secure capacities that are central to people’s acting as free and equal citizens.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

6. Conclusion I have put forward a revised liberal egalitarian account of mandatory health insurance. On this view, a mandate can be justifed because of its role in overcoming two problems arising with voluntary insurance. Te frst is adverse selection, which occurs in a particularly problematic form when we require, on grounds of equity, that those with unfairly elevated health risks pay similar premiums to those at low risk. Te second is underprovision of the public good of social equality. I have also argued that the content of the mandatory package should not be what a representative prudent individual would purchase for themselves. Instead, it should be determined by an appeal to the following values: improving people’s well-being, reducing brute luck inequality, and securing social equality. Compared to a package that a representative prudent individual would want, these values will typically prompt us to include more coverage for interventions that help the worse of, as well as more coverage for interventions that prevent the ills of domination, marginalization, and loss of self-respect. What, then, should we say to Mark and Lucinda, sensible people of average means and moderate current health risk, who are compelled to purchase a more comprehensive and therefore more costly package than they desire? First, that in order to ofer fairly priced insurance to everyone, including people who, through bad luck alone, face higher health risks than they do now, we have to require everyone to purchase this more comprehensive package. While this will raise their current premiums, it will also lower their premiums should they eventually face high health risks. By protecting others, they are therefore also protecting themselves. Second, we should ask them to consider that, because of its universal adoption, the package will shape the society they live in. Additional coverage for conditions that leave people especially vulnerable and isolated will result in less corrosive inequality and better prospects for people to participate in public life with dignity. In short, we should say that a mandate is necessary

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

186

Alex Voorhoeve

because fair access to health insurance and a less divided society are things that we can produce only together.14

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Appendix Why The Rpit Fails Even When Paired With An Unorthodox Decision Theory Te argument in §2 assumes a measure of well-being that relies on orthodox decision theory. A familiar objection to this theory (and my related assumption that prudence requires maximizing expected well-being) is that when one is evaluating a prospect from a prudential point of view, one should be permitted to be especially averse to worse outcomes. A well-known alternative to the orthodoxy, known as “rank-dependent” decision theory, indeed allows one to assign a lower-ranked outcome more weight than its probability alone would warrant (Quiggin  1993; Buchak 2013). It is noteworthy that the fundamental objections to the RPIT would remain even if one were to employ this rank-dependent decision theory. To see why, I shall consider a pair of contrasting cases. Let us start with what I will call the Equal Health Case. In this scenario, society consists entirely of people who are all “in the same boat,” in the sense that they will either all end up Ill or all end up Well and are equally likely to end up in either situation. Te government can choose between policy A and policy B. Policy A does relatively little for those who are sick; it therefore leaves Ill much worse of than Well. Policy B, by contrast, ensures complete equality between Ill and Well. However, it comes at a cost c in expected well-being. Table A1 describes this case. Note that the columns describe states of the world. Assume that 0 < c < 20, so that, compared to A, B always improves people’s situations if they fall ill. Now, on standard decision theory, A is prudentially best for each person, because it maximizes each person’s expected well-being. However, on rank-dependent decision theory, it is permissible to give some extra weight to the worst possible situation, so that for some c, B is prudentially better for each person. Let cEqual be the 14   Tis chapter was presented at Duke University, Georgetown University, the Hebrew University of Jerusalem, the Oxford Studies in Political Philosophy Conference at Universitat Pompeu Fabra, Stanford University, University College London, the University of California at Berkeley, the University of Virginia, and the U.S. National Institutes of Health. I am grateful to those present on those occasions and to Matthew Adler, Daniel Baker, Paul Bou-Habib, Luc Bovens, Joshua Cohen, Donald Franklin, Dana Howard, Niko Kolodny, Kasper Lippert-Rasmussen, Colin Macleod, Joseph Mazor, Véronique Munoz-Dardé, Michael Otsuka, Tom Parr, Debra Satz, Gopal Sreenivasan, Peter Vallentyne, David Wendler, Andrew Williams, and an anonymous referee for comments. Work on this chapter was carried out while the author was Visiting Scholar at the Department of Bioethics at the U.S. National Institutes of Health. Te opinions expressed are the view of the author. Tey do not represent any position or policy of the U.S. National Institutes of Health, the Public Health Service, or the Department of Health and Human Services.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

187

Table A1  Well-being outcomes in the Equal Health Case State of the world

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Policy

Everyone Ill (probability = 0.5)

Everyone Well (probability = 0.5)

A

40

80

B

60—c

60—c

largest cost for which one could, on prudential grounds alone, choose B in the Equal Health Case. Te rank-dependent RPIT will therefore regard A as permissible for all cases in which B is at least as costly as this, that is, for all c ≥ cEqual. Indeed, it will require A for all cases in which B is more costly than this, that is, for all c > cEqual. Next, consider the Unequal Health Case, in which there is no risk. Half the population is Ill, the other half is Well. Te government must decide between policy A, which leaves the Ill much less well-of than the Well, and policy B, which generates complete equality, again at cost c, with 0 < c < 20. Tis case is represented in Table A2. Note that the columns are now segments of the population. It should be obvious that there are important diferences between the Equal and Unequal Health Cases. First, if one chooses A in the Equal Health Case, there is no inequality. By contrast, if one chooses A in the Unequal Health Case, there will be substantial unfair inequality in both prospects and outcomes. Second, the Equal Health Case involves only a trade-of between two diferent possible futures of each individual. By contrast, the Unequal Health Case involves a trade-of between the interests of separate people. I submit that these diferences make it more important to improve the worst possible situation in the Unequal Health Case than in the Equal Health Case. It follows that there is some c for which it is permissible to choose A in the Equal Health Case but impermissible to choose A in the Unequal Health Case, because one ought to do more to improve the situation of the ill in the latter. However, the rank-dependent RPIT cannot recognize this. For in the Unequal Health Case, the RPIT will tell the government to choose the package that it would be prudent to purchase on one’s own behalf if one had average risks, that is, were equally likely to end up Ill or Well. It follows that the RPIT would draw the same conclusion in the Unequal Health Case as in the Equal Health Case: regard A as Table A2  Well-being outcomes in the Unequal Health Case People Policy

Ill (half of the population)

Well (half of the population)

A

40

80

B

60—c

60—c

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

188

Alex Voorhoeve

permissible for all c ≥ cEqual and as required for all c > cEqual. Te RPIT is, in this sense, wholly insensitive to the presence or absence of unfair inequality and to whether the trade-ofs in question are between two potential futures of the same individual or instead between the interests of separate individuals. Of course, neither of these cases represents a realistic scenario. But each of these cases represents, in an isolated form, one aspect that is present in real-world cases of universal, mandatory purchase of an insurance package. Te RPIT’s failure to respond appropriately to the diferences between these cases therefore generates a problem for this principle in real-world scenarios that consist of a mixture of the elements that each case models in isolation. Te RPIT, whether paired with orthodox or rank-dependent decision theory, will recommend exactly the same policy in all scenarios, no matter whether they are closer to the ideal type represented by the Equal Health Case or closer to the ideal type represented by the Unequal Health Case. It is therefore also unresponsive to inequality and to people’s competing claims under realistic circumstances.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Akerlof, George (1970). “Te Market for Lemons: Quality, Uncertainty and the Market Mechanism,” Te Quarterly Journal of Economics 84: 488–500. Anderson, Elizabeth (1999). “What Is the Point of Equality?” Ethics 109: 287–337. Arneson, Richard (1989). “Equality and Equal Opportunity for Welfare,” Philosophical Studies 56: 77–93. Arrow, Kenneth (1963). “Uncertainty and the Welfare Economics of Medical Care,” Te American Economic Review 53: 941–73. Barro, Josh (2010). “A ‘Scaled-Back’ Health Bill Won’t Work.” Real Clear Markets Blog January 26. http://www.realclearmarkets.com/articles/2010/01/26/a_scaled_ back_health_bill_wont_work_97609.html. Online. [Accessed January 2017]. Bognar, Greg and Iwao Hirose (2014). Te Ethics of Health Care Rationing. (London: Routledge.) Bou-Habib, Paul (2006). “Compulsory Insurance without Paternalism,” Utilitas 18: 243–63. Braun, S. Stewart (2012). “Mandatory Health Insurance, Liberalism, and Freedom,” Public Afairs Quarterly 26: 179–97. Buchak, Lara (2013). Risk and Rationality. (Oxford: Oxford University Press.) Buchanan, Allen (1984). “Te Right to a Decent Minimum of Health Care,” Philosophy & Public Afairs 13: 55–78. Carrin, Guy and Chris James (2005). Social Health Insurance: Key Factors Afecting the Transition towards Universal Coverage. (Geneva: World Health Organization.) Chase, Elaine and Grace Bantebya-Kyomuhendo, eds. (2015). Poverty and Shame: Global Experiences. (New York: Oxford University Press.) Claassen, Rutger (2013). “Public Goods, Mutual Benefts, and Majority Rule,” Journal of Social Philosophy 44: 270–90.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Government Mandate Comprehensive Health Insurance . . . ?

189

Cohen, G. A. (1977). “Robert Nozick and Wilt Chamberlain: How Patterns Preserve Liberty,” Erkenntnis 11: 5–23. Cohen, G. A. (1989). “On the Currency of Egalitarian Justice,” Ethics 99: 906–44. Cotlear, Daniel, Somil Nagpal, Owen Smith, Ajay Tandon, and Rafael Cortez (2015). Going Universal: How 24 Developing Countries Are Implementing Universal Health Coverage Reforms from the Bottom Up. (Washington, DC: Te World Bank.) Daniels, Norman (2013). “Justice and Access to Health Care,” Stanford Encyclopedia of Philosophy (Spring 2013 Edition), Edward N. Zalta (ed.), http://plato.stanford. edu/entries/justice-healthcareaccess/Online. [Accessed August 2016]. Deaton, Angus (2013). “What Does Empirical Evidence Tell Us about the Injustice of Health Inequalities?” in Nir Eyal, Samia Hurst, Ole Frithjof Norheim, and Daniel Wikler (eds.) Inequalities in Health. (Oxford: Oxford University Press): 263–81. Dworkin, Ronald (1994). “Will Clinton’s Plan Be Fair?” New York Review of Books January 13, http://www.nybooks.com/articles/1994/01/13/will-clintons-plan-befair/Online. [Accessed May 2016]. Dworkin, Ronald (2000). Sovereign Virtue: Te Teory and Practice of Equality. (Cambridge, MA: Harvard University Press.) Dworkin, Roland (2002). “Sovereign Virtue Revisited,” Ethics 113: 106–43. Dworkin, Ronald (2012a). “Why the Health Care Challenge Is Wrong,” New York Review of Books, April 2, http://www.nybooks.com/daily/2012/04/02/why-healthcare-challenge-is-wrong/Online. [Accessed May 2016 ]. Dworkin, Ronald (2012b). “A Bigger Victory than We Knew,” New York Review of Books, August 16, http://www.nybooks.com/articles/2012/08/16/bigger-victorywe-knew/Online. [Accessed January 2017]. Fourie, Carina, Fabian Schuppert, and Ivo Wallimann-Helmer (eds.) (2015). Social Equality: On What It Means to Be Equals. (New York: Oxford University Press.) Gauthier, David (1963). Practical Reasoning. (Oxford: Clarendon Press.) Gibbard, Allan (1984). “Health Care and the Prospective Pareto Principle,” Ethics 94: 261–82. Hausman, Daniel (2015). Valuing Health: Well-Being, Freedom and Sufering. (Oxford: Oxford University Press.) Kurtulmus, Ahmet Faik (2012). “Dworkin’s Prudent Insurance Ideal: Two Revisions,” Journal of Medical Ethics 38: 243–6. Lagomarsino, Gina, Alice Garabrant, Atikah Adyas, Richard Muga, and Nathaniel Otoo (2012). “Moving towards Universal Health Coverage: Health Insurance Reforms in Nine Developing Countries in Africa and Asia,” Te Lancet 380 (9845): 933–43. Lange, Benjamin (2017). “Restricted Prioritarianism or Competing Claims?” Utilitas 29: 137–52. Lippert-Rasmussen, Kasper (2001). “Egalitarianism, Option Luck, and Responsibility,” Ethics 111: 548–79. Macleod, Colin (1998). Liberalism, Justice and Markets: A Critique of Liberal Equality. (Oxford: Oxford University Press.)

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

190

Alex Voorhoeve

Mattila, Mikko, Peter Söderlund, Hanna Wass, and Lauri Rapeli (2013). “Healthy Voting: Te Efect of Self-reported Health on Turnout in 30 Countries,” Electoral Studies 32: 886–91. Menzel, Paul T. (2012). “Justice and Fairness: A Critical Element in US Health System Reform,” Te Journal of Law, Medicine & Ethics 40: 582–97. Moncrief, Abigail (2013). “Te Individual Mandate as Healthcare Regulation: What the Obama Administration Should Have said in NFIB v. Sebelius,” American Journal of Law and Medicine 39: 539–72. Nagel, Tomas (1970). Te Possibility of Altruism. (Princeton, NJ: Princeton University Press.) Olsaretti, Serena (2009). “Responsibility and the Consequences of Choice,” Proceedings of the Aristotelian Society CIX, Part 2: 165–88. O’Neill, Martin (2008). “What Should Egalitarians Believe?” Philosophy & Public Afairs 36: 119–56. Otsuka, Michael (2012). “Prioritarianism and the Separateness of Persons,” Utilitas 24: 365–80. Otsuka, Michael and Alex Voorhoeve (2009). “Why It Matters that Some Are Worse Of than Others,” Philosophy and Public Afairs 37: 171–99. Otsuka, Michael and Alex Voorhoeve (2017). “Equality versus Priority,” in Serena Olsaretti (ed.) Te Oxford Handbook of Distributive Justice. (Oxford: Oxford University Press.) Ottersen, Trygve, Reidun Førde, Meetali Kakad, Alice Kjellevold, Hans Olav Melberg, Atle Moen, Ånen Ringard, and Ole Frithjof Norheim (2016). “A New Proposal for Priority Setting in Norway: Open and Fair,” Health Policy 120: 246–51. Quiggin, John (1993). Generalized Expected Utility Teory. Te Rank-Dependent Model. (Boston, MA: Kluwer Academic Publishers.) Rawls, John (1999). A Teory of Justice, revised 2nd edition. (Oxford: Oxford University Press.) Rawls, John (2001). Justice as Fairness: A Restatement, E. Kelly (ed.). (Cambridge, MA: Harvard University Press.) Roemer, John (2002). “Egalitarianism against the Veil of Ignorance,” Journal of Philosophy 99 (4): 167–84. Rulli, Tina and David Wendler (2016). “Te Duty to Take Rescue Precautions,” Journal of Applied Philosophy 33: 240–58. Satz, Debra (2010). Why Some Tings Should Not Be for Sale: Te Moral Limits of Markets. (Oxford: Oxford University Press.) Segall, Shlomi (2010). Health, Luck and Justice. (Princeton, NJ: Princeton University Press.) Smets, Kaat and Carolien van Ham (2013). “Te Embarrassment of Riches? A MetaAnalysis of Individual-Level Research on Voter Turnout,” Electoral Studies 32 (2): 344–59. Smith, Adam (1982 [1790]). Te Teory of Moral Sentiments, D. D. Raphael and A. L. Macfe (eds.). (Indianapolis, IN: Liberty Fund.) Sreenivasan, Gopal (2014). “Justice, Inequality and Health,” Stanford Encyclopedia of Philosophy (Fall 2014 edition), Edward N. Zalta (ed.), https://plato.stanford.edu/ entries/justice-inequality-health/Online. [Accessed January 2017].

