149 109 4MB
English Pages 256 [369] Year 2024
How Can We Be Equals?
How Can We Be Equals? Basic Equality: Its Meaning, Explanation, and Scope Edited by
Giacomo Floris and Nikolas Kirby
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 © Oxford University Press 2024 The moral rights of the authors have been asserted 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 Library of Congress Control Number: 2023944405 ISBN 9780192871480 DOI: 10.1093/oso/9780192871480.001.0001 Printed and bound in the UK by Clays Ltd, Elcograf S.p.A. 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.
Acknowledgements The idea for this book originated from a conference on moral equality at the 2020 MANCEPT Workshops in Political Theory. We would like to thank all the participants for the insightful discussion. We are very grateful to all the contributors to this volume for their enthusiasm in the project, cooperative spirit, and the time and effort they put into their contributions. Finally, we are also grateful to our Editor, Dominic Byatt, for his encouragement and interest in the volume, our Project Officer, Jade Dixon, for her kind support and assistance, and an anonymous reviewer for Oxford University Press for their perceptive and constructive comments.
Contents List of Contributors
Basic Equality: An Analytical Introduction Nikolas Kirby and Giacomo Floris 1. Introduction 2. Basic Equality: Key Theses 3. The Meaning of Basic Equality: History and Import of Claiming to Be Equals 3.1 The Multiple Meanings of Basic Equality 3.2 Overview of the Chapters
4. The Explanation of Basic Equality 4.1 The End of Ecumenicism: Why the Basis of Basic Equality Matters 4.2 Grounding and Justification 4.3 The Problem of the Explanation of Basic Equality 4.4 Overview of the Chapters
5. The Scope of Basic Equality 5.1 The Scope Problem: Who Is a Moral Equal? 5.2 Overview of the Chapters
PART I
xiii
1 1 2 4 4 7
11 11 12 14 16
22 22 25
M EA N IN G: HISTO RY AND CRITIQUE
1. On the Historical Emergence of Basic Human Equality Teresa M. Bejan 1. Introduction 2. High-Road Histories 3. Political Equality: Balance and Indifference 4. Human Equality: Aequales and Pares 5. Equality-as-Parity 6. Conclusion
2. Basic and Relational Equality Jeremy Waldron 1. A Question 2. An Inadequate Answer 3. My Aim in This Chapter 4. Are Basic and Relational Equality Propositions of Different Kinds? 5. This Is Not about Moral Realism 6. Factual Similarities and Differences
35 35 37 39 46 51 53
65 65 65 67 67 68 68
viii
Contents 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Supervenience Egalitarian Commitments without Supervenience? What Does Relational Equality Supervene Upon? Supervenience and Shapelessness A Summary So Far Basic Equality as a Normative Position Immediate Normative Implications of Basic Equality Basic Equality and Human Dignity Moral Status Doesnʼt Relational Equality Do Roughly the Same Work? Concept and Conception Differences of Emphasis as Between Basic and Relational Equality Basic Equality as a Ground for Relational Equality Conclusion
3. Should People Who Are Moral Equals Relate as Social Equals? Should People Who Are Not Moral Equals Relate as Social Unequals? Kasper Lippert-Rasmussen 1. Introduction 2. Moral and Social Equality 3. How Does Moral Equality Justify Social Equality? 4. Justifying Social Equality in the Absence of Moral Equality 5. Conclusion
4. Is There a Thing Called Moral Equality? (And Does It Matter if There Isnʼt?) Andrea Sangiovanni 1. Do Consequentialism and Non-Consequentialism Accept the Same Principle of Moral Equality? 2. Abandoning Moral Equality 3. Social Status Hierarchy and Animal Ethics 4. Conclusion
PART I I
69 69 70 71 71 72 72 73 74 75 75 76 77 78
81 81 84 89 95 100
103
103 110 119 121
J USTIF ICATIO N: G ROUNDING AND SCEPTICISM
5. Basic Equality and the Contexts of Opacity Respect Ian Carter 1. Introduction: Equality and Opacity 2. Levels of Basic Equality 3. The Substantive Grounds for Opacity Respect 4. Contexts of Practical Reason 5. Evaluation versus Evaluative Abstinence, Compatibilism versus Incompatibilism 6. The Compatibilist Account
127 127 130 132 135 137 138
Contents 7. The Incompatibilist Account 8. Conclusion
6. Equal Moral Status and the Collective Nature of Rationality Thomas Christiano 1. The Puzzle 2. Autonomy and Moral Status 3. Moral Status and Public Status 4. The Puzzle and Range Properties 5. Imperfect Rational Nature 6. The Collective Character of Rational Nature 7. An Intuitive Argument for Equality Despite Different Rational Capacities 8. A More Positive Theoretical Argument for Equality 9. The Contribution Argument for Inequality 10. The Problem with the Contribution Argument 11. The Argument for Equality from the Collective Character of Rationality and Status Individualism 12. Objection from Extreme Disability to the Argument for Equal Status 13. Conclusion
7. Basic Equality: Worth, Luck, and Weight Nikolas Kirby 1. Introduction 2. The Problem of Basic Equality 3. Moral Worth 4. Moral Luck 5. Moral Weight 6. Objections 6.1 6.2 6.3 6.4
Difference in moral options Determinism Lack of alternative options Those without agency
7. Conclusion
8. Equality and Moral Status: Challenges to Their Grounding Agnieszka Jaworska and Julie Tannenbaum 1. Types of Moral Equality 2. Degrees of Moral Status 3. The Grounds of Moral Status and Why They Matter 4. Mapping Degrees of Moral Status onto Their Grounds 5. Reasons to Doubt the Continuity View 6. Conclusion
ix 142 145
149 149 150 151 152 153 156 158 162 164 165 167 168 171
175 175 176 180 185 188 192 192 193 194 194
195
199 200 204 205 210 213 217
x
Contents
9. When Equality Needs no Justification Anne Phillips 1. Equality as Commitment and Claim 2. From Moral Equality to Moral Judgement 3. The Moral/Material Distinction
PART I II
221 224 229 232
SCOPE : I NCLUSI ON AND MARG INALIZATIO N
10. The Basis of Childrenʼs Moral Equality
241
Giacomo Floris 1. Introduction 2. Moral Equality, Respect, and Children 3. Moral Inferiority and Self-Respect 4. Moral Superiority and Friendship 5. Objections
241 242 246 250 254
5.1 The question-begging objection 5.2 The sidestepping objection
254 256
6. Conclusion
11. Basic Human Moral Equality Eva Feder Kittay 1. Introduction 2. The Challenge of Moral Individualism 3. Is Species Membership Morally Irrelevant? 4. The Importance of Empirical Knowledge for Moral Assessments of Other Human Beings 5. Cognitive Ability, Speciesism, and Moral Standing 6. Substituting Hierarchies 7. Grounding Moral Individualism in a Metaphysics 8. Intrinsic Properties Versus Relational Properties 9. What About ʻUsʼ is Morally Relevant? 9.1 Who is the ʻUsʼ?
12. First Among Equals George Sher 1. Moral Standing 2. Humans and Animals 3. Human and Animal Interests 4. Treating Moral Equals Differently 5. Treating Humans and Animals Differently
257
261 261 263 265 266 270 271 273 276 279 281
287 287 290 292 295 298
Contents
13. Basic Equality, Rational Agency Capacity, and Potentiality Richard Arneson 1. The RAC Account 2. RACʼs Appeal 3. A Controversial but Defensible Implication 4. A Devastating Objection 5. A Modified RAC Account Combines Potential and Developed Capacity 6. Discussion of Other Views
Index
xi
305 310 311 312 315 318 321
325
List of Contributors Richard Arneson has taught at the University of California, San Diego, since 1973. He attained the rank of ‘Distinguished Professor’ and from 2011 to 2023 was the holder of the Valtz Family Chair in Philosophy at that institution. As of July 2023, he shifted to Emeritus status. He writes on topics in political and moral philosophy, with an emphasis on egalitarian theories of justice and on the prospects for consequentialist moral theory. In 2022 his short book Prioritarianism was published by Cambridge University Press in its Cambridge Elements series. Teresa M. Bejan is Professor of Political Theory at the University of Oxford and a Fellow of Oriel College. Her work brings historical perspectives to bear on questions in contemporary political theory and philosophy. She is the author of Mere Civility: Disagreement and the Limits of Toleration (2017) and the forthcoming First Among Equals, both with Harvard University Press, as well as numerous peer-reviewed articles and book chapters. Ian Carter is Professor of Political Philosophy at the University of Pavia, Italy. His research interests include the concepts of freedom, rights, distributive equality, and basic equality. He is the author of A Measure of Freedom (Oxford University Press, 1999) and has published in journals such as Ethics, Economics and Philosophy, Journal of Political Philosophy, Oxford Studies in Political Philosophy, Social Philosophy and Policy, and the Australasian Journal
of Philosophy. He is currently working on a book on basic equality and opacity respect. Thomas Christiano is Professor of Philosophy and Head of the Philosophy Department at the University of Arizona. He is the author of The Constitution of Equality: Democratic Authority and Its Limits (Oxford: Oxford University Press, 2008) and The Rule of the Many (Boulder, CO: Westview Press, 1996) and many papers in moral and political philosophy. He is coeditor of Politics, Philosophy and Economics. He has been a Visiting Fellow at All Souls College, Oxford, and Laurence Rockefeller Fellow at the Center for Human Values at Princeton University. Giacomo Floris is British Academy Postdoctoral Fellow in Philosophy at the University of York. His main areas of research are in contemporary moral and political philosophy, with a particular interest in theories of moral status and basic equality, and theories of relational equality and distributive justice. His work has appeared in various journals, including Philosophical Studies, Journal of Ethics, European Journal of Political Theory, and Ethical Theory and Moral Practice. Agnieszka Jaworska is Professor of Philosophy at the University of California, Riverside. She has published in Ethics, Philosophy and Public Affairs, and Philosophy and Phenomenological Research, on topics including human agency, the concepts of valuing and
xiv
List of Contributors
caring, and moral status. She has collaborated with Julie Tannenbaum on a series of articles addressing moral status—focusing especially on underdeveloped and cognitively impaired humans. She is currently working on a book project entitled Ethical Dilemmas at the Margins of Agency. Nikolas Kirby is Lecturer in Politics at the University of Glasgow. His current research interests include basic equality, the philosophy of governance, and the ethics of trust. His work has appeared in various journals including, Journal of Politics, British Journal of Political Science, Journal of Ethics, Law and Philosophy and Journal of Social and Political Philosophy. He is the editor of What is Good Government? The Philosophy of Office, Institutions and Administration, forthcoming with Oxford University Press. Eva Feder Kittay is Distinguished Professor Emeritus at the Department of Philosophy, Stony Brook University/SUNY. Her latest book, Learning from My Daughter: The Value and Care of Disabled Minds, won the PROSE award. Love’s Labor: Equality, Dependency and Women (1999) is widely read and has been translated into several languages. She was the President of the Eastern APA, the recipient of many prizes and awards, including a Guggenheim and NEH fellowship, Leibowitz APA and Phi Beta Kappa Prize, Fullbright Scholar Inaugural Prize of IMEW, and Woman Philosopher of the Year. She has authored over 100 articles and book chapters, and authored and coedited books on metaphor, feminism, disability, and the ethics of care, including Metaphor (1987), The Subject of Care (2003), Women and Moral Theory (1987), the
Blackwell Guide to Feminist Philosophy (2007), and Cognitive Disability and Its Challenge to Moral Philosophy (2010). She has a daughter with very significant disabilities, whom she regards as a collaborator in writings on disability. Kasper Lippert-Rasmussen is Professor of Political Theory at University of Aarhus, director of the Center for the Experimental-Philosophical Study of Discrimination (CEPDISC), and Professor II in Philosophy at the Arctic University of Norway-UiT. Presently, he is working on discrimination and standing to blame. His book on the latter topic—The Beam and the Mote—is forthcoming with Oxford University Press. Anne Phillips is Professor Emerita in the Government Department of the London School of Economics, where she was also previously Director of LSE Gender Institute. She has written extensively on issues of equality, including in The Politics of Presence (1995), Which Equalities Matter? (1999), The Politics of the Human (2015), and Unconditional Equals (2021). Andrea Sangiovanni is Professor of Philosophy at King’s College London. From 2018–2020, he was Professor of Social and Political Theory at the European University Institute, Fiesole. He is the author of Humanity without Dignity: Moral Equality, Respect, and Human Rights (Harvard University Press, 2017) and Solidarity: Nature, Grounds, and Value (forthcoming, Manchester University Press). He is also the recipient of a five-year ERC Consolidator Grant, which has supported his research on solidarity. George Sher is Herbert S. Autrey Professor of Philosophy at Rice
List of Contributors University, where he has taught since 1991. Previously he taught at the University of Vermont. His research interests range over ethics, social and political philosophy, and moral psychology, and his books include Desert (1987), Beyond Neutrality: Perfectionism and Politics (1997), In Praise of Blame (2006), Who Knew: Responsibility without Awareness (2009), and Equality for Inegalitarians (2014). His most recent book, A Wild West of the Mind, was published in 2021. Julie Tannenbaum is Associate Professor of Philosophy at Pomona College. Her research is in both ethical theory and bioethics, and has been published in various journals, including Ethics, Bioethics, Hastings Center Report, Oxford
xv
Studies in Metaethics, Oxford Studies in Normative Ethics, Philosophical Studies, and The Stanford Encyclopedia of Philosophy. She has published multiple articles with Agnieszka Jaworska on the moral status of animals and humans. Jeremy Waldron is Professor of Law at NYU School of Law. He has written extensively on jurisprudence and political theory, including numerous books and articles on theories of rights, constitutionalism, the rule of law, democracy, property, torture, security, homelessness, and the philosophy of international law. He is the author of One Another’s Equals: The Basis of Human Equality (Harvard University Press, 2017).
Basic Equality An Analytical Introduction Nikolas Kirby and Giacomo Floris
1. Introduction The acceptance of the idea of basic equality is widely thought to be one of the most significant achievements of modernity. Ronald Dworkin and many others have observed that the claim ‘that we are all one another’s equals’ is at the foundation of contemporary normative theory: moral, political, and legal (Dworkin 1978: 272–273; see also Arneson 2014; Kymlicka 2002: 4; Miller 2007: 28; Pojman and Westmoreland 1997: 1; Scanlon 2018: 4; Sher 2017: 31). On this view, we have reached an ‘egalitarian plateau’. Sharing the same egalitarian premise regardless of our other differences, we only disagree now about what follows from it: what can be found further up the mountain. This very idea of our ‘basic equality’ is, therefore, putatively a fundamental interpretative resource of normative theory, such that a primary test of plausibility between reasonable theories of justice is: which one best reflects the fact that we are one another’s equals? This collection advances the debate about this fundamental commitment to the principle of basic equality by analysing its meaning, explanation, and scope. In this introduction, we outline the key issues that theories of basic equality are concerned with and the most significant challenges that they face, as well as introducing some conceptual distinctions that will help to clarify important aspects of this debate. This, in turn, will provide background for understanding the questions that the chapters in this volume seek to answer. We will proceed as follows. Section 2 presents the ‘key theses’ related to the idea of basic equality. Sections 3–5 reflect the threefold structure of the book. Each section offers an explanation of the main issue addressed in the
Nikolas Kirby and Giacomo Floris, Basic Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0001
2
Basic Equality: An Analytical Introduction
corresponding part of the volume, and provides a brief summary of each of its chapters.
2. Basic Equality: Key Theses As observed above, the idea of basic equality is considered foundational to contemporary moral, legal, and political philosophy. In particular, we can distinguish between four key theses: (1) Basic Equality Thesis: All human beings are one another’s equals. (2) Disciplinary Thesis: All plausible theories are premised upon the claim that ‘A ll human beings are one another’s equals’. (3) Methodological Thesis: A theory is more plausible than another theory if it offers a better interpretation of the shared premise that ‘A ll human beings are one another’s equals’. (4) Historical Thesis: Only modern theories are consistent with the claim that ‘A ll human beings are one another’s equals’. These four key theses, widely shared in the literature, are often combined with four further claims. First, there is a cognate claim, most clearly first expounded by Amartya Sen (1979), that the fact that we are one another’s equals entails some kind of substantive egalitarianism as justice, most obviously an equal entitlement to something: (5) Egalitarian Thesis: That all human beings are one another’s equals entails some form of substantive egalitarianism. Second, following a line of work that goes from George Morgan (1943) to John Rawls (1971) to Bernard Williams (1973), it is argued that the claim that we are all one another’s equals must have some kind of ‘basis’—an explanation: (6) Explanatory Thesis: If we are one another’s equals, then there must be some explanatory basis for why we are one another’s equals. However, third—more implicit, admittedly, in the priorities and practice of analytic normative theory than explicit—there has been an assumption that, given all plausible theories share the same premise that we are one another’s equals, any explanation of such basic equality—or lack thereof—will be
Nikolas Kirby and Giacomo Floris
3
consistent with all plausible theories. This is to say that it will be ‘ecumenical’ between them—providing equal support (or not) to all theories, and thus offering no comparative dialectical advantage—no reason to choose— between them. (7) Ecumenical Thesis: Any explanatory basis (or lack thereof ) for why we are one another’s equals will be ecumenical between all plausible theories in its implications. Finally, providing such an explanatory basis for basic equality—it turns out—is very difficult; indeed, some believe ‘insurmountable’ (Husi 2017: 376; see also Arneson 2014); ‘one of the most profound problems of moral philosophy’ (Christiano 2008: 17).¹ Given the putative shared importance of the idea of basic equality for theories of justice—in the form of theses (1)– (7)—one might assume such a difficulty ought to precipitate a foundational crisis in normative theory. The reaction, however, to this difficulty has largely been deafening indifference. There appears to be little if no anxiety from those working higher up upon this putative foundation of normative theory that the foundation might be itself entirely ungrounded. This disciplinary indifference might be thought to rest upon an assumption that no one could plausibly deny that we are one another’s equals, regardless of that fact’s current state of explanation. Something must have gone awry in our substantive reasoning or methodology to explain why it is so hard to explain the truth of basic equality, but its truth cannot be denied. As John Schaar puts it, ‘Nature … shouts ‘inequality’. We reply, ‘nonetheless, equality’, (Schaar 1964: 867): (8) Negative Thesis: No one can plausibly deny the fact that we are one another’s equals.² This late twentieth-century picture of a widely shared commitment to an ‘egalitarian plateau’ is undoubtedly attractive: a common foundation that frames a normative theory as a shared discipline. The work in this volume, however, reflects its unravelling. Each and every one of the eight theses ¹ As early as 1942, Henry Spiegelberg proclaimed: ‘The situation is in fact so serious that some of the advocates of democracy are on the point of abandoning the whole doctrine of equality’ (Spiegelberg 1944: 102). ² Thesis (8) might be thought to be a mere implication of Thesis (2). But one can hold (8) and reject (2). That is, one can try to build a plausible normative theory that does not rest upon basic equality, without affirming that basic equality is false. For example, some versions of utilitarianism do not rest on a commitment to moral equality, but they do not necessarily deny that persons are each other’s equals.
4
Basic Equality: An Analytical Introduction
above is challenged. The upshot, we suggest, is that the time for disciplinary complacency about basic equality is over.
3. The Meaning of Basic Equality: History and Import of Claiming to Be Equals 3.1 The Multiple Meanings of Basic Equality The unravelling of the commitment to an ‘egalitarian plateau’ first begins upon attempting to define more precisely what the (1) Basic Equality Thesis that ‘all human beings are one another’s equals’ means, as a claim putatively shared by many historical and contemporary authors. This is not to ask what the claim should mean, but what those authors who have shaped and/or now constitute our supposed consensus do mean by it. Beginning simply with what various authors say, however, we are confronted by a dizzying multiplicity. They speak of: ‘equal worth’ (Pojman 1997; Vlastos 1984: 43ff ); ‘equal consideration’ (Benn 1967; Miller 2007: 28); ‘equal intrinsic weight’ (Haksar 1979: 17); ‘equal intrinsic value’ (Bedau 1967: 17; Frankena 1962: 14); ‘the equal primary importance of everyone’s life’ (Nagel 1991: 131); ‘entitlement to equal concern and respect’ (Dworkin 1978: 272–273; 1986: 191; 2000: 11); ‘the interests of each member of the community matter, and matter equally’ (Kymlicka 2002: 3–4); ‘fundamental equal moral status’ (Arneson 2014: 30); ‘equal basic dignity and worth’ (Arneson 2014: 30); ‘each person matters equally’, ‘the meaning of equality is specified by the principles of justice which require equal basic rights be assigned to all persons’ (Rawls 1999: 442); ‘some equality of value or rights is intrinsic in human nature’ (Morgan 1943: 116); ‘[a]ll and every particular and individual man and woman … are and were, by nature all equal and alike in power, dignity, authority, and majesty’ (Lilburne 1951: 317); ‘innate equality, that is, independence from being bound by others no more than one can in turn bind them; hence a human being’s quality of being his own master (sui juris)’ (Kant 1996: 238), and many other formulations. Prima facie, such multiplicity in locutions of (1) suggests some form of conceptual disunity beneath, rather than supposed unity, that is, the denial of (2) the Disciplinary Thesis. To clarify, let us assume that (1) involves each author saying some ‘version’ of it, such that: (1a) Basic Equality Thesis: ‘A ll human beings are one another’s equals’ in a (particular) morally significant sense.
Nikolas Kirby and Giacomo Floris
5
This, however, should be distinguished from the more agnostic claim: (1b) Basic Equality Thesis: ‘A ll human beings are one another’s equals’ in some (any) morally significant sense. Of course, it is unlikely that many (if any) theorists mean (1b) by their claims that ‘all human beings are one another’s equals’, although it does necessarily follow from (1a) a fortiori. However, distinguishing (1b) is useful for two reasons: first, as we shall see below, the problem of basic equality is the negation of (1b), not merely (1a). And second, it permits us to make the following further distinction with respect to the disciplinary thesis: (2a) Disciplinary Thesis: All plausible theories are premised upon the claim that ‘A ll human beings are one another’s equals’ in a (particular) morally significant sense. (2b) Disciplinary Thesis: All plausible theories are premised upon the claim that ‘A ll human beings are one another’s equals’ in some (any) morally significant sense. Those, like Dworkin, who most prominently espouse the Disciplinary Thesis appear to be asserting (2a). Yet we can see in the multiplicity above evidence that (2a) is false. Nonetheless, some have not lost faith. In this volume, for example, George Sher holds that ‘it is widely agreed that all normal humans have equal moral standing’ and that what is meant by this claim, by consequentialists and non-consequentialists alike, is that ‘two beings will have the same moral standing if their interests count equally toward what morality requires’ (Sher, this volume: 287). On this view, any equality in deontic incidents, like rights, will then be derived from this claim of equality of interests. Sher is not alone, but, as Kasper Lippert-Rasmussen indicates, also in this volume (85–86), it can be argued that attempting to define the concept of basic equality only with respect to interests fails to include those non-consequentialist views that see rights as fundamental, or at least as not grounded upon any assumption about interests (for example, ‘will theorists’, such as Hart (1982; 1983)). Other authors in this volume, therefore, admit multiple senses of basic equality not reducible to any one sense. The most popular version takes there to be not one but two morally significant senses of basic equality in play, consequentialist and non-consequentialist. This is a position proposed by Richard Arneson in earlier work (Arneson 2014: 30), although it is anticipated even earlier by George Morgan (1943: 115–116).
6
Basic Equality: An Analytical Introduction
In this volume, Andrea Sangiovanni puts the thought as follows: ‘There is something that all are equal in the possession of that generates significant and basic moral duties owed to beings that possess it—duties of equal respect (mostly in nonconsequentialist accounts) or equal consideration of welfare (mostly in consequentialist ones)’ (Sangiovanni, this volume: 104). Whilst this ‘bifurcated’ concept of basic equality may seem to be a plausible characterization of the concept shared at the centre of much contemporary analytic moral theory, looking even a bit beyond these bounds we quickly see this claim too to be under-inclusive. As Lippert-Rasmussen points out, it is unable to account for the sense in which a virtue theorist might have a commitment to basic equality (Lippert-Rasmussen, this volume: 85). Or, as Nikolas Kirby observes in earlier work (Kirby 2018), it fails to capture the sense shared by classical social contract theorists in which our fundamental equality is our equal fundamental authority, not our equal rights or interests. Does the acceptance of the multiplicity—the polysemy—of ‘basic equality’ entail a rejection, therefore, of (2a) and thus aspirations for unity in normative theory? Some authors are reluctant to give up quite yet. Their strategy is abstraction amidst multiplicity: to identify a concept sufficiently neutral between all these various meanings. Jeremy Waldron holds: ‘[B]asic equality … requires that all persons be given the benefit of whatever principles of social justice we come up with’ (Waldron, this volume: 72). Lippert Rasmussen contends that basic equality is a position that holds for all equals: ‘the same basic moral considerations apply to all of them’ (LippertRasmussen, this volume: 86). Ian Carter ties basic equality to: ‘equality of certain entitlement-grounding properties’ (Carter, this volume: 127; see also Carter and Page, 2003). And, finally, Kirby offers an entirely functional definition, where ‘we are all one another’s equals’ is cashed out as the claim that ‘all (or nearly all) humans are one another’s equals in their possession of some basic moral property’ where ‘basic moral property’ is simply a placeholder for any moral property that is putatively ‘natural’, ‘inalienable’, with ‘great weight or lexical priority’, and thus ‘theoretically foundational’ (Kirby, this volume: 176–177; see also, Kirby, 2024). It is plausible that such a concept of basic equality can be crafted so abstract that (almost) all contemporary plausible theories do share it, thus making (2a) true. However, the worry is that any such unity is trivial: so formal that nothing substantive is really shared between theories by virtue of this common premise (Sangiovanni, this volume: 104; see also Floris and Spotorno, forthcoming; Kirby 2018; Nathan 2014). (2b) just picks out a collection of claims that share a common feature of concepts, rather than any meaning common to those concepts. If this is true, then the Methodological Thesis
Nikolas Kirby and Giacomo Floris
7
(3) is immediately under threat, since there is no common concept that various theories can be considered to be competing conceptions.³ The plateau has begun to crack.