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Government Mandate Comprehensive Health Insurance . . . ?

191

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Stemplowska, Zofa (2009). “Making Justice Sensitive to Responsibility,” Political Studies 57: 237–59. Tawney, Richard (1964). Equality (5th edition). (London: Unwin.) Temkin, Larry (2001). “Inequality: A Complex, Individualistic, and Comparative Notion,” Philosophical Issues 11: 327–53. Vallentyne, Peter (2002). “Brute Luck, Option Luck, and Equality of Initial Opportunities,” Ethics 112: 529–57. Voorhoeve, Alex and Marc Fleurbaey (2016). “Priority or Equality for Possible People?” Ethics 126 (2016): 929–54. Voorhoeve, Alex, Tessa T.T. Edejer, Lydia Kapiriri, Ole F. Norheim, James Snowden, Olivier Basenya, Dorjsuren Bayarsaikhan, Ikram Chentaf, Nir Eyal, Amanda Folsom, Rozita Halina Tun Hussein, Cristian Morales, Florian Ostmann, Trygve Ottersen, Phusit Prakongsai, Carla Saenz, Karima Saleh, Angkana Sommanustweechai, Daniel Wikler, and Afsah Zakariah (2016).  “Tree Case Studies in Making Fair Choices on the Path to Universal Health Coverage,” Health and Human Rights Journal 18 (2): 11–22. Voorhoeve, Alex, Tessa T.T. Edejer, Lydia Kapiriri, Ole F. Norheim, James Snowden, Olivier Basenya, Dorjsuren Bayarsaikhan, Ikram Chentaf, Nir Eyal, Amanda Folsom, Rozita Halina Tun Hussein, Cristian Morales, Florian Ostmann, Trygve Ottersen, Phusit Prakongsai, Carla Saenz, Karima Saleh, Angkana Sommanustweechai, Daniel Wikler, and Afsah Zakariah (2017). “Making Fair Choices on the Path to Universal Health Coverage: Applying Principles to Difcult Cases,” Health Systems and Reform, in press. WHO (2014). Making Fair Choices on the Path to Universal Health Coverage. (Geneva: World Health Organization.)

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

8

Complicity and Coercion Toward an Ethics of Political Participation

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Alex Zakaras

It is often argued that citizens share responsibility for injustices committed by their governments. Let us call this the complicity claim. It has strong and weak versions. Te weak version assigns responsibility to citizens who actively support their government’s unjust policies—by voting or lobbying for these policies, for example, or for the candidates who sponsor them. Tough this version is usually applied to democratic citizens, it can apply elsewhere too: citizens of authoritarian states can also be active supporters of state injustice.1 Te strong version of the complicity claim, on the other hand, assigns responsibility even to citizens who do not participate in politics. Tomas Pogge argues, for instance, that citizens of afuent nations are responsible for their governments’ imposition of an unjust international order on the global poor. Tese citizens are responsible, he argues, even when they are politically passive and largely ignorant of their government’s actions (Pogge  2005, 78–80). Tomas Nagel argues that citizens share responsibility for the actions taken by their governments because these governments act in their name, even when these citizens do nothing more than obey the law and lead strictly private lives (Nagel 2005, 128–30). In fact, the strong version of the complicity claim—even more than the weak version—has become increasingly commonplace in the philosophical debates over global justice.2 Despite its increasing popularity, the strong version of the complicity claim has seldom been defended at length. It is often simply posited, or 1   Tese citizens may freely participate, for instance, in state-sponsored demonstrations in favor of unjust policies. See for instance Jubb 2014, 61–4. 2  See for instance Caney  2008, 514; Miller  2008, 388; Young  2006, 123–4; McMahan 2009, 215.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

193

defended in the space of a few paragraphs.3 Yet it clearly faces steeper philosophical obstacles than the weak version. In the weak version, individual citizens engage voluntarily in collectively harmful activity. In the strong version, the relationship between individual agency and unjust policy is more remote. Politically uninvolved citizens might reasonably ask why they should be held responsible for the behavior of politicians, and even political institutions, they do not condone and had no part in choosing. Moreover, insofar as these citizens do lend support to their government—by paying taxes, for instance—they typically do so under threat of coercion. Tis fact alone presents a serious obstacle to the attribution of moral responsibility, and it has only recently received any sustained attention in the literature.4 In this chapter, I defend the strong version of the complicity claim. I  acknowledge, moreover, that passive citizens are rightly understood as coerced accomplices to state injustice, but I argue that coercion alone does not exonerate them. Even coerced accomplices can become complicit in their coercer’s injustices. Tey become so when, as so many contemporary citizens do, they accept the legitimacy of their own subjection. I argue, furthermore, that citizens’ complicity in state injustice gives rise to an obligation to participate responsibly in politics. I suggest, in fact, that the duty to avoid or mitigate complicity in injustice can serve as the normative basis for an ethics of political participation. Te chapter proceeds in three steps. First, drawing on several recent conceptualizations of complicity, I sketch the outlines of a causal account of individual complicity in collective injustice.5 I focus on cases in which individuals become implicated in another agent’s wrongdoing.6 Second, I apply this account to the circumstances of modern citizenship; in doing so, I defend the strong version of the complicity claim. Tird, I argue that responsible political participation is—in most cases—the appropriate normative response to complicity (or the risk thereof ) in state injustice. COMPLICITY: A CAUSAL ACCOUNT To say that an agent is complicit in another’s unjust actions is to say two things: frst, the agent lends support, as an accomplice, to another who acts   Notable exceptions include Beerbohm 2012; Pasternak 2013; and Jubb 2014.   See especially Pasternak 2013. 5   I draw mainly on three accounts: Lepora and Goodin 2013; Beerbohm 2012; and Kutz 2000. 6   As opposed to cases in which, for instance, people become implicated in an unjust collective action that has no primary agent (such as pollution through the cumulative efect of tailpipe emissions). 3 4

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

194

Alex Zakaras

unjustly; second, she does so in a way that leaves her partly responsible for the primary agent’s actions. I use the term “responsible” here in a specifcally moral sense: complicity in injustice is, in my view, a form of moral wrongdoing. How, then, can one become complicit in another’s unjust actions? I begin with what we might call a perfect case—a case in which the agent’s complicity is as clear and unambiguous as possible; I then reason backward from this case, trying to isolate the features that constitute the necessary conditions of complicity. Consider a citizen named Sally who gives funds to a racist political party. Te party uses these funds to win an election and gain control of a municipal government; it then proceeds to enact racist policies in the city. As she makes the donation, Sally is well aware of the party’s racist agenda; she intends her funds to help the party realize that agenda. Furthermore, her contribution is substantial and well-timed, and the party’s victory narrow, so that it can be said with confdence that the party would not have won without her support. Intuitively, there seems little doubt that Sally is a culpable accomplice. What features of the example explain her culpability?

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

1. She gives the money freely. Since complicity connotes moral responsibility, agents who are not responsible for their actions cannot be complicit. Agents who are forced to lend support to others who act unjustly are not typically judged complicit in injustice. Te relative freedom of Sally’s action, then, is a necessary condition of her complicity. I will argue later, however, the mere presence or threat of state coercion is not always enough to violate this condition. 2. Te injustice occurs. If the primary agent—in this case the racist party— had been unable to act unjustly (say, because it lost the municipal election), then Sally would not have been complicit in injustice. She would surely have been culpable, in a diferent sense, for intending to bring about injustice. But she would not have been an accomplice in injustice. Te occurrence of injustice, then, is another necessary condition of complicity. 3. She contributes (knowingly) to a collective action. In the example at hand, Sally’s funding contributes, causally, to a collective endeavor: the party’s attempt to win the municipal election. Viewed ex ante, her contribution makes it more likely that this collective endeavor will succeed.7 If Sally’s actions had played no causal role at all—if, for instance, she had mistakenly sent the check to the Green Party instead—it would be a mistake to describe her as complicit in the ensuing injustice. As in (2), Sally could rightly be 7   Moreover, there is nothing mysterious about this causal relationship: Sally can reasonably be expected to know that her donation contributes to the party’s probable success.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

195

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

blamed for intending to create injustice. But she would not be an accomplice in wrongdoing. In his important book, Complicity: Ethics and Law for a Collective Age, Christopher Kutz sets a higher threshold for complicity: he argues that individuals become complicit in virtue of intentional participation in collective action, “when they intentionally do their part of what they conceive of as a collective project, and when their conceptions of that group project overlap sufciently” (Kutz  2000, 138). To become complicit, in his view, agents must understand themselves to be acting jointly with others.8 But as Chiara Lepora and Robert Goodin have rightly argued, this feature of Kutz’s account makes it too restrictive (Lepora and Goodin 2013, 81–3). To see why, consider a modifcation of our original example: Sally makes a bet with a friend, and both agree that the loser will have to contribute a substantial sum to the racist party. Sally loses, and so she makes her donation, even though she does not intend to “do her part” in realizing the party’s collective goals. In her eyes, moreover, she is not acting jointly with the party’s supporters at all; she is simply fulflling the terms of her bet. It seems clear that Sally’s self-conception in this case is not exonerating: she shares responsibility for the outcome in virtue of her knowing contribution to the party’s success.9 Tis is an important point for my purposes, because many of the citizens I will describe as complicit in state injustice may in fact experience considerable alienation from their government and may not think of themselves as acting jointly with it at all. 4. Te injustice is a foreseeable consequence of the collective action to which she contributes. We do not typically hold agents responsible for unforeseeable consequences, either of their own individual actions or of collective actions to which they contribute. If I help elect a public ofcial with a record of support for clean energy, and if there is no good reason for an informed and responsible observer to believe that he will change his mind, then I cannot be held responsible if he reverses course once elected. Exactly how much citizens can be expected to foresee is a central question for any ethics of political participation, and I will not attempt to answer it here. In Sally’s case, though, I take this much to be clear: provided that the 8   Robert Jubb and Avia Pasternak have recently reafrmed a version of Kutz’s view of complicity, with intentional participation as the key precondition. See Jubb 2014, 70 and Pasternak 2013, 367–9. 9   Here I follow Lepora and Goodin: “Tus a contributory action,” they write, “rather than a participatory one, is what we deem to be minimally required to constitute complicity” 2013, 81. Not all causal contributors are morally equal, of course: those who make a more substantial contribution will typically bear greater responsibility for the ensuing injustice.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

196

Alex Zakaras

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

racist political party is not a fringe group with negligible support, then its electoral victory is a foreseeable consequence of the collective activity of funding it (or voting for it, or agitating on its behalf ).10 Likewise, if the party has a record of voting for or advocating racist policies in the past, then the racist policies it goes on to implement are also a foreseeable consequence of the collective activity of funding it (or voting for it, or agitating on its behalf ). Here, then, is a fourth, necessary condition of complicity. 5. She intends the unjust outcome. Tough this is true in the perfect case, it is not a necessary condition of complicity. As we saw in (3), Sally can be complicit even if her intentions do not align with the racist party’s. Te same holds true for contributors who are culpably ignorant of foreseeable consequences: imagine another donor to the same political party who is unaware of its racist platform. He disapproves of all forms of racism, moreover, and he gives to the party simply because his pastor advises him to do so. But he fails to do any due diligence before giving; he fails, in other words, to foresee the injustice in which he implicates himself, though it was in fact foreseeable. In this case, absent extenuating circumstances, the agent is culpable. Tough he did not intend to lend support to the racist party—though he did so through negligence—he is not absolved of responsibility. Negligent wrongdoers typically deserve less blame than intentional wrongdoers, and most would agree, I think, that this second donor is less culpable than Sally. Intention does typically matter, in other words, in determining the extent of an agent’s complicity; but it is not a necessary condition of complicity. As Eric Beerbohm points out, treating this criterion as a necessary condition of complicity would also exclude most people’s contributions to structural injustice (Beerbohm  2012, 231–4). When I shop at a discount box store, for example, I may have no intention of contributing to a global system of exploitation that reaches into factories in Bangladesh. But surely the purity of my intention alone is not enough to absolve me from responsibility.11 6. Her contribution causes the injustice in a simple, counterfactual sense. In the perfect case, the racist policies would not have been enacted without Sally’s donation. And intuitively, this fact seems signifcant, for it allows us to say that Sally’s contribution had a decisive efect: without it, the 10   Consider, by contrast, someone who voted for or gave money to Ralph Nader in 2000. His electoral victory was not a foreseeable consequence of the collective action that this agent joined. Al Gore’s loss was. 11   Of course, other facts may exonerate me in this case: I may, for instance, not have resources enough to shop anywhere else.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

197

injustice would not have occurred. Counterfactual causation is not, however, a necessary condition of complicity. To see why, imagine yet another donor, Bill, who, unbeknownst to Sally, simultaneously makes an identical contribution to the racist party. As a result, Sally’s gift is no longer causal in the simple, counterfactual sense: the racist party would still have won the election without it. Rather, her donation has become a redundant cause of the party’s victory. For any injustice that actually occurs, a redundant cause is a necessary condition of its occurrence, along some (but not all) of the possible causal paths by which it might have occurred.12 In this case, Sally’s donation would have been necessary to bring about the  unjust outcome under diferent circumstances (without Bill’s intervention). But Bill’s gift made Sally’s redundant—in fact, his gift made both contributions redundant. Redundant causation is ubiquitous in mass politics. When a candidate collects more votes than she needs to win an election, for instance, no single voter’s ballot causes the outcome in a simple, counterfactual sense. Each vote for the winning candidate is a redundant cause of her victory. But the mere fact that each voter can say, “this candidate would have won even without my vote” does not exonerate any of them. If it did, the result would be genuinely puzzling, for every individual voter would be absolved of responsibility in virtually all real elections. To the question, “which individuals share responsibility for electing George W. Bush to the American Presidency in the year 2000?” the correct answer surely is not “none.”13 Likewise, it cannot be the case that Bill’s donation excuses Sally from complicity in the outcome. Agents can be complicit in injustice even when their actions are merely redundant causes. I defend this point at greater length later in the chapter. With these caveats in mind, let me now summarize the necessary conditions of complicity. Te agent acts freely; the injustice occurs; the agent contributes (knowingly) to a collective action; the injustice is among the foreseeable consequences of that collective action; and her actions cause injustice 12   Note that redundant causation, according to this defnition, is quite ubiquitous. Anything I do that might have caused ensuing event X along some possible causal path is a redundant cause of X. Foreseeability (or lack thereof ) is what frees us from responsibility for the vast majority of the events which we redundantly cause. For a detailed discussion of redundant causation and moral responsibility, see Goldman  2002; chapter two of Tuck 2008; and Beerbohm 2012, 67–72. Lepora and Goodin use the term “potentially essential” to describe redundant causation; see Lepora and Goodin 2013, 61–2. 13   For an excellent discussion of this point, see Beerbohm 2012, 68–71. Te example of a fring squad is often used as an illustration: as tempting as it may be for each individual member of the squad to absolve himself by reasoning that the victim would have died anyway, this reasoning is false.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