3.2 Overview of the Chapters Is it possible to concede the proliferation of meaning(s) of basic equality, resulting in disunity in moral, political, and legal theory, and, thus, reject (2a) and (3), but still hold that our shared commitments to basic equality are a kind of modern achievement—that is, Historical Thesis (4)—and vital to justifying any ‘egalitarian’ theorizing—that is, Egalitarian Thesis (5)? These are the questions that the chapters in Part I of this volume address. Let us begin by making a similar distinction to that above: (4a) Historical Thesis: Only modern theories are consistent with the claim that ‘A ll human beings are one another’s equals’ in a (particular) morally significant sense. (4b) Historical Thesis: Only modern theories are consistent with the claim that ‘A ll human beings are one another’s equals’ in some (any) morally significant sense. (5a) Egalitarian Thesis: That all human beings are one another’s equals in a (particular) morally significant sense entails some form of substantive egalitarianism. (5b) Egalitarian Thesis: That all human beings are one another’s equals in some (any) morally significant sense entails some form of substantive egalitarianism. In Chapter 1, ‘On the Historical Emergence of Basic Human Equality’, Teresa Bejan can be taken to be denying (4b). She traces assertions of basic equality, in different kinds of morally significant senses, all the way back to premodern classical and medieval sources. She argues that it made sense to claim that ‘we are one another’s equals’—isotes—in classical Athens, but that particular sense owed its roots to the agora. The term derived from the balance-beam scale at the centre of its commerce. Isotêta existed where there was balance between two objects on the scale. Bejan argues that this, indeed, implied the possession of some measure to the same degree, but that the fundamental meaning of equality was the balance—the harmony—achieved by the tool itself. Thus, she argues, in ancient Greece the concept of ‘equal justice’ ³ But, as we shall explore further below, so too is the Ecumenical Thesis (7).
8
Basic Equality: An Analytical Introduction
presupposed who was to be weighed, and was constituted by balance, without entailing quantitative sameness. ‘Equality-as-balance’, however, is not our only classical egalitarian inheritance. Bejan also traces two distinctly Roman ideas of ‘equality-as-indifference’ and ‘equality-as-parity’. The former is the property of being subject indifferently to the same set of rules and principles, a precursor we might think to our more modern idea of ‘equality before the law’—even though, as Bejan emphasizes, in Rome the crucial question would remain, but which law? One could be another’s equal at natural law, but not at civil law. The latter idea of ‘equality as parity’, by contrast, derives its origins from the idea of pairing or matching different objects together. Being ‘paired’ is the property of collecting together objects otherwise incommensurable, by rough comparison, as peers. Finally, Bejan argues that these classical concepts of equality—as balance, indifference, and/or parity—pass into subsequent medieval Christian thought. However, their meaning explains why such equality in the eyes of God has few implications for social structures of hierarchy. We could be equals in terms of balance and indifference, but not parity. Thus, as she denies the historical thesis (4b), Bejan also shows how past authors disprove (5b): as medieval Christian thought shows, one can have a concept of basic equality that lacks any form of substantive egalitarianism. Of course, denying (4b) and (5b) does not entail rejecting (4a) and (5a). It is plausible that there is a particular sense (or indeed sub-set of senses) in which we are one another’s equals that only begins to be asserted in modernity and does imply substantive egalitarianism. Indeed, this might be a more accurate rendering of the moral progress that several contemporary philosophers discussed above envisioned: not that authors began to claim in modernity that all humans are one another’s equals, but that they began to hold that we are equals in sense(s) that have real egalitarian implications. But what exactly is an ‘egalitarian implication’? As observed above, the most influential view of what made a conception ‘egalitarian’ is Amartya Sen’s (1979) claim that egalitarianism necessarily requires distributing something equally among us as equals—whether that be money, welfare, utility, resources, capabilities, advantages, opportunities, or so on: (5c) Egalitarian Thesis (Distributional): That all human beings are one another’s equals implies the equal distribution of something. This framing grounded the ‘currency of egalitarian justice’ debates of the 1980s and 1990s. However, in the late 1990s, a number of authors began challenging this thesis. Led by Elizabeth Anderson, Samuel Scheffler, and
Nikolas Kirby and Giacomo Floris
9
Jonathan Wolff, these ‘social, ‘democratic’, or ‘relational’ egalitarians argued that justice primarily requires equality in relations rather than some valuable good(s)—that we are all able to relate to one another as equals, rather than hold an equal amount of some valuable good(s) (Anderson 1999; Scheffler 2003; Wolff 1998). Whilst the content of such a claim was—and many argue continues to be—somewhat vague, relational egalitarianism stands as distinctly opposed to relations of inferiority, domination, arbitrary power, status, and disesteem. Accordingly, relational egalitarians maintain that distributive equality is not important in and of itself, but it matters only if—and to the extent that—it is necessary for persons to stand in relations of equality with one another. However, if relational equality stands as an alternative (or at least an addition) to distributional equality, then how is relational equality supposed to stand with respect to basic equality? Prima facie, the most intuitive proposal would be something like: (5d) Egalitarianism Thesis (Relational): That all human beings are one another’s equals implies that all human beings must relate to one another as equals. In Chapter 2, ‘Basic and Relational Equality’, Jeremy Waldron explores this thesis. In particular, he considers two different possible ways in which (5d) might be true. The first is that relational equality is implied by basic equality because the former defines the latter. On this view, simply what it means (or at least should mean) for us to be one another’s equals is to owe each other relations of equality: the absence of inferiority and superiority, dominated and dominator, esteemed and disesteemed, and so forth. However, Waldron observes that this view is unable to account for the non-relational implications that are entailed by basic equality: for instance, ‘If X is punished much more severely for a given offence than Y is, then there ought to be some sort of egalitarian concern about the disparity whatever the impact is on the relationship between X and Y (if indeed they are even known to one another)’ (Waldron, this volume: 76). It follows from this then that relational equality is only an aspect—albeit a very important one—of what it means for us to be one another’s equals. Accordingly, Waldron’s main contention is that basic equality and relational equality are to be understood as two distinct normative commitments. Specifically, the former must ground the latter: (5e) Egalitarianism Thesis (Relational grounding): The fact that all human beings should relate to one another as equals must be grounded upon the fact that we are all one another’s equals.
10
Basic Equality: An Analytical Introduction
It is at this very point that Kasper Lippert-Rasmussen, in Chapter 3, ‘Should People Who Are Moral Equals Relate as Social Equals? Should People Who Are Not Moral Equals Relate as Social Unequals?’, takes his point of departure. First, Lippert-Rasmussen rejects (5d). He argues that relational equality, at best, is a possible, indeed instrumental, implication of the commitment to basic equality, understood as a commitment to something falling within the scope of the Basic Equality Thesis (1b), whereby ‘all human beings are one another’s equals in some (any) morally significant sense’. However, secondly, and more importantly, contra Waldron, Lippert-Rasmussen also denies (5e). He aims to demonstrate that we can have sufficient reasons for relational equality even if (1b) is false. Specifically, Lippert-Rasmussen argues that relational equality is compatible with holding that although persons do not have equal moral status, they either (i) have an incommensurable moral status,⁴ or (ii) they each have a sufficient moral status, such that a sufficient number of important things cannot be done to them. Lippert-Rasmussen’s chapter, then, is an attempt to cut one contemporary theory of justice—indeed an avowedly egalitarian theory—free of needing to assume basic equality. In doing so, he comes to question all the other theses above: but most dramatically the Negative Thesis (8), as he is willing to concede that we may not be equals. However, far from being an implausible conclusion, this—so LippertRasmussen argues—is a significant result because, despite the widespread popularity of basic moral equality, ‘it is far from clear that there is any real prospect of a knock-down argument for basic moral equality being provided by those who subscribe to it’ (Lippert-Rasmussen, this volume: 100). In the last chapter of Part I, Chapter 4, ‘Is There a Thing Called Moral Equality? (And Does It Matter if There Isn’t?)’, Andrea Sangiovanni might be thought to be offering, on the one hand, a more radical decoupling of normative theory from basic equality than Lippert Rasmussen, and, on the other hand, a way of avoiding the radical implication of rejecting the Negative Thesis (8). In particular, Sangiovanni’s negative argument aims to show that the main senses in which contemporary theories claim that all human beings are one another’s equals—such as the idea that persons have equal fundamental rights, or the view that their comparable interests ought to matter equally— are ‘otiose’.⁵ Akin to Lippert-Rasmussen, his aim is to show that basic equality does no real work in most moral theorizing: both the Disciplinary Thesis (2) and the Methodological Thesis (3) are false. This, however, does not mean that we should abandon the commitment to moral equality; rather, we must ⁴ This kind of argument is anticipated by Cupit (2000) and Morgan (1943). ⁵ For further discussion, see also Lucas (1965: 298), Raz (1986: 219–220), Westen (2014), and Wollheim (1955: 288).
Nikolas Kirby and Giacomo Floris
11
reinterpret it. To do this, we need to distinguish between two distinct roles that the commitment to moral equality has historically played in political and social discussion. First, it has undermined ‘the structures that serve to enable the oppression of the “inferior” by the “superior”—structures, that is, constituted by (objectionable) social status hierarchy’ (Sangiovanni, this volume: 110). Second, it has provided the basis for extending our moral concern to at least some nonhuman animals. Crucially, Sangiovanni observes, arguments for the moral status of nonhuman animals and against social status hierarchy come apart. This has two significant implications: first, it reveals that arguments for moral equality are neither necessary nor sufficient to justify social status equality. This is a direct rejection of Waldron’s (5e) thesis above. Second, it shows that our commitment to social status egalitarianism should ultimately be understood as a commitment against relations of inequality.
4. The Explanation of Basic Equality 4.1 The End of Ecumenicism: Why the Basis of Basic Equality Matters Following the Explanatory Thesis (6), it is an intuitive assumption that if we are one another’s equals, then surely there is some kind of explanation for that fact. But as foreshadowed above, and as we shall detail further below, there is a seemingly intractable problem, or indeed, as we shall distinguish, at least two problems in providing that explanation. However, it was also noted above that current normative theorists in general are largely complacent about such a problem, since any solution (or lack thereof ) would be ecumenical with respect to them: that is, (7) the Ecumenical Thesis—no comparative dialectical advantage turns upon if and how our basic equality is explained. Yet the developments outlined in the preceding section provide two reasons to hold that (7) is false. First, if we follow Lippert-Rasmussen’s and Sangiovanni’s arguments above, then we can explain the justice of relational equality without reference to the fact that we are one another’s equals. Putatively, therefore, they offer the prospect of a plausible contemporary normative theory that does not rely upon the claim ‘that we are all one another’s equals’ in any sense. This position denies the Disciplinary Thesis (2b)—‘all plausible theories are premised upon the claim that “all human beings are one another’s equals” in some (any) morally significant sense’. This means, however, that any problem
12
Basic Equality: An Analytical Introduction
of basic equality that exists for other theories is not their problem. And thus, if others cannot find a solution to that problem, then ceteris paribus relational egalitarians have a comparative dialectical advantage over them. Alternatively, even if Lippert-Rasmussen and Sangiovanni are wrong, and (2b) remains true, there is still reason to hold that (7) is liable to be false, because (2a) is false. That is, even if, in accordance with (2b), all plausible normative theories have a premise that holds that we are all human beings in some (any) morally significant sense, if that is a different sense in each theory, then it is likely that both the explanation(s) for, and the implication(s) of, the truth of such different ways in which we are equals will themselves be different. For example, why would we expect the explanation (if any) of the fact that we have equal worth to be the same as the fact that we have equal dignity, or equal value, or interests of equal concern, or so on? And why would we expect the implications of such an explanation for the rest of normative theory to be the same? As Ian Carter is fond of reminding us in this context, citing Joseph Raz, ‘the ground of an entitlement determines its nature’ (Carter 2011: 542; Raz 1986: 223). Either way, the upshot is that assessing the competing explanations and meanings of basic equality matters for normative theory, far more than has been hitherto appreciated. Complacency about it is unjustified. Even if one continues to sustain the Negative Thesis (8)—that no one can plausibly deny that we are one another’s equals—then it still matters how one goes about not denying it.
4.2 Grounding and Justification Before introducing the problem of explaining basic equality, it is worth addressing one particular difference of terms within the existing literature, and indeed this volume. Some authors speak of the problem of ‘grounding’ basic equality, others of ‘justifying’ basic equality, and others use the terms interchangeably. Obviously, terms can be defined in any way an author wishes, but two concepts do need to be distinguished. In contemporary metaphysics, ‘grounding’ has become a technical term. In short, ‘A grounds B’ is a relationship of explanation. It is the because or in virtue of relationship. For some, causal explanations fall within its scope. For others, it’s the term we use to refer to non-causal explanations, such as constitutive or normative explanations. For some, grounding is a relationship between predicates, such that having X grounds having Y. For others, grounding is a connective between sentences such that ‘A has X’ grounds ‘B has Y’.
Nikolas Kirby and Giacomo Floris
13
Without any need, however, to adjudicate these technical disputes, and just adopting the latter usage, we can say that the search for the ‘basis of basic equality’ is looking for a fact that grounds the fact that we are one another’s equals.⁶ ‘Justification’, at least within normative theory, is typically understood as sufficient reason to promote, realize, and/or respect that state of affairs; or what ‘makes it right for someone to act in a certain way’ (Alvarez 2017). Something is unjustified, or lacks justification, therefore, if it lacks a reason to promote, realize, and/or respect that state of affairs (or nothing makes an action ‘right’). However, in the normative domain, one might take justification to be the ground of facts about normative states of affairs. For example, we might say that the fact that your life will be saved by this healthcare policy might be taken, ceteris paribus, as a justification for securing it. But we can also say that the fact that your life will be saved by this healthcare policy, ceteris paribus, grounds the fact that we should secure it. However, compare: Grounding of Basic Equality: By virtue of the fact G, it is true that we all have equal worth. Justification of Basic Equality: Fact J is sufficient reason to promote, realize, and/or respect a state of affairs where we all have equal worth.
The first claim leaves open the possibility that there is no particular reason why we should be equals. Fact G makes it true that we are equals but does not necessarily make it true that this is a valuable state of affairs. By contrast, the second claim is precisely asserting a reason why we should be equals (perhaps if we were not now, or were liable to become unequals in the future). Fact J makes it true that our equality is a valuable state of affairs, but not necessarily that we are (always) one another’s equals. With this distinction in hand, we take most but not all theorists of basic equality today to be discussing the grounding of basic equality, not its justification (if any). However, early work on basic equality does explore the latter. This is why for example George Morgan, Henry Spiegelberg, and ⁶ Until this recent analysis emerged, philosophers often used the term ‘grounding’ interchangeably with ‘supervenience’, including in debates about basic equality. But, at least now, the term ‘supervenience’ is taken to be a logical relation between properties, that is, property A supervenes upon property B, such that there is a kind of reciprocal entailment in their instantiation. If and only if X has A, then X has B, or for any object, no change with respect to A can occur without a change with respect to B and vice versa. Supervenience, therefore, unlike the grounding relations does not presuppose an explanatory relationship. Further, they have different extensions: A may ground B, without B supervening upon A. For example, when B is over-determined, such when other properties would explain B’s instantiation, if A were not instantiated (Berker 2017: 735; Fine 2012: 38; McLaughlin 1995).
14
Basic Equality: An Analytical Introduction
Isaiah Berlin all discuss basic equality as a putative ideal, not just an idea— something to be achieved, not merely explained—and thus see the need to address the apparent implication of such an ideal that we ought to equalize certain personal qualities, making us all uniform in any, or indeed all, domains (Berlin 1955; Morgan 1943; Spiegelberg 1944). In a cognate fashion, for example, in this volume Andrea Sangiovanni argues that a search only for grounding without justification would lead to a position that: ‘If it turns out that there is no morally relevant range or binary property on which moral status might be grounded, then [one] would be happy to welcome inequalities in status among people with (presumably) different psychological capacities’ (Sangiovanni, this volume: 107). Since he holds that to be an acceptably counter-intuitive conclusion, he presses that we must instead be searching for a justification of basic equality instead: why equality in status amongst human beings is always, regardless of such capacities, a state of affairs to promote, realize, and respect. Indeed, for Sangiovanni this is part of why we should turn away from the search for a grounding for basic equality at all, and instead aim to articulate a free-standing justification of relational equality. Whether basic equality needs a grounding and/or a justification is an important question. But our only point here is to highlight the need for theorists of basic equality to distinguish more carefully grounding and justification in terminology, so that they might position themselves clearly within this debate.
4.3 The Problem of the Explanation of Basic Equality In Section 3, we emphasized the apparent multiplicity of possibly conflicting senses in which theorists have asserted that we are one another’s equals, and indicated that it is hard to find a coherent way in which one theorist could assert all of them—let alone everyone do so at the same time. The startling thing about the problem of basic equality, however, is that it entails the putative negation of all of them at the same time. It is the negation of (1b), not merely some particular version of (1a), that is at stake: (1∗ ) No Basic Equality: There is no morally significant sense in which we are all one another’s equals. The argument for (1∗ ) begins with the Explanatory Thesis (6): if we are one another’s equals, then there must be some explanatory basis for this fact. This basis—this grounding—will, putatively, explain why we are one another’s
Nikolas Kirby and Giacomo Floris
15
equals, and indeed why other entities are not, such as plants, or non-sentient animals (see Section 5.1). Thus, we expect the explanation to be a fact about us, something that holds true of only us, and not others. However, when searching for such an explanation we assume that we should limit the domain of possibilities in a few ways. First, we are looking for an empirical fact: if we want to explain the difference in moral status between empirical beings, then presumably we need to ground that difference (at least in part) in an empirical difference. Second, obviously, non-relevant properties are also excluded candidates. For example, (almost) all human beings, and only human beings, have a long opposable thumb compared to finger length, but it is hard to foresee how any moral implication could be explained by this fact. So we are in search of non-normative properties all human beings have, which all other beings lack, that could plausibly be morally relevant. Candidate properties are rationality, imagination, language, and capacity for virtue, among others.⁷ But this generates the following problem: Variation Problem: Any plausible candidate basis of basic equality will come in degrees, and thus, ceteris paribus, any difference of degree in the candidate basis should ground a difference of degree in any morally significant property that it grounds.
For example, let us take what is typically assumed to be the most obvious candidate for the basis of human equality: rationality. ‘Rationality’, of course, is a complex concept, and can be defined in many ways, but in any way that one cuts it, human beings will be rational to differing degrees. Some human beings are just more reasonable, consistent, coherent, and logical in their beliefs, attitudes, and actions than others. However, so the argument goes: if the fact that we are rational is so significant that it grounds the fact that we possess some kind of basic moral import, then surely greater rationality will ground greater moral import? As Louis Pojman illustrates: ‘If P constitutes human worth, then it would seem that the more of P that a person has, the better he or she is … If reason is really all that makes us valuable, then the more of it the better … If our ability to will the good is what gives us value, then it would seem that some people are more valuable than others because they have greater ability to will the good than others’ (Pojman 1991: 484–485).⁸ There is what Richard Arneson describes as a kind of inescapable ‘pressure of reason’ to so conclude (Arneson 2014: 36). ⁷ For further discussion, see Jaworska and Tannenbaum 2023. ⁸ For further discussion, see also Carter (2011: 541), Christiano (2014: 56), Floris (2023a: 38–39), and Kirby (2024).
16
Basic Equality: An Analytical Introduction
The Variation Problem is such a significant challenge because it does not merely appear to exclude an explanation for why we are all one another’s equals, that is, the No Basic Equality thesis (1∗ ), but instead invites explanations for the fact that we are all unequals. Any argument that one might make for the fundamental moral import of any particular property that comes in degrees, like rationality, can easily serve as a premise in the argument that, by the pressure of reason, surely, we must be one another’s equals, that is: (9) Basic Inequality: We are unequals with respect to one another. Following (8), some might take (9) to be a somewhat idle conclusion: something no one, in good faith, could rely upon in normative argument. But counter-examples exist, not just in popular culture, but in contemporary philosophy (Kekes 2003; Steinhoff 2014). Yes, sexist, racist, able-ist, and other discriminatory views are included within this set. Indeed, it may be no coincidence that interest in the basis of basic equality emerged among Anglo-American authors in 1942–1945. However, as Uwe Steinhoff has powerfully argued, such a dark history of inegalitarian views should not be taken to discredit all such views (Steinhoff 2014: 142–145). Indeed, as he observes, only a respectable theory of basic inequality, grounded in a plausible basis of unequal moral worth, can explain why human beings—like Adolf Hitler—are our moral unequals due to their evil. Is that not an attractive conclusion?
4.4 Overview of the Chapters Over the last eighty years, a number of solutions, or dissolutions, of the ‘Variation Problem’ of basic equality have been proposed. One might reject the starting assumption that the possession of a morally basic property needs to be grounded in any non-normative fact about humans: it is just an ungrounded fact (Gosepath 2014; cf. Husi 2017: 388). Or, insofar as it is grounded, it is grounded in our ‘humanity’, and whatever then grounds the fact that we are human (Frankena 1962; Vlastos 1984; cf. Schaar 1964: 875–884; Wilson 1966: 93). Or, one might argue that our equal possession of a particular morally basic property is not even a fact about humans, but rather it is a proposition that we each must choose to assume about humans when engaging with them (Arendt 2006; Macdonald 1946; Phillips 2021; cf. Waldron 2017: 55–61). Or, one might accept the starting assumption of the argument above, but argue that the non-normative property may be theological (Waldron 2017: 175–214; cf. Thomas 1979: 539–540) or transcendental
Nikolas Kirby and Giacomo Floris
17
(Kant 1996; cf. Williams 1973), thus avoiding any presumption that it is likely to be spread unevenly amongst us. Or, one might argue that Basic Equality follows as a formal principle of rule application (Frankfurt 2015; Lucas 1965; Westen 2014; cf. Waldron 2017: 66–83). Or, one might look to the universal prohibition against treating others as inferiors (Sangiovanni 2017; cf. Floris 2019). Or, one might argue that the basis of moral equality lies in an independent moral requirement which offers a coherent and plausible explanation for why the variations to which the status-conferring property is held above a significant threshold are morally irrelevant (Carter 2011; Floris 2023b; cf. Arneson 2014: 44–48). Or, one might even concede that Basic Equality may be false but hold that some (many?) moral arguments need not actually rely upon it to justify many forms of so-called egalitarian rights (Husi 2017; Steinhoff 2014; cf. Waldron 2017: Ch. 1). All of these strategies have, so far, met powerful counterarguments (duly cited). Thus, Part II of this volume includes a mix of prominent contemporary arguments, revised and redefended, and new strategies to overcome the variations problem. In Chapter 5, ‘Basic Equality and the Contexts of Opacity Respect’, Ian Carter returns to defend an answer that he first proposed in his 2011 paper ‘Respect and the Basis of Equality’ (Carter 2011). Carter’s proposal begins with a position which was first put forward by John Rawls. Rawls appreciated the Variation Problem. His solution was to isolate what he called a ‘range property’. A range property is non-scalar—it does not come in degrees—unlike the other scalar candidate properties for basic equality hitherto discussed. More precisely, the range property is the non-scalar property of holding some scalar properties within a certain range; hence, it is either possessed or it is not. A non-scalar property, however, so the thought goes, may ground a property that is scalar, but in doing so ceteris paribus will ground the same degree in that latter property.⁹ However, prima facie, there appear to be no relevant non-scalar properties that all, and only, humans possess. So for Rawls the basis of basic equality was having ‘moral personality’, which is the range property of holding the capacities to form a conception of the good and have a sense of justice above a ‘minimum threshold’. Two problems, however, arise for Rawls’s range property view. The first is providing a reason for marking the threshold at any particular point (the ‘arbitrariness objection’) (Arneson 1999; 2014). The second is explaining why, even assuming the threshold is non-arbitrary, we actually have a reason ⁹ Since without relevant difference in grounding property, there is no reason for a difference in grounded property.
18
Basic Equality: An Analytical Introduction
to ignore differences in the ‘underlying’ scalar property above the threshold (the underlying Variation Problem, redux). In short, it is not clear that Rawls has made much progress at all. Carter’s proposal aims to address the second objection. His argument is that merely passing the minimum threshold of the range property of moral personality triggers, for others, in certain domains, a duty of what he calls opacity respect to refrain from taking into account the varying subvenient agential capacities above the relevant threshold. Accordingly, a commitment to opacity respect provides an independent explanation for the moral salience of the range property. In this chapter, Carter addresses one of the most pressing objections that has been raised against his view: according to this line of critique, persons do not have a fundamental duty to treat each other as opaque—indeed opacity respect is even an inappropriate moral attitude at least in some contexts, such as more intimate interpersonal relations. But if this is true, then opacity respect is unable to offer a compelling explanation for the appropriateness of treating people as equals (Arneson 2014; Wollner 2014). To address this challenge, Carter distinguishes between different ‘contexts of practical reasoning’, which provide different kinds of reasons for treating persons in a certain way. Carter argues that, on the one hand, while opacity respect is not appropriate in all contexts, neither is a commitment to basic equality; on the other hand, the contexts in which opacity respect is appropriate are also those in which we generally believe that persons should be regarded as equals. The upshot is that an appeal to opacity respect explains when and why it is appropriate to consider and treat persons as equals. In Chapter 6, ‘Equal Moral Status and the Collective Nature of Rationality’, Thomas Christiano proposes a defence of the range property view which, unlike Carter’s, does not rest on a commitment to an independent moral requirement. Christiano holds that human rationality—that is, the imperfect ability to apprehend what is valuable—grounds persons’ basic moral status. This, however, gives rise to the Variation Problem—or, as Christiano calls it, the ‘continuity argument’: if the possession of rational capacity grounds status, why does not the possession of greater rational capacity ground superior moral status? Christiano develops two arguments in response to this challenge. The first is a negative and intuitive argument, according to which it appears unfair to give greater opportunity for wellbeing to those persons who have a higher rational capacity because this amounts to ‘giving more to those who have more’. Christiano calls this the fairness consideration. The second positive argument is based on the collective nature of rationality. Specifically, the ‘Participation Argument’ runs as follows: ‘A . Each
Nikolas Kirby and Giacomo Floris
19
person’s status is grounded in their capacity to participate in the collectively rational process. B. If their status is grounded in the capacity to participate in the collectively rational process, then their status is determined by the impersonal worth of the collectively rational process. But C. If each person’s status is grounded in the same thing, then it is grounded in something of the same impersonal worth. D. If each person’s status is grounded in a thing with the same impersonal worth, then each person must have equal status. Therefore, each person has equal moral status’ (Christiano, this volume: 163). Christiano defends the Participation Argument from several objections: first, one may argue that if persons’ moral status is grounded in their capacity to participate in the collectively rational process, then unequal degrees of capacity to contribute to the collective process should generate unequal degrees of moral status. But Christiano rejects this objection by observing that it has the implausible implication of making persons’ moral status contingent on arbitrary factors—such as what society or historical moment they live in. This is the ‘Societal Variation Argument’. However, Christiano notes that the Societal Variation Argument may be unable to justify the moral equality of all persons because there may be some persons whose capacities are less valuable in any society. For example, severely disabled but still rational persons may be less capable of occupying any role in society. This is the ‘Extreme Disability Argument’. One possible answer to this objection is to expand the set of possible societies—not only socially, but also physically and logically possible—so as to have a more inclusive account of the value of a person’s participation in the collective rational activity. While this answer has some plausibility, insofar as it makes the degree of persons’ moral status invariant across possible worlds, Christiano observes that it is ultimately unclear what is the limit of the set of possible societies in which the variability of participation is relevant. The Participation Argument may, therefore, be unable to justify the moral equality of all human persons. In Chapter 7, ‘Basic Equality: Worth, Luck and Weight’, Nikolas Kirby begins with a proposal he suggests might assist both Carter and Christiano, among other ‘range property’ advocates (for example, Rawls (1971) and Waldron (2017)). This is to substitute the search for a so-called range property for what he terms a ‘bare property’. A bare property is also a non-scalar property defined by reference to a scalar property. But unlike a range property, it does not require defining any range on that scalar property by reference to some threshold(s). Instead, it is simply the ‘bare’ property of having any (some) degree of the underlying scalar property. Hence, a plausible bare property basis for basic equality would be a property—like say moral reasoning capacity—that (almost) all and only humans have to any degree at all.