198

Alex Zakaras

either in a simple, counterfactual sense or in a weaker, redundant sense. Whenever these fve conditions hold, I will maintain that the agent in question is clearly complicit in injustice. To the extent that the cases and examples I will consider deviate from these conditions, I will ask whether the deviation is enough to erase the agent’s complicity altogether. Te causal account I have just outlined is one of several ways of trying to assign responsibility to individual agents for their contribution to unjust collective action, and I have no space here to consider the full range of other possibilities. Let me simply contrast my approach to an alternative view that pertains specifcally to citizens of democratic states. Anna Stilz argues that democratic citizens share responsibility for their government’s injustices when those citizens have moral reason to accept their state’s authority.14 And citizens have moral reason to accept their state’s authority, she says, when it meets certain basic democratic conditions: it grants them a sphere of freedom, it treats them as moral equals, and it consults them in “the lawmaking process” (Stilz  2011, 202–3). Stilz thus uses the mechanism of hypothetical consent to distribute responsibility to democratic citizens: if you have moral reason to accept your state’s authority, then you share responsibility for its actions—regardless of whether you do in fact accept its authority. As I will try to illustrate in discussing Toreau in the next section, I think Stilz’s theory over-assigns responsibility in some cases: even those who, like Toreau, go to great lengths to protest and reject their state’s claim to exercise legitimate authority over them might still turn out, on her account, to be responsible for its actions. In other ways, however, Stilz’s theory under-assigns responsibility. Stilz is mainly concerned with what she calls “task-responsibility,” or responsibility “to help repair the harm caused by the state” (Stilz  2011, 205, 194–5). She holds that citizens of legitimate democratic states often share task-responsibility for their state’s unjust actions, even though they typically deserve no moral blame (and this view is reafrmed in more recent essays by Avia Pasternak and Robert Jubb). On this last point we disagree: I argue that citizens often share moral responsibility for their state’s unjust actions. One of the advantages of my account is that it better supports the intuition that democratic citizens typically bear some responsibility, not just for repairing state injustice after the fact, but also for preventing their state from becoming unjust in the frst place.15   A version of this argument has also been defended by John Parrish; see Parrish 2009.   It could of course be argued that the desire to avoid future task-responsibility gives citizens reason to prevent their states from acting unjustly in the frst place; but this is clearly not the reason that should be decisive in such cases. 14 15

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

199

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

AGENTS OF INJUSTICE It is perhaps natural to think that citizens who want to avoid complicity in state injustice need simply refrain from participating in politics. Tis is the view held by those who accept the weak version of the complicity claim but reject the strong version. In the rest of this chapter, I argue that this view is mistaken: even citizens who do not participate in politics are often complicit in the injustices committed by their governments. I draw this initial insight from what may seem like an unlikely source: Henry David Toreau, who was among the frst to give it a sustained defense. In the 1840s and 1850s, the question of ordinary Americans’ complicity in injustice became a matter of urgent debate. Like other abolitionists, Toreau believed that even citizens who held no slaves and gave no support to pro-slavery politicians were complicit in the perpetuation of slavery. He rejected outright the idea that citizens who led ordinary, private lives—even citizens of free, northern states—were simply bystanders to this injustice. What was it, then, that made citizens more than bystanders? Toreau’s answer has several parts. “Law never made men a whit more just,” he writes, in the opening paragraphs of “Civil Disobedience,” “and, by means of their respect for it, even the well-disposed are daily made the agents of injustice” (Toreau 2001 [1848], 204). Toreau thought that most of us not only obey our government’s commands but also respect them—we treat them as authoritative. Tese facts alone, in his opinion, make us something more than bystanders. To put it more directly: the agent committing the moral crime, in this case, is an agent whom we respect as authoritative, whose commands we willingly obey. Te enactment of the Fugitive Slave Law in 1850, which required free states to return runaway slaves to their southern masters, drove this point home acutely: here was a law that required New Englanders to actively abet the institution of slavery. In obeying this law, they made themselves accomplices to injustice. But Toreau was writing earlier, in the late 1840s, and his point was broader. All governments, he suggests, depend for their survival on the respect and obedience of their citizens. If enough people lost respect for any law—and refused to obey it—then it would cease to exist. It would lose all legitimacy, and government would be unable to enforce it without tremendous coercive efort.16 Tis is what Toreau means when he writes, 16   H.L.A. Hart famously argued that the rule of law depends on “a general habit of obedience” among the public, and that no government has the power to enforce compliance to even most of its laws most of the time. Governments can only sustain a rule of law if violators constitute only a small fraction of the population. Hart and Honore 1959, 23.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

200

Alex Zakaras

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

I know this well, that if one thousand, if one hundred, if ten men whom I could name—if ten honest men only,—ay, if one HONEST man, in this State of Massachusetts, ceasing to hold slaves, were actually to withdraw from this copartnership, and be locked up in the county jail therefore, it would be the abolition of slavery in America. For it matters not how small the beginning may seem to be: what is once well done is done forever. (Emphasis in original; Toreau 2001 [1848], 212)

Governments and laws survive as such by means of citizens’ deference. Once this deference is withdrawn, they fail. To restate the point, then: the agent committing the moral crime is an agent who derives his very power to act from our deference, our respectful obedience. Tax law ofers the clearest, most concrete example of this enabling relationship. Without revenue, the government cannot act, cannot even constitute itself as a collective agent by hiring ofcers to carry out its orders. And its revenue comes from us, because so many of us pay our taxes. And so—famously—Toreau refused to pay his taxes. “I simply wish to refuse allegiance to the State,” he writes, “to withdraw and stand aloof from it efectually” (Toreau 2001 [1848], 219). Toreau thought that those who hoped to avoid complicity simply by refusing to be involved in political life, while all the while respectfully paying their taxes, were fooling themselves. Tey had not withdrawn efectually. Tey had simply become negligent custodians of their own funds and allegiance. Tere is one more piece to Toreau’s story. Not only do we respect the state and fund the state, we also willingly receive—indeed, we depend on— the benefts it confers; we willingly proft from it. He blames the “hundred thousand merchants and farmers here, who are more interested in commerce and agriculture than in humanity, and are not prepared to do justice to the slave and to Mexico, cost what it may” (emphasis in original; Toreau 2001 [1848], 207).17 Te state confers important material benefts—it protects our property, for instance, and oversees and enforces our contracts—and virtually all of us take these benefts willingly. In any case, we take no pains to avoid or forfeit them. In accepting such benefts without objection, we lend further legitimacy to the regime; we signal our acquiescence to its authority.18 Te agent committing the moral crime, then, derives its power from us, through a combination of our deferential obedience, our funds, and our willingness to proft from and depend on it. 17   He also maintains that those who accumulate property will always have trouble following their conscience against the state, because they have so much to lose. 18   Tis should not be mistaken for a fair-play argument. Te claim is simply this: citizens who willingly proft from their association with their state are, in so doing, expressing some measure of acquiescence to its authority.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

201

Does Toreau’s argument succeed in establishing citizens’ complicity? Consider the fve necessary conditions I outlined earlier, in reverse order. First, Toreau thought that docile compliance (including the payment of taxes and the receipt of benefts) was a form of causal infuence on which the state depended. When the state succeeded in acting unjustly, in other words, each individual’s compliance was a redundant cause of its success. Second, Toreau thought that slavery and war were foreseeable consequences of citizens’ compliance. Tese were among the widely acknowledged aims of the U.S. government in the late 1840s (so much so that many abolitionists thought that northern states should themselves withdraw from the union). Tird, citizens at the time knew that their docile compliance, and their payment of taxes, contributed to successful state action.19 It was widely understood at the time, for instance, that citizens’ voluntary adherence to law was a precondition of functional governance. Fourth, the injustices he decried were actual, not just potential. And fnally, Toreau went to great lengths to emphasize the fact that citizens could freely choose not to comply; in fact, his own act of disobedience was calculated to dramatize this freedom. Toreau believed that citizens were therefore obliged to withdraw their support for injustice by withdrawing their compliance from government. Toreau’s treatment of this last condition—the freedom to disobey—is unpersuasive, and I will have more to say about it in the next section. But before turning to this problem, let me return to another important objection, which applies especially to causal accounts of complicity. Many people believe that the actions of (most) individual citizens in mass democracies are simply too small and too distant from government injustices to carry moral responsibility in the frst place. Tis concern is best expressed by what Kutz calls the “Individual Diference Principle”: I am accountable for a harm only if what I have done made a diference to that harm’s occurrence . . . I am accountable only for the diference my action alone makes to the resulting state of afairs.  (Kutz 2000, 116)

Tis principle is frmly grounded in commonsense moral intuitions, and it  is relevant to Toreau’s argument. It suggests that since no individual citizen’s deference or tax money is necessary, either to the conduct of the Mexican American War or to the perpetuation of slavery, individual citizens cannot be held accountable for these unjust state actions. No individual could, simply by withdrawing support for the government, have made any diference, either to slavery or to the war. On this view, agents cannot act 19   Such knowledge is still commonplace today: one need only consider the common lament that some absurd, inefcient, or overly bureaucratic government initiative shows “our tax dollars” at work.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

202

Alex Zakaras

wrongly in virtue of what Kutz calls “marginal participation” (Kutz 2000, 125). Unless they can exert causal control in a simple, counterfactual sense, or unless they have a nonnegligible chance of doing so, citizens are not responsible for the actions taken by their governments. Tose who are thoroughly convinced of the validity of the Individual Diference Principle will not be persuaded by the argument of this chapter. Tey will hold that complicity itself is relatively hard to come by—especially in large polities, where individual citizens wield very little causal power over political events. In my opinion, the strongest response to this view is simply that it leaves unacceptably large accountability “gaps,” and that such gaps are themselves inconsistent with the experience of many refective moral agents. Kutz drives these points home with a detailed discussion of the frebombing of Dresden by Allied aircraft in 1945. Te deliberate destruction of large residential areas of the city was deliberately planned and, as Kutz puts it, “as overdetermined as can be imagined”:

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Te city was bombed in three raids, and at least 1,000 planes and 8,000 crewmen were directly involved . . . Te frestorm was already raging when many crews dropped their bombs. Each crewman’s causal contribution to the confagration, indeed each plane’s, was marginal to the point of insignifcance.  (Kutz 2000, 118)

Te Individual Diference Principle suggests that none of the individual contributors to the raid shared responsibility for the outcome, for their own  actions made no morally signifcant diference. Here, then, is the accountability gap: who was responsible for the bombing raids if not the volunteer pilots who carried them out?20 Was it strictly the planners—or perhaps, strictly those planners whose contributions were indispensable to the eventual outcome? And if no individual planners were indispensable to the outcome, does that mean that responsibility for the raids vanishes entirely? Tis view suggests, perversely, that agents can exonerate themselves from crimes whenever they are joined by enough accomplices to overdetermine the result. In fact, as Kutz elaborates, many of the contributors do and did feel responsible, after the fact, for having facilitated or carried out what is now widely considered an atrocity (Kutz  2000, 118–22). Likewise, many abolitionists in Toreau’s day came to feel complicit in slavery in virtue of their knowing contributions—as taxpayers and obedient citizens—to unjust 20   It might be argued that the pilots faced sanctions had they refused to go along, and they were therefore efectively coerced into obeying. Kutz writes that pilots who refused ran the risk of being “labelled ‘LMF,’ for ‘Lacking Moral Fibre,’ and drummed out or sent to desk duty” (Kutz 121). But this is beside the point here: even absent these relatively mild sanctions, the Individual Diference Principle would exonerate the individual pilots and crew members, simply because their individual contributions made no morally decisive diference to the outcome.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

203

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

collective action. Tey came to feel, in other words, that the intuitions that underlie the Individual Diference Principle only concealed or rationalized their own culpability. And so Toreau, like Kutz, would have rejected the Individual Diference Principle. Of course, the principle’s inadequacy does not in itself show that Toreau’s account of complicity is correct. Tere are many ways of trying to close the accountability gaps that the Individual Diference Principle leaves behind.21 But it does suggest that one of the most obvious, intuitive responses to Toreau’s argument is mistaken. One further observation is worth making about Toreau’s account of complicity: it is not unique to democratic polities.22 Democratic governments may be especially dependent on their citizens’ deference, but no government can survive without it. “Notice,” one philosopher observes, “that for a political unit to act in a given fashion, it is not enough merely to have highlevel ofcials actively push for such action or even to have many citizens wholeheartedly support it; it is also necessary that the vast majority of constituents acquiesce rather than actively resist” (Wellman 2007, 740).23 Virtually all governments depend on their citizens’ acquiescence, expressed both through their obedience and their willing receipt of benefts.24 All governments also depend on the taxes that their citizens pay. Toreau’s account of complicity does not, moreover, depend on the act of voting or any other participatory gesture. It is therefore diferent from the weak version of the complicity claim, which is more commonplace in democratic theory. John Rawls invoked this weak version when he asked, When may citizens by their vote properly exercise their coercive political power over one another when fundamental questions are at stake? Or in the light of what principles and ideals must we exercise that power if our doing so is to be justifable to others as free and equal?  (Rawls 1993, 217)   See for instance Kutz’s discussion of Derek Parft, pp. 129–32.   In this sense, it difers from Beerbohm’s account. Beerbohm has given the subtlest and most comprehensive account of citizens’ complicity in state injustice to date (an account which has received far too little attention in the recent literature). He focuses on the causal power of the vote: the right to vote, he argues, puts democratic citizens’ hands on the causal levers of political power, and they bear responsibility for both what they do with it and what they fail to do; see Beerbohm 2012, 63–81. My own argument follows Toreau in appealing to diferent causal pathways: citizens become complicit in state injustice in virtue, mainly, of their obedience to law and payment of taxes. My argument therefore runs more squarely into the problem of state coercion. In the end, though, I  believe that it ofers a stronger, broader foundation for the complicity claim. 23   Jubb makes a similar point in discussing authoritarian regimes that rely on “the acceptance of the mass of the population”; Jubb 2014, 61. 24   Te dramatic events in Tunisia, Egypt, and Libya in the winter of 2010–11 bear witness to the power that citizens possess to abruptly withdraw their acquiescence, even from authoritarian governments, and force political change. 21 22

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

204

Alex Zakaras

When we vote for a candidate or a law, we exercise (potentially) coercive power over others. In virtue of this exercise of power, it might be said, we become morally implicated in the result. Toreau would have agreed. Tose who voted for or otherwise supported pro-slavery candidates became, in virtue of this support, complicit in any injustices those candidates committed or created. Te degree of their complicity, moreover, varies in proportion to the extent of their support. But Toreau was interested in a prior form of complicity—one that obtains between all obedient citizens and their governments. Tis form of complicity casts a broader net that enmeshes even nonvoters, nonparticipants. In fact, representative democracy involves its citizens in two diferent causal relationships, both of which threaten to render them complicit in injustice. Tese two causal relationships correspond to the strong and weak versions of the complicity claim. First, in virtue of the fact that they respect the law, pay taxes, and accept state benefts, citizens lend causal support to their government. If their government does wrong, they risk becoming complicit in these wrongs—call this political complicity. Second, as participants in politics whose participation lends support to their government’s behavior— by voting, giving money, organizing, writing letters, etc.—citizens help authorize and direct their government’s actions. If they support unjust policies or candidates, citizens risk becoming complicit in these wrongs—call this participatory complicity.25 Troughout this chapter, I focus on complicity of the frst type.26