20
Basic Equality: An Analytical Introduction
Without the need to define a threshold on the underlying scalar property, then, one of the problems facing the range property strategy disappears: the arbitrariness objection. We are still left, however, with the Variation Problem. However, unlike Carter, but somewhat akin to Christiano’s intuitive ‘fairness argument’, Kirby’s strategy is an attempt to explain how we can base equality upon a bare property in a way that properly considers variation in the underlying scalar property, rather than explaining why we should ‘ignore it’. Kirby’s strategy is in fact inspired by a suggestion found within one of the original papers on the problem of basic equality by Henry Speigelberg (1944). Drawing upon Spiegelberg, Kirby argues that (almost) all and (perhaps) only human beings have rational agency to some (any) degree. On this basis, (almost) all and (perhaps) only human beings can live lives of any degree of moral worth in the sense of worthy of genuine praise, resentment, and other reactive attributes. Of course, in part as a function of differences in degrees of moral agency, human beings will lead lives of different degrees of moral worth. But Kirby argues that the prohibition on circumstantial moral luck entails that the former differences should discount the latter. The result, he argues, is that each human being, because of having some (any) degree of moral agency, has equal ability to attain any degree of moral worth, regardless of what particular degree of moral agency they possess. Whilst the arguments in Chapters 5, 6, and 7 constitute different direct solutions to the Variation Problem of basic equality, the subsequent three chapters constitute more ‘indirect’ (dis)solutions. Each aims to undermine, in some way, the very assumptions that create the problem in the first place. In Chapter 8, ‘Equality and Moral Status: Challenges to Their Grounding’, Agnieszka Jaworska and Julie Tannenbaum develop a comprehensive account of different kinds of moral equality, and elaborate a novel response to the Variation Problem. First, they distinguish between four different kinds of moral status: Moral Status #1 concerns regard for an entity’s existence and welfare; Moral Status #2 concerns regard for an entity’s autonomy; Moral Status #3 concerns opacity towards an entity’s agential capacities; Moral Status #4 concerns treatment of individuals by social systems (political, legal, economic, linguistic, etc.). Jaworska and Tannenbaum explain that an entity can have more than one type of moral status, and that while different kinds of moral status identify different areas of moral concern, ‘degrees of status mark variations in the extent to which we should have that type of concern’ (Jaworska and Tannenbaum, this volume: 204). This gives rise to two questions: (1) In virtue of what does an entity have (at least one type of ) moral status? (2) Do entities have the same degree of moral status?
Nikolas Kirby and Giacomo Floris
21
As to the former, Jaworska and Tannenbaum argue that a plausible theory of moral status should be neither under-inclusive nor over-inclusive (see discussion of the Scope Problem below, Section 5). Hence, drawing on their previous work (Jaworska and Tannenbaum 2014; 2015), they argue that a high degree of Moral Status #1 should be grounded in not only complete but also incomplete realizations of sophisticated cognitive and emotional capacities: such an account provides a principled and compelling justification for why (almost) all human beings have a high degree of Moral Status #1, and their moral status is superior to that of (almost) all nonhuman animals. This view, however, seems vulnerable to the Variation Problem: if the relevant capacities are held to unequal degrees, how can their possession ground equal degrees of moral status? In response, Jaworska and Tannenbaum observe that the Variation Problem implicitly rests upon what they call the ‘Continuity View’: ‘if B is the basis of an M degree of X, then a greater degree of having or exhibiting B corresponds to a greater degree of X’ (Jaworska and Tannenbaum, this volume: 213). This view, however, cannot simply be assumed, but must be argued for.¹⁰ Thus, Jaworska and Tannenbaum attempt to show through several examples that it is intuitively plausible to maintain that the possession of different degrees of status-conferring properties can nonetheless give rise to rights of equal strength. Hence, they conclude that we have compelling reasons to hold that what matters is that entities possess a significant property up to a sufficient degree, and that they have equal moral status regardless of the variations above the relevant threshold. In Chapter 9, ‘When Equality Needs No Justification’, Anne Phillips put forwards a radical and powerful response to the problem of explaining equality: Phillips’ main contention is that we should reject the Explanatory Thesis (6)—‘if we are one another’s equals, then there must be some explanatory basis for why we are one another’s equals’. In her view, while we can, and arguably should, engage in arguments about why equality matters, we should refuse to seek a justification for why we ought to regard other humans as our equals. Basic moral equality is ‘a commitment we make to ourselves and others to regard one another as equals’ (Phillips, this volume: 225). In particular, Phillips argues that current theories of the basis of moral equality fail in several ways: first, by anchoring moral equality to the possession of some intrinsic features, they embrace a logic of justification which has historically legitimized the exclusion and the subjugation of several groups of human beings—such as women and people of colour—deemed to possess morally relevant features to an inferior degree. Second, the language of ‘moral ¹⁰ A similar line of argument has recently been proposed also by Zoltan Miklosi (2022).
22
Basic Equality: An Analytical Introduction
equality’ deprives equality of much of its normative force by suggesting that what matters is that people are considered and treated as moral equals in a minimal sense, despite severe material inequality between them. Finally, current theories of moral equality rely on the language of ‘moral worth’ and ‘moral status’ which makes individuals’ possession of basic human rights conditional on meeting some presumed standard of sufficient worthiness, thereby inevitably opening the possibility that some people have greater moral worth than others.
5. The Scope of Basic Equality 5.1 The Scope Problem: Who Is a Moral Equal? Let us concede that a solution of some kind, direct or indirect, exists for the Variation Problem of basic equality. This, however, is not the only challenge that theories of basic equality must face: indeed, a theory of basic equality must also define the scope of the beings that have equal moral status, that is, it must determine who is a moral equal. Theories of basic equality must therefore address the following problem: Scope Problem: For any plausible candidate basis of basic equality, there exists (or possibly exists) a human who lacks that basis, and/or a non-human entity who has that basis.
Different views of the scope of basic equality have been proposed in the literature. Some have defended the following version of (1), the Basic Equality Thesis: (1c) Basic Equality (of Human Species): All human beings are one another’s equals qua human beings. This idea is prominent in the Jewish and Christian traditions, for example, which maintain that human beings are one another’s equals (and superior to all other entities) because they are God’s special creatures.¹¹ But (1c) has also been defended on secular grounds. Thus, some have argued that human equality is based in the membership of the species homo sapiens: all human beings are moral equals simply by virtue of being human beings ¹¹ For instructive discussions of Christian accounts of moral equality, see Sangiovanni (2017: 27–31), and Waldron (2017: Ch. 5).
Nikolas Kirby and Giacomo Floris
23
(Williams 2006: Ch. 13). The speciesist account of moral equality, however, has been strongly criticized: many have pointed out that species membership is simply a biological feature that has no moral significance in and of itself. Hence, speciesism is a wrongful form of discrimination, which can justify neither human moral equality nor human moral superiority (McMahan 1996; Singer 2011). In light of this criticism, several philosophers have attempted to identify an intrinsic property which is morally significant and therefore can be a plausible candidate for the basis of moral equality. According to these views, then, those beings that possess the relevant status-conferring property are one another’s equals in virtue of being ‘moral persons’: (1d) Basic Equality (of Persons): (Almost) all human beings are moral equals qua moral persons. Many have claimed that the possession of highly sophisticated cognitive capacities—such as the capacity for rational and moral agency—offers a plausible justification for why persons are each other’s equals and superior to nonhuman animals and other living entities.¹² Critics, however, have pointed out that these views are objectionably under-inclusive. Indeed, grounding basic equality in the possession of sophisticated capacities has the disturbing implication of excluding many human beings who do not possess such capacities, such as infants and severely cognitively disabled human beings, from the scope of basic equality. This is known as the under-inclusiveness objection (Floris 2021). Accordingly, in response to the under-inclusiveness objection, it has been argued that the scope of basic equality should be broadened: the possession of highly sophisticated cognitive capacities is not necessary for being a moral equal. This line of argument has been defended most prominently by animal ethicists—such as Peter Singer and Tom Regan—who hold that basic equality is grounded in ‘sentience’ (Singer 2011) or ‘being-subject-of-a-life’ (Regan 1983). Since both human beings and (most) nonhuman animals are sentient beings and ‘subjects-of-a-life’, they ought to be considered and treated as each other’s equals. Some environmental ethicists have gone a step further, contending that the scope of basic equality should be expanded even more widely: this is because an entity counts morally simply if it is ‘alive’ (Schweitzer 1929) or is ‘a teleological-center-of-life’ (Taylor 2011 [1986]; cf. ¹² Most notably, this position has been defended by Kantian theories of the basis of basic equality. See Christiano (2014), Kant (2002 [1785]), and Rawls (1971). For a recent attempt to defend a more inclusive Kantian view capable of ascribing moral status also to nonhuman animals, see Korsgaard (2018).
24
Basic Equality: An Analytical Introduction
Floris and Porro forthcoming). It follows from this that not only nonhuman animals but also all living entities have equal moral status. These theories of animal and environmental ethics, then, reject 1(d) in favour of: 1(e) Basic Equality (of Living Entities): All human beings and (most) nonhuman animals (as well as (most) other living entities) are moral equals. However, some have observed that these theories of basic equality avoid the under-inclusiveness objection only at the price of being over-inclusive. As we have seen, in fact, these theories entail that nonhuman animals (and other living entities) have a moral status that is equal to that of some human beings. But many find this conclusion hard to accept. What we need, instead, is a theory of the scope of basic equality which is capable of including (almost) all human beings within its scope while at the same time providing a principled justification for why human beings have a moral status that is superior to that of nonhuman animals and other living entities. Thus, for instance, Jaworska and Tannenbaum have argued that the possession of the capacity to engage in ‘person-rearing relationships’ is a necessary and sufficient condition for having full equal moral status (Jaworska and Tannenbaum 2014). Others have affirmed that beings who have the potential to become rational agents are moral equals (Arneson, this volume; Kagan 2019). Since the capacity to engage in ‘person-rearing relationships’ and the potentiality to become rational agents are significant properties that almost all human beings possess—but nonhuman animals and other living entities lack—these views might be able to account for (i) the moral equality of (almost) all human beings, and (ii) the human moral superiority over all other living entities. In other words, they offer an attempt to rescue 1(d) from the under-inclusiveness objection without endorsing the overinclusive 1(e). The Scope Problem is a pressing challenge that any theory of basic equality must face. As illustrated above, the literature on ‘basic equality’ shares this Scope Problem with a cognate literature of practical ethics on ‘moral status’, inclusive of animal rights, the rights of the unborn, disability rights, and now emerging theory on artificial intelligence. Somewhat surprisingly, however, despite sharing this problem, theorists of basic equality and moral status have rarely engaged with one another. In order to initiate just such exchange, the last set of chapters aims to explore how to address this problem, in light of particular solutions to the Variation Problem. If nothing else, they demonstrate just how much any theorizing about the so-called margins of basic
Nikolas Kirby and Giacomo Floris
25
equality—such as with respect to children, human foetuses, humans with severe mental disabilities, the truly evil, higher-primates, ancestor species, super-intelligent robots, and the like—will depend upon the explanation of our equality in central cases.
5.2 Overview of the Chapters Part III begins with Giacomo Floris’ chapter, ‘The Basis of Children’s Moral Equality’, which examines how the Variation Problem can be overcome when this applies to the question of the explanation of children’s moral equality. According to Floris, the basis of moral equality is to be found in a fitting, basic, and independent moral attitude that offers a coherent and plausible rationale for why the variations above the threshold for moral status are morally irrelevant (Floris 2023b). Floris’ chapter then starts with a critical discussion of what is arguably the most influential attitude-based view of the basis of moral equality: Carter’s (2011) theory of opacity respect. Floris argues that while Carter’s view offers a plausible explanation for the equal moral status of persons—understood as fully competent adults who hold a wide range of sophisticated agential capacities—it is unable to account for the moral equality of children. This is because opacity respect is either not an appropriate response to children’s moral status, or it is not basic to what we owe to children. Hence, opacity respect does not have enough normative weight to justify a solid commitment to children’s moral equality. Floris thus contends that children’s moral equality must be grounded in another kind of moral attitude which is owed to children qua moral status holders. In particular, children’s moral equality is entailed by a commitment to a duty of what he calls ‘positive respect’, which requires providing children with those social conditions that foster their developmental process into well-adjusted adults endowed with a proper and stable capacity for moral personality. The reason for this is that unequal consideration and treatment precludes, or at least is a significant obstacle to, the provision of some goods that are crucial to the cultivation of proper and stable capacity for moral personality, namely, a robust sense of self-respect and access to integrated friendship. Therefore, it is inconsistent with a duty of positive respect towards children. Hence, children’s moral equality is a constitutive requirement of what is owed to children qua moral status-holders. In Chapter 11, ‘Basic Human Moral Equality’, Eva Feder Kittay defends the 1(c) Basic Equality (of Human Species) Thesis: ‘A ll human beings are moral equals qua human beings’, by arguing that being human is a sufficient—yet
26
Basic Equality: An Analytical Introduction
not necessary—condition for having full equal moral status. First, Kittay develops a critique of moral individualism, the view according to which moral status is ascribed not on the basis of species membership, but in virtue of a being’s intrinsic properties. She rejects Singer’s (2011) theory of moral equality by showing that it rests on an inadequate knowledge of the human beings that he excludes from the scope of equal moral status. This leads Singer to attribute to cognitively disabled human beings fewer and less developed capacities than those that they actually possess, thereby making implausible comparisons between them and some nonhuman animals—such as, for example, normally functioning chimpanzees. Kittay also criticizes McMahan’s (2003) moral individualist account, which measures the degree of a being’s moral status according to the degree of its psychological unity—the higher the degree of psychological unity, the richer a being’s life is. She argues that this view is vulnerable to two pressing objections: first, it is reasonable to maintain that the amount of goodness and the capacity for goodness in a being’s life are not fixed. But if this is true, does that mean that our moral standing varies throughout our lives? Few would accept this conclusion. Second, even assuming that capacities are fixed, who determines what counts as good—especially in the case of interspecies comparisons? McMahan’s theory, then, does not offer a principled and plausible way of determining the degree of a being’s moral status. In light of the difficulties, Kittay argues that we should reject moral individualism in favour of a relational view which grounds our moral obligations on the special relationships we stand in towards each other. For example, Kittay argues that her disabled daughter has equal moral status not because she holds some valuable properties, but because she stands in a special relation to her. Crucially, this kind of special relationship ought to be recognized as morally significant by everyone—not only by those who partake in it— thereby grounding one’s equal moral status in the eyes of everyone. More generally, then, Kittay’s main contention is that all human beings are moral equals because they are all a ‘mother’s child’. In previous work, George Sher argued that humans are moral equals because they all have a subjective perspective of their own, that is, humans have—or are—a separate consciousness (Sher 2014). However, as several critics have pointed out, Sher’s view seems to entail that at least some nonhuman animals have a moral status that is equal to that of human beings: after all, cats and dogs, for example, also have subjective perspectives. So, why ought humans to be considered and treated as morally superior to nonhuman animals? Sher’s view, then, appears to be committed to a version of 1(e), Basic Equality (of Living Entities): ‘A ll human beings and (most) nonhuman
Nikolas Kirby and Giacomo Floris
27
animals are moral equals’ (Arneson 2014: 44–45). In Chapter 12, ‘First among Equals’, Sher defends his theory of basic equality from this pressing objection. Sher does not deny that this implication follows from his view, but contends that this is not an implication that we should resist because it is compatible with favouring humans over nonhuman animals. The reason for this is that the fact that all persons are moral equals and facts about persons’ circumstances relevant to whichever principles we accept enter at different stages of our moral deliberations. For instance, we can consistently claim that all persons are moral equals, but that those persons who have committed crimes must be punished, whereas those who have not should not similarly be punished, because ‘committing a crime’ is a relevant fact that must be taken into account when determining what is owed to persons as equals. Analogously, then, holding that humans and nonhuman animals are moral equals is consistent with affirming that they ought to be treated unequally in accordance with the ways in which they differ. But what exactly are the factual differences that justify unequal treatment between human beings and nonhuman animals, and what kind of unequal treatment might they licence? Sher observes that it seems reasonable to hold that the greater mental powers of humans provide us with a plausible explanation for why our duties to humans are not the same as those that we owe to animals. However, what exactly is the difference in the content of these duties will, in turn, depend on the moral outlook—for example, act-consequentialism or contractualism—that a normative theory is based upon. In Chapter 13, ‘Basic Equality, Rational Agency Capacity, and Potentiality’, Richard Arneson elaborates a defence of the Rawlsian range property view, according to which persons are moral equals because they all possess some significant rational agency capacities within a certain range. As observed above, the range property view has been subject to several criticisms: first, it is difficult to determine the minimum degree of rational agency capacities necessary to qualify for personhood in a non-arbitrary way. Second, it looks as if the range property view simply stipulates that the variations above the relevant threshold do not matter, thereby begging the question against the Variation Problem. In response, Arneson notes that the appeal of the range property view— or what he calls the ‘rational agency capacities’ (RAC) account—stems from the fact that it offers a convincing explanation for a wide array of commonsense moral beliefs, such as: the special and unique value of human beings; the unequal moral considerability of nonhuman animals; and the rejection of speciesism. In this sense, the RAC account allows us to reach a ‘reflective
28
Basic Equality: An Analytical Introduction
equilibrium’ between our moral principles and our considered judgements about particular cases. However, Arneson observes that the RAC account runs up against the ‘young childicide’ problem: late-stage foetus and new-born humans lack developed rational agency capacity and therefore are not protected by the deontic constraints associated with the full moral status of personhood. To avoid this disturbing implication, Arneson proposes a revision of the RAC account, which incorporates the potential to develop rational agency capacity, whose possession increases a being’s moral considerability. This, Arneson argues, provides us with the theoretical resources to rule out ‘young childicide’, while at the same time affirming that young children have a moral status that is superior to that of nonhuman animals, who hold some developed rational agency capacities to the same degree but lack the potential for rational agency capacities. The RAC account, then, offers the most plausible defence of the 1(d) Basic Equality Thesis.¹³
References Alvarez, M. 2017. ‘Reasons for Action: Justification, Motivation, Explanation’. In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. https://plato. stanford.edu/cgi-bin/encyclopedia/archinfo.cgi?entry=reasons-just-vs-expl Anderson, E. 1999. ‘What Is the Point of Equality?’. Ethics, 109/2: 287–337. Arendt, H. 2006. On Revolution. New York: Penguin Books. Arneson, R. J. 1999. ‘What (If Anything) Rende.rs All Human Persons Morally Equal?’ In Singer and His Critics, edited by Dale Jamieson, 103–128. Oxford: Blackwell. Arneson, R. J. 2014. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Bedau, H. A. 1967. ‘Egalitarianism and the Idea of Equality’. In Equality, edited by J. Roland Pennock and John W. Chapman, 3–27. New York: Atherton Press. Benn, S. 1967. ‘Egalitarianism and the Equal Consideration of Interests’. In Equality Nomos; 9, edited by J. Roland Pennock and John W. Chapman, 61–78. New York: Atherton Press. Berker, S. 2017. ‘The Unity of Grounding’. Mind, 127/507: 729–777. Berlin, I. 1955. ‘Equality’. Proceedings of the Aristotelian Society, 56: 301–326. ¹³ Both authors contributed equally to the chapter; we have adopted the convention of altering the order in which our names appear on published work. For helpful written comments, we would like to thank George Sher, Julie Tannenbaum, and an anonymous Oxford University Press reader.
Nikolas Kirby and Giacomo Floris
29
Carter, I. 2011. ‘Respect and the Basis of Equality’. Ethics, 121/3: 538–571. Carter, I., and Page, O. 2023. “When Is Equality Basic?” Australasian Journal of Philosophy, 101/4: 983–97. Christiano, T. 2008. The Constitution of Equality: Democratic Authority and its Limits. Oxford: Oxford University Press. Christiano, T. 2014. ‘Rationality, Equal Status, and Egalitarianism’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 53–75. Oxford: Oxford University Press. Cupit, G. 2000. ‘The Basis of Equality’. Philosophy, 75/1: 105–125. Dworkin, R. 1978. Taking Rights Seriously. Cambridge (MA): Harvard University Press. Dworkin, R. 1986. A Matter of Principle. Oxford: Clarendon Press. Dworkin, R. 2000. Sovereign Virtue. Cambridge (MA): Harvard University Press. Fine, K. 2012. ‘Guide to Ground’. In Metaphysical Grounding: Understanding the Structure of Reality, edited by Fabrice Correia and Benjamin Schnieder, 37–80. Cambridge: Cambridge University Press. Floris, G. 2019. ‘On the Basis of Moral Equality: A Rejection of the Relation-First Approach’. Ethical Theory and Moral Practice, 22/1: 237–250. Floris, G. 2021. ‘A Pluralist Account of the Basis of Moral Status’. Philosophical Studies, 178/6: 1859–1877. Floris, G. 2023a. ‘Are Adults and Children One Another’s Moral Equals?’, The Journal of Ethics, 27/1: 31–50. Floris, G. 2023b. ‘Contingency, Arbitrariness, and the Basis of Moral Equality’. Ratio, 36/3: 224–234. Floris, G., and Porro, C. Forthcoming. “The Idea of Equality in Environmental Ethics”. Environmental Ethics. Floris, G., and Spotorno, R. Forthcoming. ‘What Does It Mean to Be Moral Equals? The Special Case of Children’s Moral Status’. Social Theory and Practice. Frankena, W. K. 1962. ‘The Concept of Social Justice’. In Social Justice, edited by Kenneth E. Boulding and Richard B. Brandt, 1–32. Englewood Cliffs: PrenticeHall. Frankfurt, H. G. 2015. On Inequality. Princeton, N.J.: Princeton University Press. Gosepath, S. 2014. ‘On the (Re)Construction and Basic Concepts of the Morality of Equal Respect’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 124–141. Oxford: Oxford University Press. Haksar, Vt. 1979. Equality, Liberty, and Perfectionism. Oxford: Oxford University Press. Hart, H. L. A. 1982. Essays on Bentham: Studies in Jurisprudence and Political Theory. Oxford: Clarendon Press.
30
Basic Equality: An Analytical Introduction
Hart, H. L. A. 1983. Essays in Jurisprudence and Philosophy. Oxford: Clarendon Press. Husi, S. 2017. ‘Why We (Almost Certainly) are Not Moral Equals’. Journal of Ethics, 21/4: 375–401. Jaworska, A., and J. Tannenbaum. 2014. ‘Person-Rearing Relationships as a Key to Higher Moral Status’. Ethics, 124/2: 242–271. Jaworska, A., and Tannenbaum, J. 2015. ‘Who Has the Capacity to Participate as a Rearee in a Person-Rearing Relationship?’. Ethics, 125: 1096–1113. Jaworska, A., and J. Tannenbaum. 2023. ‘The Grounds of Moral Status’. In The Stanford Encyclopedia of Philosophy, edited by E. N. Zalta and U. Nodelman, 1–47. Metaphysics Research Lab: Stanford University. https://plato.stanford.edu/ archives/spr2023/entries/grounds-moral-status/. Kagan, S. 2019. How to Count Animals, More or Less. Oxford: Oxford University Press. Kant, I. 1996. Practical Philosophy. Kant, Immanuel, 1724–1804. Works. English. 1992, edited by Mary J. Gregor and Allen W. Wood. Cambridge: Cambridge University Press. Kant, I. 2002 [1785]. Groundwork for the Metaphysics of Morals. New Haven and London: Yale University Press. Kekes, J. 2003. The Illusions of Egalitarianism. Ithaca, N.Y.; London: Cornell University Press. Kirby, N. 2018. ‘Two Concepts of Basic Equality’. Res Publica, 24/3: 297–318. Kirby, N. 2024. ‘The Problem of Basic Equality: A Constructive Critique’. Journal of Ethics and Social Philosophy, 26/3: 495–522. Korsgaard, C. 2018. Fellow Creatures: Our Obligations to the Other Animals. Oxford: Oxford University Press. Kymlicka, W. 2002. Contemporary Political Philosophy. Oxford: Oxford University Press. Lilburne, J. 1951. ‘The Free-man’s Freedom Vindicated’. In Puritanism and liberty: being the Army debates (1647-9) from the Clarke manuscripts with supplementary documents, edited by A. S. P. Woodhouse, 317. Chicago: University of Illinois Press. Lucas, J. R. 1965. ‘Against Equality’. Philosophy, 40/154: 296–307. Macdonald, M. 1946. ‘Natural Rights’. Proceedings of the Aristotelian Society, 47: 225–250. McLaughlin, B. P. 1995. ‘Varieties of Supervenience’, 16–59. Cambridge University Press. McMahan, J. 1996. ‘Cognitive Disability, Misfortune, and Justice’. Philosophy & Public Affairs, 25/1: 3–35.