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

COERCED ACCOMPLICES Te most important objection to Toreau’s argument, in my view, is that refusing to obey the state is costly—so costly that citizens are efectively forced to obey, to pay taxes, and to receive state benefts. Tey do not act freely. Tere are two possible forms of refusal at issue here, both of which are costly. Te frst is selective refusal: refusal to obey unjust laws, to pay taxes that support these laws, and to accept state benefts. Toreau’s own example illustrates the costs associated with these forms of civil disobedience.27 Tose who refuse to pay taxes, or to obey the law, go to jail. And those 25   I use the term “participatory” here to refer to political participation, not to Kutz’s idea of intentional participation in collective wrongdoing. 26  Again, this focus distinguishes my view from Beerbohm’s; Beerbohm ofers a specifcally democratic, participatory account of complicity. 27   Toreau’s own withdrawal was selective in this sense. He continued to pay highway taxes, for instance, claiming that he used the roads himself.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

205

without the good fortune of having relatives who will pay their taxes for them (as Toreau did) stay in jail.28 Even democratic laws are, after all, backed by the threat of coercion. And though the growing number of people going “of the grid” might suggest otherwise, the cost of systematically renouncing state benefts is, in the world as we know it, still steep.29 Te second kind of refusal is, of course, total withdrawal: emigration and the renunciation of citizenship. Such renunciation is perfectly legal in many countries; citizens are not therefore coerced into remaining members of their polities. Still, the costs of emigration are typically very high. Leaving aside, for a moment, everything that one renounces when one leaves one’s own country, it is often difcult and costly to fnd another country to emigrate to. We do not ordinarily hold people responsible for failing to absorb immense personal costs in order to avoid making a small contribution to collective injustice.30 Indeed, Socrates’ famous argument notwithstanding, the possibility of emigration is not enough to render citizens’ compliance with state commands voluntary. Citizens who do not wish to comply with unjust commands are, in efect, told: you must either submit to legal penalties (fne or imprisonment) or you must leave your life here permanently. Tose who had defended or posited the strong version of the complicity claim have not grappled with this problem sufciently.31 Te moral signifcance of state coercion can be clarifed by considering the diference between the state and a voluntary association. If I joined an association among whose explicit purposes were the perpetuation of slavery and the conduct of unjust foreign aggression, if I paid my membership dues, obeyed the association’s rules, and benefted from its services, I would clearly share responsibility for its actions.32 Even if I had inherited my membership in the unjust association—if I had not joined of my own volition—as soon as I was old enough to be responsible for my actions and 28  Interestingly, Toreau condemned those who paid the tax on his behalf for neglecting the common good: “If others pay the tax which is demanded of me, from a sympathy with the State, they do but what they have already done in their own case, or rather they abet injustice to a greater extent that the State requires. If they pay the tax from a mistaken interest in the individual taxed, to save his property, or prevent his going to jail, it is because they have not considered wisely how far they have let their private feelings interfere with the public good” (Toreau 2001 [1848], 220). 29   And of course, certain benefts—such as national defense and the rule of law—cannot be renounced without leaving the country altogether. 30  Such sacrifce might well be obligatory in some cases. For the purposes of this chapter, however, I proceed on the more modest premise that we have an obligation to avoid complicity in injustice when we may do so at no great cost to ourselves. 31   Pasternak’s essay (2013) is a notable exception. 32   Tis would be true even if my contribution was very small and causally redundant, and this fact too shows the inadequacy of the Individual Diference Principle.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

206

Alex Zakaras

as long as I remained a member, I would share responsibility for the association’s crimes. I would be complicit in its wrongdoing—at least as long as the costs of my withdrawal were not prohibitively high. Toreau argues that, despite the coercive nature of the state, responsibility fows just as it would in a voluntary association: so long as he remained an obedient member of the state, he thought, he would share responsibility for its crimes, just as he would for the crimes of any other association to which he belonged. He had to extract himself, whatever it might cost. I have already suggested that this is too exacting a standard. But this fact hardly restores citizens to the status of bystanders. Instead, it suggests that citizens are coerced accomplices, contributing to unjust actions under duress. Te key claim that needs attention, then, is the claim that coerced accomplices are, in virtue of their coercion, absolved from complicity in injustice. In this claim is true, then Toreau’s account of complicity fails. In my view, however, it cannot withstand careful scrutiny: in fact, coercion absolves citizens from complicity only under certain conditions. Toreau himself would have been quick to point out that very few of his fellow New Englanders behaved like coerced accomplices. Tey respected the state; they freely acknowledged its authority (even as they grumbled privately about slavery); they welcomed its benefts. Accomplices who gladly cooperate with their coercer, or who willingly do more than their coercer demands, are no longer fully absolved by the presence of coercion. Likewise, an accomplice who accepts the legitimacy of his coercer remains (at least partly) responsible for his own actions.33 And this is precisely the situation of most citizens today: they continue to accept their government’s legitimacy despite the injustices it commits. In an intriguing footnote to “Te Problem of Global Justice,” Tomas Nagel argues that even subjects of an oppressive colonial regime might share responsibility for the injustices it perpetrates: But I think it can be said that if a colonial or occupying power claims political authority over a population, it purports not to rule by force alone. It is providing and enforcing a system of law that those subject to it are expected to uphold as participants, and which is intended to serve their interests even if they are not its legislators. Since their normative engagement is required, there is a sense in which it is being imposed in their name.  (Nagel 2005, 129)

By “normative engagement,” Nagel seems to mean this: even the subjects of  an oppressive regime are asked to do their part in maintaining social and political order. Tey are asked, at a minimum, to accept the regime’s 33   Tere are some obvious exceptions, including for instance victims of Stockholm syndrome.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

207

legitimacy and to obey its rules willingly (even when they could disobey without being detected).34 Nagel argues that this normative engagement can render citizens complicit in their government’s injustices even in relatively oppressive states; he believes, therefore, that state coercion is not enough to exonerate them. To evaluate this view, we must ask what it is, exactly, about coercion that is supposed to exonerate agents who contribute to injustice. Consider the following example: a member of the racist political party comes to my door and persuades me to give a monthly donation. He makes it clear, moreover, that once I sign up, there’s no backing out: I will be coerced into fulflling this obligation, every month, as long as I am able to pay. Suppose that I agree to these terms and sign on. From then on, I face a constant threat of coercion, and in some cases this threat moves me to act in ways that I would not otherwise have acted. In some months, when my budget is tight and I would not otherwise have paid, the threat of violence is what moves me to contribute. In these instances, I act under duress. Suppose, however, that I never cease to accept the party’s coercion as legitimate: like the other members, I understand that coercion is necessary to assure members that everyone is doing their fair share. For coercion to be exonerating, it must force me to act against my will. In this example, it does this only in a limited sense: it compels me to act, in some cases, in ways I might not have otherwise. But because I accept the coercion to which I am subject as legitimate, there is a higher-order sense in which these actions are not against my will at all. Harry Frankfurt argues that coercion typically compels us to act on motives that we would prefer not to act on (Frankfurt 1988 [1973], 42–4). In our example, this condition is not satisfed, for I accept that the threat of coercion is necessary to the party’s success, and I so accept being coerced. Fear of punishment is an inducement that I would prefer, all things considered, to be subject to.35 Let me now modify the example to make it more analogous to the colonial subjects in Nagel’s footnote. Suppose that rather than being persuaded to give 34  I use the term acceptance in keeping with L. Jonathan Cohen’s defnition: “to accept that p is to have or adopt a policy of deeming, positing, or postulating that p—that is, of going along with that proposition (either for the long term or for immediate purposes only) as a premise in some or all contexts for one’s own and others’ proofs, argumentations, inferences, deliberations, etc.” (1989, 368). Te principal context I have in mind here is moral: to accept that my state is legitimate—that is, to accept that it has a right to rule me—has a number of implications for my own deliberations as to how I ought to act. To reject p, on the other hand, is to adopt a policy of deeming, positing, or postulating that not-p. 35   For my purposes here, I am agnostic as to whether we should say, about such cases, that I am not coerced at all, or simply that coercion is not exonerating. Either alternative suits my argument.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

208

Alex Zakaras

monthly contributions, I am simply born into this arrangement (my father was a monthly donor, as was his father, etc.). Suppose, moreover, that over time I come to accept the racist party’s work as legitimate, and I feel glad that I was born into it. My approval of the party is not the result of brainwashing or manipulation; it refects my own considered judgment. It seems clear, in this case, that the party’s coercion does not exonerate me for contributing to its injustices. Once I come to accept the party as legitimate—and, by extension, once I accept its coercion of its own members as a legitimate means of accomplishing its ends—I am responsible for my participation in it, for precisely the same reason that proved decisive in the earlier case: I accept the coercion to which I am subject as legitimate. Nagel’s colonial subject stands in a morally analogous position: so long as she accepts her regime as legitimate, and so long as her acceptance refects her own considered judgment, she cannot claim that it forces her to act against her will. To accept a political regime as legitimate is to accept the legitimacy of its coercion, specifcally. It is to accept that it has the right to rule—not just others, but also oneself. And the right to rule entails the right to issue commands backed by the threat of force.36 Supposing that the colonial subject meets the other conditions of complicity, then, she shares responsibility for her regime’s injustices.37 One of the peculiar features of Nagel’s argument, however, is that he seems to treat complicity as an unavoidable feature of political membership. He fails, in any case, to clarify that citizens of any regime can withhold their normative engagement and in doing so avoid responsibility for its actions.38 In the next section, I consider how this can be accomplished. If my argument in this section is correct, then coercion or the threat thereof—even when very substantial penalties are involved—is not enough to absolve citizens from complicity in their government’s injustices. Coercion does not absolve those who continue to accept their regime’s authority as legitimate. And since most citizens of afuent modern societies do regard their governments as legitimate, this argument is sufcient to show that 36   Tere are, of course, limits to this right. To accept a government’s legitimacy is to accept its right to command me even when I disagree with its decisions; it does not require, however, that I accept its right to issue any command whatever. 37   Beerbohm says this about Nagel’s colonial subject: “She lacks the most basic power of the citizen . . . ‘the present and future capacity to infuence politics’” (Beerbohm 2012, 49–50). Beerbohm concludes that she cannot therefore be thought to co-author the colonial regime’s policies, and that she cannot be complicit. As I have already suggested, I think Beerbohm takes too restrictive a view of complicity here; see note 22. 38   Nagel describes normative engagement as “a special involvement of agency or the will that is inseparable from membership in a political society;” (emphasis mine; Nagel 2005, 128).

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

209

they cannot appeal to state coercion to defeat Toreau’s argument and avoid complicity in political injustice. Of course, some regimes are so thoroughly controlling that they extinguish their subjects’ responsibility altogether. Totalitarian regimes, for example, typically control their subjects, not only through pervasive coercion and surveillance, but also through intense propaganda and information control. Such regimes, as Robert Jubb puts it, “might be usefully described as substituting their will for their citizens’ by preventing them from forming one in the frst place” (Jubb  2014, 67). Where citizens have virtually no room for free thought or action, they cannot be complicit in injustice at all—even if they fully accept their regime’s right to rule.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

WITHHOLDING NORMATIVE ENGAGEMENT What must citizens do, then, to withhold or withdraw their normative engagement? Toreau’s answer is clear: they must disobey the state’s commands. For Toreau, this was especially clear in deeply unjust polities: in the face of severe injustice, says Toreau, the only refuge for the just citizen is in jail. I have already rejected that view as too demanding (except perhaps in the most extreme cases). But what are the alternatives? Is it enough for citizens to privately reject their government’s right to rule them, even while outwardly obeying its commands?39 Is this rejection alone sufcient to avoid complicity in its injustices? Te answer, I will suggest, depends on the extent of the coercion to which they are subject. Consider, again, Nagel’s idea of normative engagement. Te colonial regime, he writes, “is providing and enforcing a system of law that those subject to it are expected to uphold as participants” (Nagel 2005, 129). Te regime’s demand here is both attitudinal and behavioral. Subjects are asked to accept that the regime has the right to rule. But more importantly, they are asked to act as though they accepted it: to obey its law and other commands, to pay taxes, to help identify troublemakers, to withhold criticism of the regime, etc. Of subjects who comply with the behavioral demands but not the attitudinal demands, we might say that though they privately reject the regime’s claim to legitimacy, they publicly accept it. In some contexts—when agents are subject to acute and comprehensive coercion—public acceptance of even profoundly unjust regimes is morally permissible. If I am forced, with a gun to my head, to comply with my regime’s every demand, I will typically bear no responsibility for my behavioral compliance. Te comprehensiveness of my coercion makes it 39

  For a defnition of reject, see note 34.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

210

Alex Zakaras

unreasonable to expect any outward deviation from my coercer’s will. In this case, my private judgment alone is decisive: so long as I regard the regime as illegitimate—that is, so long as I reject its right to rule me—I am not responsible for the actions it forces me to take. But even authoritarian regimes are not typically this repressive. Many allow some space for dissent. And where there is room for dissent, private attitudes alone are no longer decisive.40 For simplicity’s sake, consider another stylized example. Suppose that I was born into a village governed by an authoritarian junta that I have come to regard as wholly illegitimate. Te  junta is considering launching an unjust and unprovoked attack on neighboring villagers. Te junta’s leaders hold a public meeting, which I attend, to canvass local opinion, because they are hesitant to pursue any course of action that is highly unpopular. Tey make a credible commitment to allow free and open debate. Te debate is held, but public opposition is not strong enough to persuade the junta to forego the attack. I say nothing and leave the discussion to others. I am subsequently forced to contribute, in some way, to the unjust attack. When I later claim that coercion absolves me from responsibility for the ensuing crimes, my claim is likely to be received with some skepticism, for I had the opportunity to try to prevent the injustice altogether, at little cost to myself, and I did nothing. To bring the moral signifcance of this omission into focus, imagine that I am later challenged by some of the villagers who were attacked. Tey hold me responsible, they say, in virtue of my contribution to the attack against them. In my defense, I claim that I was forced to contribute against my will. No one would be surprised if the victims then asked: “Really? Did you speak out against these crimes when you had the opportunity?” And if not, “Why not?” It is important to see what the victims are asking here. Tey are trying to ascertain the extent of my normative engagement with the junta. Like Toreau, they suppose that, in most contexts, strictly private rejection of the regime’s legitimacy is not enough to achieve normative disengagement. Where there exist (safe) opportunities for dissent, silent compliance with unjust directives is not a morally adequate response. Silent compliance is 40   Pasternak makes this point too: “I presume,” she writes, “that when citizens do not engage in activities that signal their resentment of their state . . . it is reasonable to think that they are intentional participants” in their state’s injustices. “Te onus is on them,” she continues, “to signal that they are not its intentional members” (Pasternak  2013, 375). I have already argued that intentional participation is too high a threshold for complicity in injustice, but her broader point here is sound. It needs further justifcation, however, to overcome the skepticism that many citizens may naturally feel when told that they are responsible for the actions of public ofcials that they never consciously endorsed and may in fact actively despise.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

211

what my unjust coercers want, and I have given it to them freely.41 In such cases, normative disengagement can be achieved only through public action. Consider these possible answers that I might give to the victims’ questions: “No, I did not speak out, I was too busy playing with my latest iPhone app.” “No, I was preoccupied with my work.” “No, I don’t like to involve myself in contentious discussions.” Tese answers constitute clear evidence of moral failure on my part, and they would surely be received as such by the victims.42 Te failure in question is not, however, merely a bystander’s failure of  omission. Te victims’ point of view is instructive in clarifying the diference between the coerced accomplice’s position and the bystander’s. Te victims have standing to demand specifcally that I—one of their aggressors—stop harming them. As Beerbohm puts it, “wronged individuals seem to have a special kind of authorization to address and make demands upon their aggressor” (Beerbohm 2012, 44). Te victims’ moral demands are addressed to me, particularly, with an urgency that does not extend to mere bystanders. As a coerced accomplice, my frst line of defense is coercion itself: “I was forced to contribute to the attack against my will.” But as I have been arguing, certain conditions must be met for this defense to be persuasive. And in the example I have just given, they are not met. In this case, my victims can rightly claim that I had an opportunity to try to avoid acting unjustly (by lending my voice to the opposition), and I didn’t even try. From their point of view, the existence of such an opportunity must change my default position, as it were, from innocent to guilty. Whether I intended it or not, my silence triggered normative engagement with the local regime. So far, I have been arguing that citizens who wish to withdraw normative engagement from their regime, and hence avoid complicity in its injustices, must typically do more than privately reject its claim to legitimacy. It remains to be asked: what is the appropriate normative response, in the frst place, to complicity (or the risk thereof ) in state injustice? Ought we to strive, in all cases, to avoid such complicity by withholding our normative engagement? For citizens of illegitimate regimes—such as the authoritarian junta—the answer is clearly yes. Tose who accept the legitimacy of illegitimate regimes, and so knowingly lend them support, do wrong.