Nikolas Kirby and Giacomo Floris
31
McMahan, J. 2003. The Ethics of Killing: Problems at the Margins of Life. Oxford; New York: Oxford University Press. Miklosi, Z. 2022. ‘The Problem of Equal Moral Status’. Politics, Philosophy & Economics, 21/4: 372–392. Miller, D. 2007. National Responsibility and Global Justice. Oxford University Press. Morgan, G. 1943. ‘Human Equality’. Ethics, 53/2: 115–120. Nagel, T. 1991. Equality and Partiality. Oxford: Oxford University Press. Nathan, C. 2014. ‘What is Basic Equality?’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 1–16. Oxford: Oxford University Press. Phillips, A. 2021. Unconditional Equals. Princeton: Princeton University Press. Pojman, L., and R. Westmoreland. 1997. Equality: Selected Readings. New York; Oxford: Oxford University Press. Pojman, L. P. 1991. ‘A Critique of Contemporary Egalitarianism: A Christian Perspective’. Faith and Philosophy, 8/4: 481. Pojman, L. P. 1997. ‘On Equal Human Worth: A Critique of Contemporary Egalitarianism’. In Equality: Selected Readings, edited by Louis P. Pojman and Robert Westmoreland, 282–299. New York; Oxford: Oxford University Press. Rawls, J. 1971. A Theory of Justice. Cambridge (MA): Harvard University Press. Rawls, J. 1999. A Theory of Justice. Rev. ed. Cambridge (MA): Belknap Press of Harvard University Press. Raz, J. 1986. The Morality of Freedom. Oxford: Clarendon Press. Regan, T. 1983. The Case for Animal Rights. Berkeley and Los Angeles: University of California Press. Sangiovanni, A. 2017. Humanity without Dignity: Moral Equality, Respect, and Human Rights. Cambridge, MA: Harvard University Press. Scanlon, T. M. 2018. Why Does Inequality Matter? Oxford: Oxford University Press. Schaar, J. H. 1964. ‘Some Ways of Thinking about Equality’. Journal of Politics, 26/4: 867–895. Scheffler, S. 2003. ‘What Is Egalitarianism?’ Philosophy & Public Affairs, 31/1: 5–39. Schweitzer, A. 1929. Civilization and Ethics: The Philosophy of Civilization Part II. London: A. & C. Black. Sen, A. 1979. Equality of What? The Tanner Lecture on Human Values. Stanford University. Sher, G. 2014. ‘Why We Are Moral Equals’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 17–29. Oxford: Oxford University Press. Sher, G. 2017. Me, You, Us. Oxford: Oxford University Press. Singer, P. 2011. Practical Ethics. Cambridge: Cambridge University Press.
32
Basic Equality: An Analytical Introduction
Spiegelberg, H. 1944. ‘A Defense of Human Equality’. The Philosophical Review, 53: 101–124. Steinhoff, U. 2014. ‘Against Equal Respect and Concern, Equal Rights, and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 142–172. Oxford: Oxford University Press. Taylor, P. W. 2011 [1986]. Respect for Nature. A Theory of Environmental Ethics. Princeton and Oxford: Princeton University Press. Thomas, D. A. L. 1979. ‘Equality Within the Limits of Reason Alone’. Mind, 88/352: 538–553. Vlastos, G. 1984. ‘Justice and Equality’. In Theories of Rights, edited by Jeremy Waldron, 41–76. Oxford: Oxford University Press. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, Massachusetts: The Belknap Press of Harvard University Press. Westen, P. 2014. Speaking of Equality: An Analysis of the Rhetorical Force of ‘Equality’ in Moral and Legal Discourse. Princeton, N.J.: Princeton University Press. Williams, B. (1973). Problems of the Self. Philosophical Papers 1956–1972. Cambridge: Cambridge University Press. Williams, B. 2006. Philosophy as a Humanistic Discipline. Princeton: Princeton University Press. Wilson, J. 1966. Equality: Philosophy at Work. London: Hutchinson. Wolff, J. 1998. ‘Fairness, Respect, and the Egalitarian Ethos’. Philosophy & Public Affairs, 27/2: 97–122. Wollheim, R. 1955. ‘Equality’. Proceedings of the Aristotelian Society, 56: 281–300. Wollner, G. 2014. ‘Basic Equality and the Currency of Egalitarian Justice’. In Distributive Justice and Access to Advantage: G. A. Cohen’s Egalitarianism, edited by A. Kaufman, 180–203. Cambridge: Cambridge University Press.
PART I
MEANING History and Critique
1 On the Historical Emergence of Basic Human Equality Teresa M. Bejan
1. Introduction The basic equality of human beings has long functioned as an ‘underlying major premise’ of moral and political philosophy (Waldron 2017: 13). The American Declaration of Independence (1776) proclaimed it to be simply ‘self-evident’ that ‘all men are created equal’. Two centuries later, Ronald Dworkin (1978: xii) likewise described people’s right to equality as ‘axiomatic’.¹ The long-standing language of Euclidean geometry deployed in these discussions suggests that while equality’s implications for our social, political, and economic arrangements may be up for debate, the claim that every human person is somehow ‘equal’ is not.² Basic human equality (hereafter, BHE) has, nonetheless, recently come under increased philosophical scrutiny from both critics and defenders (for example, Steinhoff 2015; Kirby 2018; Phillips 2021; Waldron 2017).³ Commentators on all sides, however, have tended to leave several of Dworkin’s governing assumptions intact. Firstly, they assume that BHE is, in some sense, a relation of sameness according to which every human being enjoys exactly the same fundamental status and/or intrinsic worth, despite their many differences (for example, Dworkin 2000: 126; cf. Phillips 2021: 93). Secondly and relatedly, most political philosophers assume that BHE (whatever its basis) must therefore operate as a normatively egalitarian principle, in some sense opposed to hierarchical relations of superiority and
¹ See also Barry (1999): 96–97. ² For the purposes of this chapter, I set aside debates over whether basic equality (BE) pertains to human beings as such, or only to ‘persons’ as independent sites of moral agency, and thus also whether it might or might not include nonhuman animals. I shall treat instead of BHE, as a claim to basic human equality, in keeping with my historical sources. ³ These accounts build upon and/or respond to earlier objections pressed by J. R. Lucas (1965; 1977), Bernard Williams (1973), Peter Westen (1982), and Louis Pojman (1997), among others. Teresa M. Bejan, On the Historical Emergence of Basic Human Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0002
36
On the Historical Emergence of Basic Human Equality
inferiority.⁴ In other words, those who are ‘equal’ or the same with respect to their fundamental status or worth should, as a matter of justice, be treated ‘as equals’, too, enjoying the same political and/or social status, rights, and respect. Thirdly and finally, many philosophers follow Dworkin by treating BHE as a definitively modern idea based on a more or less simple story about its historical emergence. As per Richard Arneson (2015): ‘Claims about basic human equality … mark a divide in moral thinking between a premodern world in which nobles are regarded as having greater worth than peasants … and a modern world that repudiates these crude prejudices’ (30). More historically minded political philosophers point to early modern England as the pivotal period and place in which the equality of human beings became widely accepted as a premise of political theorizing (for example, Anderson 2017: 1–36; Pettit 2014: 3; Waldron 2002: 94). Versions of this history are ubiquitous, but they are also false.⁵ The idea that all humans are somehow ‘equal’ is, in fact, an ancient one. For example, the sixth-century CE Digest of Roman Law compiled under the emperor Justinian includes a third-century extract referring to an even earlier text according to which omnes homines aequales sunt (‘all humans are equal’). From Roman jurisprudence this claim made its way into Christian theology and medieval theories of natural law. It seems that by the seventeenth century in Europe, human equality was not a new idea seeking acceptance, but rather a legal, theological, and political commonplace—one seen for centuries as consistent with hierarchies of all sorts (Bejan 2019 and 2022; Hoekstra 2013). That philosophers make poor or partial historians is not surprising. Still, in what follows I argue that the antiquity of assertions of human equality presents a challenge not only to BHE’s presumptive modernity, but to philosophical assumptions about its association with sameness and its normative egalitarianism as well. To this end, this chapter will explore the political ideals of isotes and aequalitas in democratic Athens and republican Rome, respectively, before turning to discussions of human equality in Roman legal and Christian theological traditions. I conclude my historical survey by taking a closer look at the seventeenth-century English Levellers identified by contemporary political philosophers as the preeminent group of ‘early egalitarians’. ⁴ See, for example, Anderson (2017), Waldron (2017), and Kolodny (2022), as well as the exchange between Angle et al. (2017) and Pettit (2017). On the precise relationship between BHE and social or relational equality, see Waldron’s chapter in this volume. ⁵ As Phillips (2021: 40) reminds us, modernity has produced plenty of its own ‘crude prejudices’— including forms of scientific racism and sexism alien to premodern eras.
Teresa M. Bejan
37
While the following survey is far from exhaustive,⁶ it should establish three things: (1) While philosophers understand BHE today primarily as a relationship of quantitative identity or sameness, the word ‘equality’ and its cognates have been applied to many different, often competing, conceptual relations.⁷ I highlight three historically significant equalityrelations below—equality-as-balance, equality-as-indifference, and equality-as-parity. (2) Because of this plurality of equality-relations, scholars’ tendency to dismiss the many hierarchies and exclusions—of gender, class, and race, for example—seen throughout history as consistent with human equality as ‘blind spots’ in otherwise normatively egalitarian theories is inadequate; (3) Moreover, distinguishing between these different ideas of equality helps to explain: (a) Why human equality operated for so much of its history as a normatively hierarchical principle; as well as (b) How it began to have identifiably ‘egalitarian’ consequences in early modern England—as, for instance, through the Levellers’ peculiar understanding of equality-as-parity, according to which all humans were naturally ‘Peers’ on the model of English aristocrats. By way of conclusion, I consider the upshot of this history for debates about BHE in contemporary moral and political philosophy. In particular, I suggest that conceiving of BHE as a matter of human beings’ fundamental parity—as opposed to their sameness or indifference—illuminates the so-called gradability objection pressed by its contemporary critics.
2. High-Road Histories Let’s take a closer look at the just-so stories about equality embraced by philosophers. Today, versions of the claim derived from Dworkin (1983: 25, 31) that all modern political theories occupy the ‘egalitarian ⁶ Stuurman (2017) rightly challenges philosophers’ exclusive focus on the history of Western political thought with respect to equality. I do not attempt to rectify it here, however, because I am interested in the history of a Latinate word, ‘equality’, with distinctive mathematical and juridical resonances, as attached to the idea that every human being shares some basic status or intrinsic worth. ⁷ Here, I am in sympathy with Kirby (2018), who distinguishes between BHE understood as ‘equal worth’ or ‘equal authority’. Both of Kirby’s conceptions, however, assume that ‘equality’ is itself a relation of sameness with respect to either property.
38
On the Historical Emergence of Basic Human Equality
plateau’ with respect to their premises abound (for example, Bidadanure 2021: 15; Gosepath 2021; Kymlicka 1990: 4; Swift 2006: 93; Scanlon 2018: 4). Some scholars lend this assertion historical texture by quoting the Leveller sympathizer, Col. Thomas Rainborough, at the Putney Debates in 1647— ‘Really, I think that the poorest he that is in England hath a life to live as the greatest he’ (Christiano 2002: 32; Waldron 2002: 122; White 2007: 25).⁸ Notice that what may, at first, seem like historical window-dressing, in fact, does normative work. Acknowledging BHE as a historical achievement saves philosophers the trouble of arguing for it, while also allowing them to take the wrongness of hierarchical institutions like patriarchy or slavery for granted. But appealing to history in this way can also cause problems. When confronted with the myriad sexist, classist, and racist positions held by historical thinkers otherwise committed to human equality, philosophers tend to dismiss these as ‘blind spots’—that is, as philosophically uninteresting failures of reasoning or nerve that should not distract us from equality’s essentially egalitarian implications (Anderson 2017: 16, 151; Rosanvallon 2013: 71). Some philosophers do, of course, acknowledge that equality did not always have egalitarian consequences, particularly in religious contexts. Waldron (2002; 2012a; 2017: 16–18), for instance, recognizes that Locke’s ‘egalitarian’ arguments in the seventeenth century had deep roots in Jewish and Christian theology—specifically, in the creation of men and women in the ‘image of God’ (Gen. 1:26–7). But this divine resemblance evidently did not preclude the equally divinely sanctioned institutions of slavery and patriarchy found elsewhere in the Bible.⁹ Other philosophers blame Christians’ historical failure to understand the egalitarian implications of their beliefs on other doctrines—for example, ‘the authoritarian doctrine of Original Sin’ (Anderson 2017: 13). Samuel Fleischacker (2004) suggests that many religious assumptions ‘will block the inference from [human] equality in principle … to a presumption that efforts should be made to equalize people political, socially, or economically’, including the belief that poverty is punishment for sin, or an ascetic disdain for the material world (9–10).¹⁰ That philosophers continue to view human equality’s failure to produce egalitarian consequences as incidental, rather than indicative of deeper conceptual difficulties, reflects a broader failure of historicism. The term ⁸ For Rainborough’s intervention in the Putney Debates, see ‘Extract from the Debates at the General Council of the Army, Putney. 29 October 1647’ (1998: 122). ⁹ Waldron (2002) describes Locke’s own much-debated support for patriarchal marriage as an ‘inconsistency’ that does not materially impact the ‘egalitarianism’ of his political theory (31–32). ¹⁰ See also Kirby (2018: 300).