41   Tey would prefer vocal support, of course. But most illegitimate regimes ultimately thrive on passive, silent acquiescence. 42   Tis is not to say that there are no satisfactory answers, e.g.: “no, I had good reason to believe that the junta would retaliate against dissenters in spite of its commitment to allow open debate.”

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

212

Alex Zakaras

For citizens of legitimate regimes, however, the answer is more difcult. It is possible, of course, to hold that we ought to avoid complicity in all cases, and that we ought therefore to withhold normative engagement from any state that is even partly unjust. But the implications of this view are truly radical, for it forbids normative engagement in any existing state.43 It also contravenes widely accepted ideas about political obligation: most plausible accounts of political obligation hold that even governments that are not perfectly just can give rise to binding political obligations. Since there is no space to assess the adequacy of these accounts here, I will simply suppose, for the sake of argument, that they are correct on this point. I will suppose, in other words, that citizens of legitimate regimes may have compelling moral reasons to remain normatively engaged, even in the face of their own complicity in state injustice. And in the next section, I argue that the appropriate normative response to such permissible complicity is responsible political participation.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

THE DUTY TO PARTICIPATE By responsible participation I mean political participation that tends to make the polity more just. By political participation I mean the full range of public actions that citizens can take to exert a causal infuence on political “outcomes,” including voting, protesting, funding campaigns or advocacy groups, organizing, staging acts of civil disobedience, and more. In my usage, participation is an inclusive category; it does not connote playing by the rules, joining existing political parties, or even taking part in mainstream political procedures or institutions. Citizens who take to the streets rather than cast their ballots in elections whose legitimacy they contest are still, in my sense, participating in politics (even as they refuse to participate in elections). My claim in this section is simply that permissible complicity in state injustice, or the threat thereof, gives rise to an obligation to participate responsibly in politics. Consider again the way responsibility accrues to members of voluntary associations: if I continue to associate (permissibly) with a legitimate group despite the fact that some of its policies are unjust, I am obliged to take action to oppose the group’s unjust policies. Tis much seems fairly uncontroversial. If I fail to take action, I am subject to moral blame. Tis blame arises, moreover, from the claims of the victims in whose victimization I participate. As in the example of the village junta, the victims’ claims 43   It also seems likely to forbid membership in any state (or statelike entity) that might come to exist in the world as we know it.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.



Complicity and Coercion

213

generate urgent, agent-relative moral reasons: I must stop treating them unjustly.44 I may do so by leaving the group; or, if I choose to stay, I must act on these reasons by working to stop the injustice to which I am an accomplice. Te same reasoning also applies to citizens who are permissibly complicit in state injustice: if they choose to remain normatively engaged with their government, they are obliged to take action to oppose the state’s unjust policies. To bring this argument into clearer focus, consider a fnal example. Noah  lives in small-town Vermont and attends town meeting in his local community. Te town is preparing to enact a policy that will unjustly harm members of several neighboring communities. At town meeting—which Noah silently attends—the townspeople deliberate openly and then approve the unjust policy. Afterward, Noah remains normatively engaged in his town. He rightly accepts that his municipal government remains legitimate despite the unjust policy. He pays local taxes; he sits on the local school board; he works with town ofcials on local charity events. If the main argument of this chapter is correct, then Noah is complicit in his town’s injustice. Now imagine that some neighboring townspeople (who are unjustly harmed) approach Noah and demand that he account for his contribution to the injustice. What might he say in his own defense? He cannot claim that his contribution was obtained against his will; as I have argued, Noah’s normative engagement vitiates this line of response. Still, suppose that he responds as follows: “While it is true that I accept my government’s right to coerce me in general, I never accepted its right to force me to contribute to this particular unjust policy. I therefore maintain that my contribution to injustice was, in this case, extracted against my will.” Tis response runs squarely into the problem that I explored in the previous section: provided that the other conditions of complicity are met, coerced accomplices who have ample opportunity to resist their coercers, at no great cost to themselves, and fail to do so, cannot plausibly disclaim all moral responsibility for their coercer’s unjust actions. Noah might try to defect blame in a diferent way, by arguing that he was obliged to contribute to his government’s injustice—through his general obedience and his taxes—because legitimate governments give rise to a duty of obedience. But this response, too, is clearly inadequate, because obedience and disobedience are not Noah’s only options. After all, he could have spoken out against the injustice before it was enacted. He could also have invested time and energy working (within the law) to reverse the unjust policy. Had he done so, he could say to the victims: “I recognize that I have 44  See Beerbohm  2012, 42–4. In the case of new unjust policies still awaiting ratifcation, it is the potential victims who have a distinctive moral claim on me.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

214

Alex Zakaras

an obligation to stop treating you unjustly; I am acting on this obligation, but my actions are being overcome by the political infuence of others.”45 Such actions would not have erased Noah’s complicity altogether—in polities that are less than fully just, complicity is one of the unavoidable costs of normative engagement. But they would have mitigated his complicity and, if the extent of his opposition were commensurate with the scope and severity of the injustice, they might even have excused his complicity. What seems undeniable, in any case, is that from the victims’ point of view, Noah’s political opposition would have mattered morally. Notice that Noah’s responsibility here is not exhausted by what Stilz calls “task-responsibility.” When victims reproach Noah for his complicity in his town’s injustice, they are not merely trying to assign responsibility for repair or remediation (though they may be doing this, too). Tey are assigning moral blame. Te blame they are assigning, moreover, is diferent from the blame they might have assigned to a bystander who failed to intervene on their behalf. In their eyes, Noah is a contributor to injustice who has failed to excuse his contribution. Tere is some similarity, here, between my view and expressive justifcations of voting and participation. Expressive justifcations conceive of participation as “a form of moral self-expression”; the central idea is that such expressions can be morally signifcant acts: when we step forward and publicly condemn our government’s injustice, we are taking a stand and repositioning ourselves, morally, relative to our government’s actions (Benn 1988, 19). But here again, the victims’ point of view is instructive: from their vantage point, merely expressive justifcations are bound to seem insufcient. From the victims’ point of view, what matters is not strictly that Noah expressed disapproval of the injustice, but that he threw his causal weight, so to speak, against the injustice. He tried to stop treating his victims unjustly, and had enough of his fellow townspeople done so as well, the injustice would have ended. From the victims’ point of view, merely symbolic expressions ring hollow, for they seem self-involved. It is in the making of a causal contribution—in actually trying to end or prevent injustice—that the victims’ claims are appropriately acknowledged and incorporated in the complicit agent’s practical reasoning. At this point, the Individual Diference Principle may resurface as grounds for objection: critics might still argue that the causal insignifcance of any ordinary citizen’s participation undermines the obligation to participate. After all, in most polities, most individual citizens cannot rationally hope to 45  In this instance, Noah would be negotiating two competing obligations: the obligation to obey a legitimate government and the obligation to avoid injustice. I am arguing that responsible participation is the only permissible way to do so.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

215

correct any political injustice; their causal contribution is too insignifcant. But again, this objection misses the mark. Suppose that Noah had been the only one in his village to speak up against the injustice and suppose he had done so knowing that he would fail to persuade a majority of his peers. From the victims’ point of view, his action loses none of its moral signifcance. If anything, it gains value because of the moral courage it embodies.

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

CONCLUSION In the frst half of this chapter, I set out to defend what I have called the strong version of the complicity claim, or the view that politically passive citizens often fnd themselves complicit in state injustice. I outlined fve necessary conditions of complicity and argued that citizens who do no more than pay taxes, respect the state’s authority, and receive state benefts often meet these conditions. I called the form of complicity to which these citizens fall victim political complicity (as distinct from participatory complicity). I suggested, furthermore, that political complicity threatens citizens in two diferent situations. First, citizens who reject the legitimacy of their state—or indeed, citizens who simply go along with it without forming any opinion of its legitimacy—can become complicit if they fail to make use of opportunities to safely oppose its unjust actions. Such failures, as in the example of the village junta, signal normative engagement with their governments. Only by taking political action against injustice can such citizens achieve normative disengagement; only by achieving normative disengagement can they plausibly claim that their contributions to injustice were obtained against their will. Second, I have argued that citizens who, like Noah, afrm the legitimacy of their state are unavoidably complicit in the injustices it commits. Tey can, however, mitigate or excuse their complicity through responsible political participation. In both cases, passive compliance is morally unacceptable. Of course, citizens in both situations will sometimes fnd themselves exonerated by circumstances that violate one or more of the fve necessary conditions of complicity. Consider the foreseeability condition alone: when citizens cannot reasonably be expected to foresee particular state injustices, they cannot be held responsible for them. Tis exemption may apply, for instance, to injustices conducted under cover of state secrecy.46 It may also apply, at least initially, to policies whose efects are so complex that citizens cannot reasonably be expected to anticipate them. (At what point in time,   See Jubb 2014, 67.

46

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

216

Alex Zakaras

for example, did American citizens become responsible for the U.S.’s disproportionate role in causing global climate change? Surely not before the emergence of a broad scientifc consensus on the issue.) Tis exemption certainly applies to citizens whose lives are such that they cannot be expected to stay informed about politics at all—citizens who have to work several jobs, for example, just to meet their family’s basic needs.47 Several other related questions fall beyond the scope of this chapter. I  have said very little about how much citizens must participate to either avoid or excuse their complicity in injustice. Te answer depends, of course, on the extent of their complicity and the severity of the injustice.48 My own view is that, in most contemporary democracies, casting an informed ballot is not nearly enough. I have also said virtually nothing about the division of democratic labor. Can I excuse my complicity in one injustice by working harder to remedy another? In complex modern societies, the answer is surely yes. I need not work to remedy all of the injustices for which my government is responsible. I can excuse my complicity by doing my fair share of this work, and this share may involve focusing on a few injustices in particular, or even just one.49 Tere is an important diference, however, between this view and the claim that I can fulfll my civic duties simply by doing some good in my society. Jason Brennan has argued, for example, that citizens of modern liberal societies can fulfll the demands of civic virtue by making some contribution to the common good. In his view, such contributions can be entirely apolitical: we can make them by working as “artists, entrepreneurs, small business owners, venture capitalists, teachers, physicians . . . ” (Brennan 2011, 52). All of these vocations, he says, contribute to the “bundle of goods” that citizens receive from their societies (these goods, he clarifes, are not just political but also “economic, cultural, social”) (Ibid.). Political participation, then, is not obligatory; it is only one of many ways in which citizens can contribute. If my argument in this chapter is correct, Brennan’s argument is sound only in perfectly just societies that stand in no foreseeable danger of committing future injustices (in other words: it is sound nowhere). To see why, suppose that Noah answered the aggrieved neighboring villagers as follows: “I know that my town is harming you unjustly, and I know that I am complicit in this injustice. But my complicity is ofset 47   See Beerbohm 2012, 246–9. If these citizens participate in politics in a way that lends support to unjust candidates or policies, however, they become complicit in a diferent sense: they accrue participatory complicity. 48   Lepora and Goodin suggest a number of helpful criteria for assessing the extent of an agent’s complicity; see Lepora and Goodin 2013, 102–13. 49   Tis division of labor has the advantage of substantially reducing the epistemic barriers to responsible participation.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -



Complicity and Coercion

217

by all of the good that I bring into the world as a successful venture capitalist who has helped expand my town’s economy and create jobs. I have no further obligation to you.” Tis response is clearly inadequate and would be received as such by the villagers, for it does nothing to address their moral claims.50 To further sharpen this point, consider the example of a colonialist who presides over the oppression and exploitation of a community of foreigners, but who also creates great works of art that contribute immeasurably to humankind. Just as this cultivated colonialist cannot discharge his moral obligations through art—no matter how great his work or how vast its infuence—Noah cannot excuse his complicity in injustice by investing in the local economy (or by working as a dermatologist or a schoolteacher, etc.).51 He is obliged to take political action, specifcally. Finally, just as responsible participation is a means of mitigating or avoiding complicity, irresponsible participation will only deepen it.52 Citizens who agitate for unjust political causes will only add participatory complicity to their existing political complicity and so do themselves further moral damage. Tis is why I have been careful to argue that responsible participation, not participation as such, is the appropriate normative response to complicity in political injustice.53

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

References Beerbohm, E. (2012). In Our Name: Te Ethics of Democracy. Princeton, NJ: Princeton University Press. Benn, S. (1988). A Teory of Freedom. Cambridge, UK: Cambridge University Press. Brennan, J. (2011). Te Ethics of Voting. Princeton, NJ: Princeton University Press. 50   Even if some of the jobs he created had gone to the victims of his town’s injustice, this would do nothing to mitigate Noah’s complicity. Te victims are demanding, frst of all, that he stop treating them unjustly. Ofering some of them a job is no answer to this demand. 51   In fairness to Brennan, he does not argue that apolitical contributions to the common good can excuse complicity in political injustice. Rather, he supposes that citizens who abstain from political participation (in modern liberal states) are not complicit in their government’s injustices. Tese citizens, he suggests, can be fully virtuous by making strictly apolitical contributions to their society. 52   On this point, I agree with Brennan. Tere is nothing morally laudable about political participation as such. 53   I would like to thank Shmulik Nili, Randall Harp, Tyler Doggett, Eric Beerbohm, Patrick Neal, Bob Taylor, and Alan Wertheimer for their generous help with this chapter, as well as the participants in the 2016 Oxford Studies in Political Philosophy Workshop and the political theory workshops in Portland, Oregon and at Washington University in St. Louis.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

218

Alex Zakaras

Caney, S. (2008). “Global Distributive Justice and the State,” Political Studies 56, 3: 487–518. Cohen, L. J. (1989). “Belief and Acceptance,” Mind 98, 391: 367–89. Frankfurt, H. (1988 [1973]). “Coercion and Moral Responsibility,” in Te Importance of What We Care About, pp. 26–46. Cambridge, UK: Cambridge University Press. Goldman, A. (2002). “A Causal Responsibility Approach to Voting,” in Democracy, ed. David Estlund, pp. 267–86. Malden, MA: Blackwell. Hart, H. L. A. and A. M. Honore. (1959). Causation in the Law. Oxford, UK: Clarendon Press. Jubb, R. (2014). “Participation in and Responsibility for State Injustices,” Social Teory and Practice 40, 1: 51–72. Kutz, C. (2000). Complicity: Ethics and Law for a Collective Age. Cambridge, UK: Cambridge University Press. Lepora, C. and R. Goodin. (2013). On Complicity and Compromise. Oxford, UK: Oxford University Press. McMahan, J. (2009). Killing in War. Oxford, UK: Oxford University Press. Miller, D. (2008). “National Responsibility and Global Justice,” Critical Review of International Social and Political Philosophy 11, 4: 383–99. Nagel, T. (2005). “Te Problem of Global Justice,” Philosophy and Public Afairs 33, 2: 113–47. Parrish, J. (2009). “Collective Responsibility and the State,” International Teory 1, 1: 119–54. Pasternak, A. (2013). “Limiting States’ Corporate Responsibility,” Te Journal of Political Philosophy 21, 4: 361–81. Pogge, T. (2005). “Reply to Critics: Severe Poverty as a Violation of Negative Duties,” Ethics & International Afairs 19, 1: 55–83. Rawls, J. (1993). Political Liberalism. New York: Columbia University Press. Stilz, A. (2011). “Collective Responsibility and the State,” Te Journal of Political Philosophy 19, 2: 190–208. Toreau, H. D. (2001 [1848]). “Civil Disobedience,” in Henry David Toreau: Collected Essays and Poems, ed. Elizabeth Hall Witherell, pp. 203–24. New York: Library of America. Tuck, R. (2008). Free Riding. Cambridge, MA: Harvard University Press. Wellman, C. H. (2007). “Responsibility: Personal, Collective, Corporate,” in A  Companion to Contemporary Political Philosophy, vol. 2, ed. Robert Goodin, Philip Pettit, and Tomas Pogge, pp. 736–44. Malden, MA: Blackwell. Young, I. (2006). “Responsibility and Global Justice: A Social Connection Model,” Social Philosophy & Policy 23, 1: 102–30.

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Index accountability  62–4, 67, 202–3 action  31, 33–4, 36, 40n.14, 49, 72, 97, 102n.33, 127–9, 134, 142 and n.4, 143, 144 and n.10, 148–9, 152, 154, 157, 159, 164, 181, 193–5, 197–8, 204, 206–7, 209, 212, 215, 217 ability to take  148 and n.17, 149 and n.20, 150, 159 collective  183–4, 193 and n.6, 194–8, 204n.25, 205 freedom of  141 and n.4, 142 and nn.4 and 6–7, 143, 147–8, 153n.25, 154, 160, 163, 178–9, 183–5, 194, 197, 201, 209 see also liberty, to spend as one sees ft and rights, to act (free from interference) lack of  142, 152–5, 156 and nn.30 and 32, 165, 183–4, 204, 208–9 harmful 193 legitimate  143, 145–6, 149, 157 permissible  142, 144n.10, 159n.37 see also rights, to act (free from interference) possible  147, 148n.17, 149 and n.20 see also action, ability to take prevention of  146, 151, 153n.25, 154, 156 and n.32, 165 see also poverty and prison unfreedom of  142 and n.7, 143 and n.7, 144, 147–9, 152–3, 160, 163 unjust  60, 193n.6, 207, 211, 213, 217 see also state, the, actions of, unjust action-tokens  156 and n.32, 157, 165 action-types  156 and n.31 Adler, Matthew  186n.14 “adverse selection”  170 Anderson, Elizabeth  81 and nn.1–2, 82–4, 91, 98n.27, 99, 100 and

n.29, 101, 106, 110n.1, 111 and nn.3–4, 114n.10, 115n.12, 125 and n.27, 126 and nn.31–2, 127 and n.32, 128–9, 130 and n.35 Arneson, Richard  100 and n.29, 107n.37, 114n.10 association  63, 75, 206, 212 assurance  33, 37, 41, 46, 47 and n.23, 48 and n.25 Assurance Game, the  182n.12, 183n.12 Astier, Cristina  107n.37 attitudes  180–1, 182n.12, 209 authoritarian regimes  192, 203nn.23–4, 207, 209–10, 211 and nn.41–2 authority  16n.16, 20, 21 and nn.23–4, 28, 31 and n.5, 32 and n.5, 33–4, 39, 42n.18, 45–6, 48, 49 and n.28, 50, 54, 63, 70, 73, 94, 127, 134 and n.43 claims to  39, 198 of decisions see decisions, authority of democratic see democracy, authority of natural  44, 45 and n.20 political  29, 40n.16, 41, 50 sources of  16, 20n.20, 29, 32, 44 of the state  29, 35, 39, 45, 50, 198–200 Baker, Daniel  186n.14 Beerbohm, Eric  196, 203n.22, 204n.26, 208n.37, 211, 217n.53 Beitz, Charles  136n.44 beliefs  9n.9, 16n.16 benefts 175 accorded by institutions  54, 56, 58–9, 71, 74, 76 mutual  179–80, 185–6 receipt of  203–4, 205n.29 Bentham, Jeremy  151n.23, 161n.40 Berlin, Isaiah  147 and n.15 Besson, Samantha  77n.2

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

220

Index

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Blackburn, Marsha  167–8 Bou-Habib, Paul  186n.14 Bovens, Luc  186n.14 Brennan, Jason  217nn.51–2 Buchanan, Allen  119n.18 Burke, Edmund  89 and n.14 Bush, George W.  197 Caney, Simon  13n.12 Carlsmith, Joseph  165n.46 Carter, Ian  107n.37, 136n.44, 152, 162n.42, 165n.46 causation  196, 197 and n.12, 198, 201–2, 204 choice  42, 100, 177–8, 180–4, 186–7, 201 unilateral  32–3, 38–9, 41 of others see other people, choices of Christiano, Tomas  6n.3, 17–18, 66, 77 and n.2, 123n.24, 136n.44 Christiansen, Andreas  107n.37 citizens  13, 15–16, 18–19, 23, 36, 67, 86n.8, 101, 116, 167, 179 and n.10, 180–2, 184–5, 194, 199, 203nn.22 and 24, 209, 216 see also noncitizens claims against/in opposition to one another see individuals, claims against/in opposition to one another actions of  192 and n.1, 193–9, 203n.22, 205 and nn.28–30, 206, 210n.40, 211–17 as coerced accomplices 193–4, 203–8, 211 of democracies  192, 212 interests of  17, 67, 180 responsibilities of  192–9, 201–2, 206, 209, 213–15, 216 and n.49, 217 see also responsibility, for injustice and the state  192 and n.1, 193, 198, 199 and n.16, 200–2, 203 and nn.22 and 24, 204, 205 and n.28, 206, 208–15, 217 and n.51 support of  47, 193–201, 203 and n.22, 204–6, 208, 211–15

treatment of  16, 18 citizenship  20, 193, 205, 212n.43 civil disobedience  199, 201, 204 and n.27, 207, 212 Clayton, Matthew  22n.25, 26n.28 coercion  3, 5, 10, 16, 26, 53–6, 64, 69, 192, 199, 203 and n.22, 205–6, 207 and n.35, 208–11, 213 see also decisions, coercive and enforcement, coercive justifed/permissible  4, 145 lack of  77 power of see power, coercive presumption against  158n.34 reliance upon  69, 76 subjects of see subjects, of coercion threat of  193–4 coercive efort  199 Cohen, G.A.  85 and n.7, 90nn.15–16, 114n.11, 148–9, 151, 156n.30m, 159n.37, 178n.9 Cohen, Jerry  84 and n.6 Cohen, Jonathan  207n.34 Cohen, Joshua  183n.12, 186n.14 collectivity  13, 66, 74 see also decisions, collective and judgements, collective common good, the  21n.23, 89, 205n.28 see also goods, public community  114n.11, 177, 217 “community rating”  170–1 compliance  5, 36–7, 46, 54–5, 57, 66, 69–70, 74, 77, 201, 209 see also noncompliance complicity  192–4, 195 and nn.8–9, 196–202, 203 and n.22, 204, 205 and nn.28–9, 206–7, 208 and n.37, 209, 210 and n.40, 211–15, 216 and nn.47–8, 217 and n.50 see also culpability avoiding/attempting to avoid  199–200, 203 and n.24, 204 and n.27, 205 and n.28–30, 206–7, 209–13, 214 and n.45, 215, 217 and n.51 complicity claim, the  192, 215 strong version  192–3, 204, 215 weak version  192–3, 203–4

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Index 221

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

computers 42–4 conductivity 43–4 confict/disagreement  14–15, 16 and n.16, 17, 70, 98, 106, 163 and n.44, 174, 184 see also individuals, claims against/in opposition to one another resolution of  12, 106, 163n.44 conformity 53 consent 198 constitutions  6, 45, 60, 67 see also norms, constitutional constraint  30, 32–3, 40, 42, 45n.21, 46, 49, 97–8, 121, 131, 141n.4, 142, 147n.12, 148 and n.18, 149, 156, 183 see also egalitarian deliberative constraint, the and interference contracts  178, 183, 200 contractualism  13, 106, 112, 123n.25 cooperation  63, 133n.42, 179, 181, 186 coordination  56, 59, 61, 63, 73–4 Cordelli, Chiara  118n.17, 124n.26, 136n.44 correctness 13n.12 corruption  3, 6n.4 courts  61, 64–5, 74–6 Crisp, Roger  107n.37 Culp, Julian  77n.2 culpability  194, 196, 204 Dagger, Richard  51n.31 Damgaard Taysen, Jens  107n.37 “death spiral”  170 decision-making  5, 12, 23 see also decisions mistakes in  11, 23n.26, 25 procedures  4–6, 11, 17, 23, 25 democratic  3–6, 7 and n.7, 8 and n.8, 9, 10 and n.10, 11–13, 15–17, 20, 25 inclusive  14, 16 legitimate  10, 13 public 179 decision-theory  172, 186 “rank dependent”  186–8 decisions authority of  5

autonomous 22 coercive 4–5 collective  12–13, 19 content of  4–6, 17, 25 just  4, 15, 22 unjust  20, 22–3 gravely 5–7 democratic  15–16, 19, 21, 23 see also decision-making, procedures, democratic enforceability of  4–6, 11, 16, 21n.23 see also enforceability enforcement of see enforcement, of decisions infuencing 121 insurance 171 legitimacy of  4–5, 7, 9, 14, 16, 23–5 see also enforceability, permissible making of see decision-making political  23–5, 208n.36 provenance of  4–6 see also decision-making, procedures public  14n.13, 39–40 reversing 23n.26 status of  5 deliberation 121–3 demands 7 see also under justice and legitimacy democracies  18, 67–8, 184, 192, 198, 203, 212 see also democratic bodies democracy  3, 7n.7, 19–20, 23, 40, 45, 60, 66–9, 75–7, 198, 203–4, 216 authority of  21 and n.23, 45n.20 and decision-making see decision-making, procedures, democratic and equality see equality, democratic justifcation of  17 lack of  14 see also processes, non-democratic democratic bodies  11, 22, 69 deontology 92–3 Dresden 202 desire-dependence  147 and n.12

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

222

Index

desires  147–8, 150 and n.21 see also citizens, desires of directives 28 discrimination 117n.14 see also non-discrimination distribution  81n.2, 83, 85, 87 and n.10, 94–6, 98–9, 105–6, 110–11, 112 and n.5, 113–16, 118, 121, 123–5, 127–8, 130–1, 133, 135, 178n.9 see also justice, distributive and demands, distributive and egalitarianism, distributive and freedom, external, distribution of and redistribution and resources, distribution of equal/equality of  82n.4, 83, 84 and nn.5–6, 85 and n.7, 86 and n.9, 87–8, 95–6, 105 fair  122, 124 of goods/resources  10, 83–4, 87, 88 and n.11, 91, 110–11, 112–13, 116, 118, 123n.25, 125, 135, 171 prioritarian 111 sufcientarian  111, 117 unequal/inequality of  86n.8, 87, 91, 94, 101, 103, 112, 126–7, 164n.45 Doggett, Tyler  217n.53 domination and exploitation  217 see also social relations, oppressive absence of  179–81, 185 duties and obligations  23, 28, 31, 39, 57, 64, 68–70, 77, 119, 130, 132n.39, 181n.11, 212–13, 214n.45, 217 Dworkin, Ronald  86n.8, 114n.11, 133n.41, 171n.2, 177–8, 183 and n.13 economics  112, 114n.11, 200 Edmundson, William  63 education 176 egalitarian deliberative constraint, the  94, 95 and n.22, 96–7, 98 and n.28, 99, 101–2, 121, 122n.24 see also deliberation, egalitarian egalitarianism  40, 81 and nn.1–2, 82 and n.4, 83, 86 and n.9, 89 and

n.13, 90, 92nn.18–19, 93, 95 and n.21, 96n.25, 97–101, 103, 118, 120–3, 125n.28, 127, 131, 132 and nn.38 and 40, 133, 134 and n.43, 135–6, 169, 184 see also inegalitarianism and nonegalitarianism brute luck  177–8 dispositional  82, 98, 102, 103n.34, 104–5, 127 distinctive  94n.21, 95 distributive  82 and n.3, 84 and nn.5–6, 86–8, 90, 92 and n.18, 95–6, 98–9, 100 and n.29, 101, 110, 111 and n.5, 113n.8, 114 and n.10, 115, 117, 118 and n.16, 119, 120 and n.21, 121–3, 131 see also distribution, equal/ equality of and justice, distributive ecumenical  82 and n.4, 102 and n.33, 103–4, 106 insufcient 172 liberal  168–9, 173, 185 luck  81 and n.1, 82 and n.3, 83, 84 and n.6, 85 and n.7, 86 and n.8, 87n.10, 90–1, 92 and n.19, 95n.23, 99, 100 and n.29, 101–2, 103 and n.34, 104–7, 111 and nn.3–4, 121–2, 126n.31 see also luckism non-luckist 101 non-relational 101 outcome  83, 101 pluralistic  169, 174 and n.7, 175–6 relational  81 and n.1, 82, 83–6, 88–91, 92 and nn.18–19, 93 and n.20, 94, 95nn.21 and 23, 96, 98 and n.27, 99–102, 103 and n.34, 106–7, 110 and n.1, 111n.3, 112, 114–15, 118n.16, 119, 121 and n.22, 126 and n.30, 131–3, 135 and n.43, 178n.9 see also justice, relational reductionist challenge to  82, 86–7, 88 and n.11, 90, 92, 94, 95 and n.21, 96, 99, 101, 102n.32, 103, 106

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Index 223 social  169, 176, 178–9, 181 and n.11, 183 and n.13, 184–5 views about  85, 91, 119 core diference(s) regarding  85, 102 transmission principle 104–5 social 91 Egypt 203n.24 emergencies 6n.3 enforceability see also enforcement permissible  4 and n.1, 5–6, 8 and n.8, 9–12, 14, 16n.16, 19, 21n.23, 23–4, 26 balancing view regarding  8–12, 17, 22–4 conventional view regarding  6–7, 9, 12, 14, 16, 19–22, 25 lack of  5–7, 16, 24 “recommendation view” of  11–12 sources of  20 enforcement  10 and n.10, 11–12, 39, 200, 209 see also enforceability agents of  29 see also state, the, agents of apparatus of  56 coercive 3–4 see also coercion of decisions  18–19, 22, 23 and n.26, 24 of laws  34, 200 of policies  22, 24 environment, the  176 epistocracy 44 Equal Consideration of Interests (ECI)  122n.24, 123n.24 equalisandum  83, 86, 93n.20, 104, 118n.16 equality  9, 16, 18–20, 35, 40–1, 44, 46–7, 50–1, 64, 81n.2, 83, 87, 89n.14, 95 and n.24, 96, 101–5, 119, 120 and n.21, 121, 124, 126–7, 132, 134–6, 168, 185, 188, 203 see also egalitarianism and inequality defenses of  81n.1 demands/requirements of  122 democratic  83, 125 and n.27 of distribution see distribution, equality of of fortune/luck  83

of freedom  28–9, 31–3, 39–40, 43–4, 157, 164 innate  29–30, 32, 41, 46 of opportunity  45nn.20–1, 178 relational  93, 127 social  20 and n.22, 21 and n.24, 22, 67, 85, 87, 117, 126, 180, 181 and n.11, 182, 184–5 see also status of welfare  87 see also welfare Estlund, David  4n.1, 6 and n.3, 10n.10, 13 and n.11, 15n.14, 26n.28, 107n.37, 136n.44 ethics  105, 192, 195 see also morality and  values and virtue evidence relativity of  9 and n.9, 23 exceptions 15 exoneration/absolution  196 and n.11, 197 and n.13, 202, 206, 207 and n.35, 208–10, 212, 215 exploitation standing  91 and n.17, 92 extensionality 93 Fabre, Cécile  51n.31 facts  10, 13, 16, 24 relativity of  9 and n.9, 23 fairness  82 and n.4, 102n.33, 104, 105 and n.36, 106, 117, 122, 168, 186, 200n.18 see also distribution, fair and  justice and unfairness Flanagan, Jessica  107n.37 Flinch Midtgaard, Søren  107n.37 Forst, Rainer  13n.12 Frankfurt, Harry  207 freedom  28, 30–1, 34–5, 40 and n.14, 43–4, 50–1, 141 and n.2, 201, 203, 209–10 see also action, freedom of and equality, of freedom and independence and liberty equal see equality, of freedom external  30 and n.2, 31–3, 36, 42–3, 46, 49 distribution of  44, 50 innate 31–3 unequal 46

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

224

Index

futures 198n.15 potential  173–4, 201

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

gender 10 Gheaus, Anca  107n.37, 118n.16 Gibbard, Allan  171n.2, 172n.5 good, the/goodness  131–2, 133 and n.42, 134n.43, 135, 160, 217 Goodin, Robert  195 and n.9, 216n.48 goods  83, 88, 91, 93n.19, 110, 112, 125, 135, 179–80 see also common good, the and money and resources access to  119, 186 claims to  112, 174 distribution of see distribution, of goods nonrelational  112–13, 116, 118, 132, 133 and n.42, 135–6 possession of  90 public  169, 178–80, 181n.11, 183–5, 205n.28 relational 118n.16 social  90–2, 112, 180, 181 and n.11, 183 and n.13, 185–6 supply/provision of  56, 118, 184 vital 116 Gore, Al  196n.11 grievances 23–4 guarantee problem, the  36, 45–7 harm  162n.43, 193 see also action, harmful moralization of  162n.43 Harp, Randall  217n.53 Hart, H.L.A.  48, 199n.16 health  169, 171–3, 184–8 poor 179 health care/services  170, 175, 183n.13, 185 equality of  186–8 spending on  168–70, 175, 177 equal 168 health insurance  179–80, 183n.13, 184–8 see also risks, to health decisions about see decisions, insurance fnancing of  167–8 free market  169–70, 178, 183n.13

mandatory  167–72, 174–5, 177, 180, 181n.11, 183n.13, 184–5 content of  168 justifcation of  168–9 minimum 168 see also Representative Prudent Individual Test (RPIT) packages  173–4, 177, 181, 185 extra large  175 large  172–8, 180–1, 182 and n.12, 183–4 medium  172–4, 176–8, 180–1, 182 and n.12, 183n.13, 184 small  172–4, 176, 181, 184 Hohfeld, Wesley Newcomb  31, 146, 152, 155n.29 Holtung, Nils  107n.37 Hooker, Brad  107n.37 Howard, Dana  186n.14 human beings  29, 36, 41, 63, 119, 200, 217 rights of see rights, human humanitarianism  118–19, 130n.35, 168, 181n.11 Hume, David  98n.27, 105 illegitimacy  26, 159, 210, 211n.41 grave 9 see also injustice, grave impartiality 62 inclusivity  12, 14, 16 income 169 see also poor, the independence 31 indeterminacy  31, 32n.5, 33 and n.8, 49–50 indignity  180, 185 indoctrination 147 individuals/agents  12–13, 16, 33n.9, 34–9, 41–2, 46–7, 65, 73, 75, 84n.5, 87, 93n.20, 94, 97, 110, 112, 126n.32, 128–30, 132, 134 and n.43, 135, 146–9, 152, 153 and n.25, 154, 171, 176, 178, 183, 185, 187–8, 193, 195, 197n.13, 201, 202n.20 see also citizens and other people actions of see action claims against/in opposition to one another  11–12, 17, 98, 112, 174

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Index 225 freedom of see action, freedom of and liberty and institutions  56, 58, 66–7, 71, 75–7 interference with see interference lives of  131–5, 176, 179–80, 184–5 relations between  40 and n.16, 67, 85, 86 and n.8, 88 and n.12, 89 and n.13, 89 and n.13, 90–2, 93n.20, 94, 95 and n.22, 96–8, 101–3, 105 and n.36, 106, 112, 120n.20, 123n.25, 127, 132, 133 and n.41, 134 see also social relations rights of see rights, of individuals situations facing  169 and the state  47, 67–9, 197–9, 201, 211 “Individual Diference Principle”, the  201, 202 and n.20, 203, 205n.32, 214 inegalitarianism  84, 96n.25, 132, 182 inequality  28, 47n.22, 85n.7, 91–2, 94, 126–7, 174, 177–8, 183, 188 see also equality choice-based 177–8 distributive see distribution, unequal/ inequality of economic  114n.11, 117 justifcation of  177, 183 less/lessening  87, 175–6, 184–5 relational  126n.30, 127 social  21–2, 92 unjust/unfair  86n.9, 91, 169, 171, 173, 177 infuence  114n.11, 121, 214, 217 injustice  6, 18, 23, 25–6, 50, 60, 74, 84n.6, 86n.9, 91, 102, 116–17, 126, 128–30, 192, 194–7, 199, 201, 204–9, 210n.40, 211–16, 217 and nn.50–1 see also decisions, unjust complicity in see complicity of decisions see decisions, content of, unjust foreseeable  195–6, 201 grave  3, 6 and n.4, 7 and n.5, 8–10, 12, 15–16, 21n.23, 24, 64,

209 see also decisions, content of, unjust, gravely and illegitimacy, grave avoidance of  9, 209 of laws see laws, unjust obvious 15 of outcomes see outcomes, unjust perpetration of  4, 192, 194, 198–9, 201, 204n.25, 206, 217 see also policies, unjust of policies see policies, unjust responsibility for see responsibility, for injustice slight  6, 18 social  3, 20, 85 structural 196 institutions  28, 34–5, 38, 40, 53 and n.1, 54–63, 65–77, 112, 128, 132n.39, 193 benefts of see benefts, accorded by institutions functioning of  55–7, 62, 69, 74 legitimacy of see legitimacy, institutional support for  53–60, 62–3, 65–6, 68–70, 72–3, 77 instrumentalism 105 insulation problem, the  36–8, 41 intention 196 interests  67, 95n.23, 102, 121, 132, 187–8 advancement of  62, 95n.23 best 57 diferent  174, 187–8 equality of see Equal Consideration of Interests (ECI) private  38, 57, 187–8 see also self-interest interference  141 and n.4, 143n.8, 144, 146, 149, 153 and n.25, 157, 159, 165 see also poverty and prison justifed  144–6, 151–4, 158n.35, 160, 162 overcoming 149n.20 unjustifed  144–6, 150n.22, 151, 159, 161n.39 wrongness of  157, 160

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

226

Index

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

international order, the  192 interpretation  32, 41 intervention  170, 175, 185, 197 Jensen, Kristian  107n.37 Jubb, Robert  195n.8, 198, 203n.23, 209 judgements  10n.10, 16, 26, 31 and n.4, 39, 42–4, 55, 164, 208, 210 burdens of  50 collective 13–14 combined 14 confdence in  14–15 correctness of  10–11, 24 inclusive 14 independence of  11 individual/private  17, 34–7, 38 and n.13, 41–4, 48 instrumental 32 about justice  9–10, 12, 15, 19, 26 mistaken  19, 25 about legitimacy  15, 26, 59 making of  13 privileged  32, 41–2, 44–5, 48–50 public 36 social 14n.13 superior 45 judges  42, 49n.27, 64, 164 justice  6, 9, 10n.10, 11–12, 14, 19, 21n.24, 24, 26, 36, 42, 55, 59, 64, 82, 92n.19, 103, 106, 110, 113–14, 119, 126 and nn.31–2, 127 and n.32, 128–9, 130 and n.35, 131, 133 and n.41, 134–6,  199 access to  47n.22 advancement/promotion of  4, 8, 20 associational 112 of decisions  5 see also decisions, content of, just demands/requirements of  10–11, 13, 14n.13, 15, 83–4, 112, 119n.18 distributive  110–17, 118 and n.16, 119, 120 and n.21, 121–5, 126 and n.29, 127 and n.32, 128–9, 131, 135 challenges to  110–12, 113 and n.9, 114–16, 118n.16, 124–5, 128–9, 133 Core Distributive Tesis, the  113–14

relational exclusivism  113, 115 and n.12, 125–6, 128–9, 131, 133–5 distributive exclusivism 114 egalitarian  81, 83–4, 98n.27, 100, 102–4, 106, 116, 124, 171 evading 46 global  7n.6, 192, 196, 206 imperfect 9 judgements about see judgements, about justice loss of  23 political  5, 25 and n.27 principles of  13, 17 “procedural” 5 reasons of  128, 129 and n.33, 130–1, 171, 177, 184 relational  110–16, 117 and n.15, 118–19, 125–7, 131 see also justice, distributive, challenges to, relational exclusivism social  5, 7–11, 13, 15, 17–19, 20 and n.22, 21–2, 24, 25 and n.27, 26 see also society, just delivery of  4 spheres of  81n.2 “substantive” 5 understandings/interpretations of  4 views about/accounts of  19, 26, 70, 81n.2, 102n.33, 106, 110, 111 and nn.2 and 5, 112–13, 115–19, 122n.23, 124, 128, 131, 192, 199 see also judgements, about justice “ideal theory of ”  24–5 Kant, Immanuel  31, 34–5 Kantianism  6n.3, 29–32, 40 and n.16, 42n.18, 43–4, 48n.25, 163–4 see also state, the, Kantian Kern, Adam  51n.31 Kirby, Nikolas  51n.31 Kis, Janos  129n.34, 136n.44 knowledge  21, 29, 44, 133, 172n.3, 194, 197 Kolodny, Niko  6, 16, 20 and n.21, 38n.13, 40, 51n.31, 186n.14 Kramer, Matthew  148n.18

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Index 227

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Kurtulmus, Ahmet Faik  171n.2 Kutz, Christopher  195 and n.8, 201, 202 and n.20, 203 and n.21, 204n.25 Complicity: Ethics and Law for a Collective Age 195 Lægaard, Sune  107n.37 lawfulness  61–2, 64–5, 67 laws  30, 31n.5, 33–6, 46, 48–9, 61–9, 72, 74, 199 and n.16, 200, 205n.29, 209 see also courts and legislation abidance by  53, 61, 63–6, 68–9, 74–5, 200 application of  28, 31–2, 34, 36, 38, 42, 48 see also enforcement, of laws agreement upon  48 coercive 10 content of  32n.5, 36, 69 just 35–6 legitimacy of  6, 8, 10 making/promulgation of  31, 34–5, 42, 44, 48–50, 66 permissible enforceability of  6, 24, 159n.37 unjust  6 and n.4 legality  61, 64 legislation  29, 36, 44 see also laws pieces of Afordable Care Act (2013)  167–8 Emergency Medical Treatment and Labour Act  183n.13 Fugitive Slave Law (1850)  199 legislators  42, 44, 49n.27 legitimacy  4–6, 9, 16–17, 21n.24, 22, 25 and n.27, 26, 54–5, 61–73, 76–7, 124, 193, 206, 207 and n.34, 208, 215 see also illegitimacy of actions see action, legitimate advancement/promotion of  4, 8, 20 assessment of see legitimacy assessments of authority see authority, legitimacy of corruption of  6n.4, 7 creation of  71–2 criteria for  67, 72, 74–7

of decisions see decisions, legitimacy of defnition of  4–5, 72, 77 demands/requirements of  15, 71 democratic  5–9, 14, 24–5, 45n.20, 66 see also decision-making, procedures, democratic disagreements about  15, 69–71, 210 discourse of  55 exclusive 58 function of  54, 61 ideal  22, 24–5 implausibility of  22 institutional  53–5, 56–62, 65, 67–77 Claim-Right view of  68–70, 75–7 Metacoordination view of  53–8, 60, 62, 66, 68–9, 71–5 Razian view of  68, 70–3, 75 Weak view of  76–7 judgements about see judgements, about legitimacy of laws see laws, legitimacy of loss of  23 non-ideal 25–6 normative  58, 63, 74, 76 of outcomes see outcomes, legitimacy of outweighing of  9 of policies see legitimacy of political 13n.12 of procedures see procedures, legitimacy of sociological  58, 63, 74, 76 of states see state, the, legitimacy of theories of  25, 70–1, 74–5 value of  22 legitimacy assessments  53, 58–9, 62, 69, 71, 73, 75–6 legitimation 75 Lepora, Chiara  195 and n.9, 216n.48 Lever, Annabelle  26n.28 liabilities 28 liberalism  168, 216 liberty  76, 141 and n.2, 142–7, 148 and n.17, 150, 151n.23, 152, 155–6, 157 and n.33, 158–60, 161 and nn.39–41, 161n.40, 162 and n.42, 163 and n.44, 164–5, 178–9, 201 see also action, freedom of and freedom

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

228

Index

liberty (cont.) conceptions of  141 and nn.3–4, 142–3, 146–7, 151 and n.24, 158 and n.36, 164 moralized of  141 and nn.3–4, 142–6, 147n.12, 149, 151nn.23–4, 152–6, 158–9, 161, 162n.43, 163 and n.43, 164–5 negative  141 and n.3, 142n.5, 143n.7, 158, 163 nonmoralized  141 and n.4, 142n.4, 157, 162–4 normative  154, 157, 158 and n.35, 159, 161 and n.39, 162 and n.43 positive  141 and n.3 rights-based  143, 158n.36, 164 degrees of  147, 154n.26 increasing  147n.14, 148 infringement of  141, 143–6, 150–2, 153 and n.25, 154 and n.26, 155–7, 158 and n.35, 159, 161–3, 165, 177, 183–4 see also action, prevention of and constraint and interference and  poverty and prison physical 154 and rights  161 see also rights, to act (free from interference) and rights, violation of signifcance of  160 to spend as one sees ft  167, 178 type-1 154–5 type-2  154, 155 and n.28, 157 type-3  154, 155 and n.28, 157 y-parameter  144, 146 z-parameter of  142 and nn.4–6, 143–4, 146–7, 148 and nn.17–18, 149–50, 153–4, 155 and n.27 Libya 203n.24 license  141–2, 143 and n.8, 144, 146, 155 Lippert-Rasmussen, Kasper  186n.14 Lo Re, Stefano  33n.7, 51n.31 Locke, John  31 luck bad  174, 185 bad brute  111n.4, 168–9

brute  176, 177 and n.8, 178, 184 see also egalitarianism, brute luck “option”  177 and n.8 luckism  100n.29, 101 Mack, Eric  153n.25 Macleod, Colin  186n.14 Mazor, Joseph  186n.14 Medicare 167 Menzel, Paul  171n.2 Mexico 200–1 Miklosi, Zoltan  107n.37 Miller, David  81n.2, 87, 93–4 mistakes 11 see also decision-making, mistakes in Møller Lyngby Pedersen, Viki  107n.37 money  156n.30, 175, 194, 200 see also health care, fnancing of and income and resources, personal and taxation moral permissibility  104, 142, 209 see also action, permissible moralism 122 morality  9, 17, 30–1, 34, 53–9, 62, 64–9, 71–3, 77, 94, 101, 103–5, 112–13, 118–19, 124, 128, 129 and n.33, 130, 132 and n.39, 134 and n.43, 135 and n.43, 136, 145, 147n.14, 159, 161n.40, 162, 177, 194, 195n.9, 197n.12, 198–200, 202n.20, 205, 208, 211, 213–15, 217 and n.52 see also ethics and moralization and power, moral and principles, moral and rights, moral and values and virtue moralization  141–2, 163 and n.43 of harm see harm, moralization of of liberty see liberty, moralization of Munoz-Dardé, Véronique  186n.14 Nader, Ralf  196n.11 Nagel, Tomas  112n.6, 131n.37, 192, 207, 208 and nn.37–8, 209 “Te Problem of Global Justice”  206

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Index 229

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Neal, Patrick  217n.53 necessity 201 negligence 196 Neumann-Morgenstern axioms  172n.5 Newey, Charlotte  107n.37 Ng, Geraldine  107n.37 Nielsen, Lasse  107n.37 Nili, Shmulik  217n.53 noncitizens  179n.10, 217 noncompliance 57 non-discrimination  66, 117 nonegalitarianism 89 non-social relations see individuals, relations between “normative disengagement”  210–11 “normative engagement”  206–7, 208n.38, 209–13 norms  41, 45, 49 constitutional 48 institutional  28–9, 44 see also institutions Nozick, Robert  161 and n.41, 178n.9 obedience  200, 203–4, 213, 214n.45 see also  civil disobedience obstacles 150n.21 Olsaretti, Serena  107n.37, 136n.44 Olsen, Tore Vincent  107n.37 opportunity  95n.23, 158 and n.36, 159n.37, 160, 162, 178 equality of see equality, of opportunity other people  18, 21–2, 30, 33n.9, 36, 42, 47–9, 97–8, 133n.42, 175, 179–80, 182, 187, 194, 214 see also individuals choices of  30 expectations placed upon  46 interference with see interference preferences of  13 treatment of  19, 49, 102, 105, 131, 180 views of  13–15, 17, 19, 31n.4, 45 outcomes  14, 24, 26, 101, 148, 187, 202n.20, 204 determination of  122n.24 equality of  83 see also egalitarianism, outcome inequality of  177 just 24

legitimacy of  4 lack of  7 permissible enforceability of  8n.8 unjust  6–7, 16, 196, 202n.20 Otsuka, Michael  51n.31, 107n.37, 174n.6, 186n.14 Pallikkathayil, Japa  33n.9 Parft, Derek  203n.21 Parr, Tom  107n.37, 136n.44, 186n.14 particulars 33 Pasternak, Avia  195n.8, 198, 205n.31, 210n.40 Peter, Fabienne  16n.16 Pettit, Philip  5, 6n.3, 10, 25 and n.27 pluralism  81n.2, 82n.4, 84n.6, 132, 169 see also egalitarianism, pluralist Pogge, Tomas  192 policies  4, 21, 172–4, 174n.7, 186–7 enforcement of  3–4, 22 legitimacy of  6–8 permissible enforceability of  6, 24 unjust  4, 6, 22, 192 and n.1, 193, 212, 213 and n.44 political organisations  194–7, 207–8, 212, 217 political participation  192, 194, 195 and n.8, 197, 199–200, 202, 204 and n.25, 208 and n.38, 210n.40, 212, 214–15, 216 and nn.47 and 49, 217 and n.52 lack of  192–3, 215 political passivity see political participation, lack of politicians 193 politics  24, 25 and n.27, 26, 50, 60, 67, 112, 114n.11, 158, 165, 179, 193–7, 202–4, 208–9, 214, 217 and n.51 see also authority, political and decisions, political and justice, political and rights, political poor, the  171 poverty  150, 156 and n.32, 157, 161n.39, 165 see also poor, the 171

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

230 power  20, 28–9, 35, 46, 51, 58, 60, 71–2, 203 and n.24 coercive  29, 55, 76, 203–4, 207, 208n.36, 210 Hohfeldian 155n.29 wielding of  21, 62 see also state, the, agents of, power of of the state see state, the, power of predictability  62, 64 preferences  13, 171, 172n.5, 178 of other people see other people, preferences of principles  31–3, 55 prison/imprisonment  150–2, 153 and n.25, 154 and n.27, 155 and n.29, 156–7, 160, 161n.39, 209 justifed see interference, justifed Prisoner’s Dilemma, the  182n.12, 183n.12 privileges 143n.8 procedures  5–7, 10–11, 13–14, 23, 25–6, 34, 60 see also processes for decision-making see decision-making, procedures democratic  20, 40 see also decision-making, procedures, democratic legitimacy of  4–5, 6 and n.4, 7–10, 12, 16–17, 24–6 processes  13–14, 19, 45, 50, 67 property  29, 31, 34, 200n.17, 205n.28 see also rights, property proportionality 4n.1 protection 38 public, the  18, 34–6, 62, 179, 185, 199 and n.16 see also decision-making, public and decisions, public and goods, public and will, public will of see will, public public/ofcial positions  28–9, 38, 41, 44–5, 48, 50–1 see also state, the, agents of public order  61, 63–4 punishment 151 Quong, Jonathan  6n.3

Index racism  194, 196, 207–8 rationality  134 and n.43, 178 Rawls, John  6 and nn.3–4, 7n.6, 8, 13, 50, 100, 111n.3, 122n.23, 128, 132n.39, 179, 203 A Teory of Justice 100 Raz, Joseph  49n.28, 53, 70–2 see also legitimacy, institutional, Razian view of reason  35, 58, 62, 66, 72–3 reasoning  13, 14n.13, 120n.22, 214 reasons  10, 55, 57–8, 68–9, 71–2, 102, 119, 124, 198 see also justice, reasons of reciprocity  95n.21, 119n.18 recommendations  9, 10n.10 reconciliation 50n.30 redistribution 179 Representative Prudent Individual Test (RPIT)  168–9, 171 and n.2, 173, 186–8 objections to  172 and n.3, 174–6, 186–8 Diferential Prospects Case  172–5 Identical Prospects Case  176, 178, 183 resources  112, 122n.24, 135, 156 and n.30, 159, 167 see also goods distribution of see distribution, of goods/resources respect  14, 16–17, 19–20, 56, 65, 76, 81n.2, 86n.8, 95n.21, 159–60, 178, 179n.10 see also self-respect responsibility  83, 85, 93n.20, 94, 99–103, 105 and n.36, 106, 111, 128, 141n.2, 177n.8, 193–6, 197 n.12, 198, 201–2, 209, 212, 214–15, 216n.49, 217 see also citizens, responsibilities of and culpability for injustice  192–8, 201, 205–6, 208–9, 212–14 shared  195–6, 205–6 “task”  198 and n.15 retributivism 151 rights  16, 28, 30, 41, 54, 66, 94, 141 and n.1, 142, 143 and n.8, 145, 149, 158 and n.36,

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Index 231 162, 164 see also liberty, conceptions of, rights-based acquired  28–32, 33 and n.9, 34–5, 37–8, 45–6, 48–9 to act (free from interference)  141, 143 and n.8, 144 and n.10, 145–6, 149, 150 and n.22, 151–2, 153n.25, 154, 155n.29, 159, 163–4 bodily integrity  30n.2 claim  143n.8, 146 conditional 152 contractual  28, 30 human  3, 60–1, 67 individual 67 innate  29–32, 35, 37, 39, 42, 50 legal 65 moral  28, 141n.1 negative 163 political 7 positive 163 property  28, 30, 161 violation of  3, 39, 46, 59, 61, 143–5, 151–2, 153n.25, 154 and n.27, 158n.35, 161n.39, 165 see also interference Ripstein, Arthur  33n.9, 34–5, 37, 38 and n.12, 39, 45 Force and Freedom 37 risks  58, 71–2, 77, 187 equality of  169, 176 to health  168–71 high 168 low 185 Rousseau, Jean-Jacques  42n.18 rules  11, 30, 53–4, 56–7, 62–6, 68, 71, 73, 207 see also institutions breaking of  207 following of  29, 48, 51 “rules of recognition”  48 Sangiovanni, Andrea  77n.2 Satz, Debra  186n.14 Scanlon, Tomas  117–18, 123n.25 Schefer, Samuel  83–5, 88, 90n.16, 94 and n.21, 95 and n.21, 96–7, 98 and n.28, 99, 100 and n.29, 101–2, 106,

110n.1, 111n.3, 114n.11, 119, 120 and nn.21–2, 121 and n.22, 122–3, 124 and n.26, 125 Schemmel, Christian  116–17 security 66–7 self-conceptions 195 self-interest  54–5, 57, 180 self-respect  179–80, 185 Sen, Amartya  14n.13, 66 Simmons, John  70 simpliciter 142n.6 Simpson, Tomas  51n.31 Sinclair, Tomas  6n.3, 107n.37 slavery  199–202, 204 Slavny, Adam  107n.37 Smith, Adam  180 Sobel, David  107n.37 “social choice theory”  14n.13 social interaction  20n.21, 88n.12 social practice  53, 65, 70, 73, 77 social relationalism  89 social relations  81n.2, 83, 84, 85 and n.7, 86 and n.8, 87, 88n.12, 89 and nn.13–14, 90–4, 96–9, 101, 110–11, 112 and n.5, 113 and nn.7–8, 114, 117–18, 120, 125 and n.28, 126 and n.28, 127, 131, 132 and n.39, 134 and n.43, 135, 169 see also individuals, relations between oppressive  84, 217 see also domination and exploitation quality of  115, 118 social standing see status society  18, 25, 40, 64, 76, 81n.2, 84 and n.5, 85–90, 101, 120n.20, 130, 157, 171–2, 176, 178–80, 181 and n.11, 183–6, 208n.38, 216, 217 and n.51 see also justice, social cohesion of  180, 186 of equals  121, 134, 181 see also status, equality of hierarchy of  85n.7, 89 and n.14, 93–4, 97, 103n.34 just 110 order of  64 structures of  112, 115

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

232

Index

Socrates 205 solidarity 171 Sommer Hansen, Rasmus  107n.37 Stanczyk, Lucas  136n.44 “Standard Gamble”, the  172n.5 state, the  3, 34, 38, 40, 44, 47 and n.22, 60–2, 66–9, 71, 116, 192, 199 and n.16, 200, 201 and n.19, 203 and nn.22–4, 211n.41, 212n.43, 213, 215 see also states actions of  3–4, 46, 60, 67, 167–8, 183n.13, 186, 192 and n.1, 194, 197, 199, 201–2, 204–6 unjust  192–3, 197, 199–201, 204–6, 208–9, 210n.40, 211–15, 217 and n.51 agents of  28–9, 34–5, 37, 38 and n.12, 39, 41, 48, 74, 199–200, 210n.40 see also enforcement, agents of and public/ofcial positions power of  29, 39, 51, 200 subjugation to  29, 200 authority of see authority, of the state and its citizens  47, 93n.20, 116, 167, 178, 192–3, 197–202, 203 and n.22, 204, 205 and n.28, 206–13, 214 and n.45, 215, 217 see also citizens, and the state constitution of  34–5, 42, 51 functioning of  46, 61, 67, 69 Kantian account of  29–30, 34–42, 47 and n.23, 48n.25, 49n.27, 50–1 legitimacy of  64–5, 74, 76, 206, 207 and n.34, 208 and n.36, 210, 211 and n.41, 212, 214n.45, 215 nature of  28, 30, 35, 50, 199–200, 206 necessity of  36 norms of  29–30 opposition to  200 and n.17, 203 and n.24, 205 and n.28, 206–7, 208n.36, 209–15 see also civil disobedience policies of see policies power of  29, 35, 39, 41, 46–7, 199, 203n.22, 209–10

support for  193–4, 196–7, 199–201, 203–6, 208, 212–15 state of nature, the  29–33, 35–6, 38–45, 48, 50 ideal  36–7, 41, 47–8 status  40 and n.16, 87, 88n.11, 93n.20, 94, 135, 169, 179 equality of  67, 87, 113n.7, 117, 134n.43, 135 see also society, of equals Steel, Sandy  51n.31 Steenivasan, Gopal  186n.14 Stemplowska, Zofa  107n.37, 118n.16, 133n.42 Stilz, Anna  31, 33n.9, 198 Stockholm Syndrome  206n.33 Stratton-Lake, Philip  107n.37 subjects  23, 209 of coercion  5, 209 subordination/subjugation  16, 29, 49n.28, 193 see also state, the, agents of, subjugation to acceptance of  193, 200, 206 and n.33, 207 and n.34, 208–9, 211n.41 avoidance of  19, 209 sufcientarianism 89 see also distribution, sufcientarian sympathy  180, 205n.28 Tadros, Victor  19n.17, 26n.28, 107n.37 Tasioulas, John  70 taxation  167, 193, 200, 201 and n.19, 203, 204 and n.27, 205 and n.28, 213 Taylor, Bob  217n.53 Toreau, Henry David  198–202, 203 and n.22, 204 and n.27, 205 and n.28, 206, 209–10 Tomlin, Patrick  85, 107n.37 trade-ofs 180 interpersonal  169, 173 intrapersonal 169 Tunisia 203n.24 unfairness  105, 168, 177 United Kingdom, the  25–6 United States, the  25, 183n.13, 197, 199–201, 206, 213

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -

Index universality  40, 46, 164 and n.45, 185 utilitarianism 122 Valentini, Laura  6n.3, 10, 17, 25 Vallentyne, Peter  82, 107n.37, 178, 186n.14 value  131, 132n.40, 159–60, 178, 184 values  35, 55, 66, 84n.6, 102, 120n.21 Van der Vossen, Bas  77n.2 Viehof, Daniel  6 and n.3, 20 and nn.20 and 22, 21 and n.23, 51n.31 violence  29, 60 virtue  64–5, 69, 75, 89n.14, 126n.32, 216 see also morality voluntariness  145, 193 Voorhoeve, Alex  107n.37, 136n.44

Walzer, Michael  81n.2, 92n.19 war  201–2, 205 Waxman, Dan  165n.46 welfare  85n.7, 87, 111n.4, 113n.8, 122n.24, 175 well-being  89n.13, 132 and n.40, 133 and n.41, 134, 172 and n.5, 173–6, 178, 180–7 Wellman, Christopher  77n.2 Wendler, David  186n.14 Wertheimer, Alan  217n.53 will  31, 33n.9, 34 public 34–6 Williams, Andrew  4n.1, 22n.25, 26n.28, 107n.37, 186n.14 Wolf, Jonathan  81n.2, 110n.1, 117n.15 Wollheim, Richard  147n.15 Zubof, Arnold  148n.19

Copyright © 2018. Oxford University Press USA - OSO. All rights reserved.

Waldron, Jeremy  156n.32, 159n.36, 161, 163 Wall, Steve  107n.37

233

Oxford Studies in Political Philosophy Volume 4, edited by David Sobel, et al., Oxford University Press USA -