Teresa M. Bejan
39
‘egalitarian’ is itself a late nineteenth-century coinage.¹¹ Nevertheless, scholars deploy it routinely today as an anachronistic and strategically vague evaluative-descriptive term. One might expect more from intellectual historians. Yet in tracing the origins of human equality even earlier—for instance, to early Christianity (Holland 2019; Siedentop 2014), Judaism (Berman 2011), or the Axial Age (Stuurman 2017)—they also suggest that its intrinsic ‘equality effects’ would or should have prevailed, had not some contrary prejudice or atavistic impulse intervened (Stuurman 2017: 7–8). Whether purveyed by philosophers or historians, these high-road histories assume that people in the distant past must have meant what we do when we say that human beings are ‘equal’—namely, that they are the same in fundamental status and/or worth and should be treated accordingly. A more contextual approach reveals that this was not the case. ‘Equality’, as we shall see, could describe many different possible relations, and because of this its relationship with hierarchy was (and is) much more complicated than philosophers generally allow.
3. Political Equality: Balance and Indifference To see this, let us begin with an ancient philosophical tradition preoccupied with political equality from which a commitment to human equality is conspicuously absent. In ancient Greek, the idea of to ison (as ‘the equal’, ‘even’, or ‘level’) derived from an ancient technology, the balance-beam scale, occupying a central position in the agora or marketplace. Accordingly, the Greek word for equality (isotes) placed one firmly in the realm of economic exchange, recalling the practice of ‘balancing’ goods, as well as the precious metals used in payment, against a political community’s standard weights and measures. In Greek, ‘the equal’ (to ison) was thus related to, but distinct from, ‘the same’ (to tauton) or ‘the like’ (to homoion). It implied the possibility of measurement, as well as the exigencies of evaluation and exchange. The distinctiveness of equality is evident in Aristotle’s Metaphysics, which contrasted isotes—as a relation of oneness with respect to magnitude—from sameness and likeness as relations of oneness with respect to essence and to quality, respectively (1933: 1021a).¹² Likewise, Euclid’s Elements presented equality as a specifically quantitative relationship between different objects sharing a unit of measurement in contrast with relations of similarity or ¹¹ See ‘egalitarian, adj. and n.’ (2022). ¹² See also White (1971).
40
On the Historical Emergence of Basic Human Equality
sameness (for example, 1.Prop.35–36, 6.Def.1, 1.Prop.35).¹³ Objects were therefore unequal (anisotes) when they differed in some measurable degree, as either greater or lesser, whereas equal things were strictly indifferent. The quantitative conception of what I call equality-as-indifference found in these Greek sources may sound a lot to modern audiences like sameness with respect to some scalar property. But again, commentators like Aristotle and Euclid took care to distinguish ‘the equal’ from ‘the same’ or ‘the identical’ in their discussions. ‘Indifference’ better captures the distinctiveness of equality as described by Aristotle elsewhere in the Metaphysics (1933: 1056a) as a kind of absence or lack—what he calls ‘contradiction by privation’: ‘The equal, then, is that which is neither great nor small, but would naturally be either great or small; and it is opposed to both as a privative negation’. On this view, there was something exceptional—and indeed, artificial—about equality, as the absence of difference where one would normally expect to find it.¹⁴ The distinctively privative nature of equality-as-indifference, I suggest, reflects the continuing conceptual priority of equality-as-balance for Greek authors. Consider, again, the balance-beam scale. Assessing whether goods or unminted specie were of equal weight in the agora required a delicate operation of contrastive comparison, that is, counterbalancing, until the pans of the scale were even or level. The pans of an ‘equal’ (balanced) scale contained objects of ‘equal’ (indifferent) weight. But practically speaking, the evenness of the parts (the pans) was assessed with reference to the balance of the whole (the scale itself ), through careful calibration and counterbalancing. The practices of balancing and exchange in the agora also, I think, help us make sense of the close association in Greek, as in Latin, between equality and justice. While Aristotle (1932: 1280a) noted in the Politics that ‘the equal’ (to ison) was synonymous with ‘the just’ (to dikaion), his Nicomachean Ethics echoed Plato (1926: 6.757b–c) by distinguishing between two, different kinds of equality (Aristotle 1926: 1131b–1132a).¹⁵ Aristotle dubbed these influentially ‘proportional’ (or ‘geometric’) and ‘arithmetic’, respectively, and associated each with ‘distributive’ and ‘corrective’ (or commutative) justice, in turn. While the latter implied quantitative indifference in restoring like-for-like, regardless of the parties’ relative worth (axia), the ¹³ I am grateful to Jinxue Chen for drawing my attention to these various discussions. ¹⁴ This privative sense of equality was preserved in Samuel Johnson’s Dictionary of the English Language, which defined ‘equal’ inter alia as ‘neither greater nor less; neither worse nor better’, ‘indifferent’. ¹⁵ See also Fleischacker (2004: 19–28).
Teresa M. Bejan
41
former connoted a decidedly evaluative assessment of proportionality that distributed ‘equal’ things only to those who were themselves ‘equals’ (isoi) with respect to worth. Modern commentators sometimes dismiss Aristotle’s distinction as sophistical, reflecting his hostility to the democratic argument that Athenian citizens (and only Athenian citizens) were isoi by birth (Vlastos 1997; cf. Frank 2005). Yet to understand his distinction, it is important to keep the priority of balance and the model of economic exchange in mind. One might assume that objects of the same or indifferent weight would be of equal value, but this was incorrect. Determining a just exchange, in fact, required two, successive measurements—that of the goods to be sold and that of the precious metals to be exchanged, both of which must be assessed through careful counterbalancing against the standard weights issued by a polis. Aristotle and other aristocratic commentators could therefore observe that, yes, justice was a kind of equality. But distributive justice involved four terms, at least, hence two, successive acts of balancing. Like metals in the marketplace, men too must be weighed. Here, however, a democrat might reasonably point out that another, somewhat less ancient technology—that of coinage—had simplified matters by substituting the weighing of metals with the counting of coins, the value of which had been predetermined by political authority. Why shouldn’t a democracy likewise insist that its citizens were of strictly equal (indifferent) weight or value, and so—like the Athenian silver ‘owl’ coins—be counted instead of weighed (Kurke 1999: Ch. 8)? Indeed, Aristotle was theorizing isotes in the context of an Athenian democracy in which several iso-prefix terms operated in precisely this way. The institutions of isonomia and isegoria, for example, treated all adult male citizens as isoi by affording them equal access to the law-courts and assembly, and counting their votes (and voices) as of strictly equal weight therein, despite their sometimes extreme differences of wealth and social status (Lombardini 2013; Ober 2017: 18–33). Today, philosophers and historians often present these institutions anachronistically as egalitarian ones granting equal citizens the same—or strictly equal—political dignity and rights (Christiano 2002; Ober 2012). Some assume, therefore, that a belief in BHE must be lurking in the background, making Athens’ political exclusion of Athenian women, non-citizens, and slaves—who were, after all, ‘the same’ with respect to their humanity—a matter of ‘historical baggage’ or bigotry (Ober 2017: 33). That Aristotle himself did not subscribe to BHE is obvious; he argued that the rule of superior
42
On the Historical Emergence of Basic Human Equality
over inferior (including master over slave and male over female) was not only just, but natural (1932: 1254a). But one need not take Aristotle’s views on slavery and the sexes as representative to see that Athenian isonomia and isegoria also operated on an assumption of basic human inequality—that is, of the differential worth and weight of particular human beings and citizens, be they male or female, rich or poor, well-born or ‘base’. To see this, consider that democratic Athens regarded all adults of Athenian parentage, men and women, as citizens (polites), but counted as isoi only adult men over the age of 18 who were formally enrolled in the citizenbody. These men’s status as isoi had been ‘proofed’ or tested ceremonially on the basis not only of their birth, but their prospective military service, in a process called dokimasia linked explicitly with the regulation of domestic coinage (Kurke 1999: 309–310). By contrast, Athenian women, as well as male citizens convicted of sex-crimes and bribery, were regarded as unworthy of being assigned political weight. In the case of isegoria, especially, the comparative weightiness of men’s voices, even the poorest (who in Athens served militarily by rowing in the naval fleet), relied heavily on the weightlessness of women’s (Beard 2014; Bejan 2020). Difference, again, was the norm. Equality, a carefully curated exception. While it may be tempting, then, to read the disagreement between Athenian democrats and their aristocratic critics as arising from conflicting conceptions of equality, with the former favouring ‘arithmetic’ indifference and the latter prioritizing ‘geometric’ proportion, Aristotle insisted rightly that this was not the case. The democracy’s constitutive exclusions revealed that an overarching, evaluative conception of equality-as-balance remained normative for them, too. Democrats regarded adult male Athenian citizensoldiers in good standing as so far superior, so much ‘weightier’, than other people that they adopted a peculiar balancing strategy of ‘counting’ them like coins, on the model of avowedly aristocratic bodies like the Athenian Areopagus or the Spartan Gerousia or senate (Schwartzberg 2013: 19–48). Hannah Arendt (2006) was thus right to characterize Athenian institutions like isonomia as expressing a Greek idea of political equality as an artificial and exclusionary convention adopted among ‘those who form a body of peers’ (30). Accordingly, these institutions did not reflect some underlying moral or metaphysical commitment to the basic equality of human beings or of Athenians. Rather, like coins, adult male Athenians’ value as isoi was exclusive and backed by political authority, like the other conventional weights and measures employed in the agora. This, in turn, explains why critics like Plato and Aristotle could criticize democratic institutions so effectively as
Teresa M. Bejan
43
unbalanced, arguing that the differences of wealth, birth, virtue, and ability that they were designed to obscure were not, in fact, matters of indifference.¹⁶ Politically speaking, then, in ancient Greece ‘equal justice’ remained a question of who should be ‘weighed’ or ‘counted’ politically—as well as when, where, and how much—in order to balance the whole. Given this, we should not be surprised to find that democratic institutions like isonomia and isegoria also rejected BHE, operating instead as artificial instruments of human ingenuity designed to ‘equalize’ a polity’s unequal parts, particularly with respect to the relative power of rich and poor. Tellingly, isonomia carried similar connotations in Greek cosmology and medicine, in which the essentially different or unequal elements of the universe or parts of the body must be balanced, lest one overwhelm the rest (Lombardini 2013: 414).¹⁷ Turning now from Greece to Rome, we find a similar set of conceptual associations. In classical Latin, the abstract noun aequalitas (from which we get the modern English word ‘equality’) derived from the verb aequo, meaning to balance, level, or make even.¹⁸ The operative image, again, was that of a ‘level’ set of scales or libra evenly counterbalanced between two weights. The conceptual priority of balance in Roman thought, as in Greek, helps to explain equality’s close association with equity (aequitas) and the scales of justice. Beyond the law, aequalitas was also celebrated as an ideal of ‘balanced’ harmony in the fields of music, cosmology, and medicine (Kaye 2014). Unsurprisingly, we also find similar ideals of equality-as-balance being applied politically in analyses of the Roman constitution. For instance, the second-century-BCE Greek historian and prisoner-of-war Polybius described the Roman republic influentially as a mixed regime, in which the powers of the consuls, the senate, and the tribunate—the last of which represented the Roman people—were so equally balanced ‘that no one could say for certain … whether the constitution as a whole were an aristocracy or democracy or ¹⁶ See, for example, Plato’s (2013: 561c–e) discussion of the ‘life of some isonomic man’ (bion isonomikou tinos andros): ‘And does he not … also live out his life in this fashion, day by day indulging the appetite of the day, now wine-bibbing and abandoning himself to lascivious pleasing of the flute, and again of drinking only water and dieting; and at one time exercising his body, and sometimes idling and neglecting all things, and at another time seeming to occupy himself with philosophy. And frequently he goes in for politics and bounces up and says and does whatever enters his head … and there is no order or compulsion in his existence, but he calls this life of his the life of pleasure and freedom and happiness and cleaves to it to the end’. Whether his desires are noble or base, healthy or harmful, the isonomic man thus ‘avers are all alike and to be esteemed equally’—that is, he counts them strictly equally, without weighing their relative merits. ¹⁷ Cf. Vlastos (1947). ¹⁸ ‘Equality, n.’ (2019). Like Greek, classical Latin also distinguished things that were aequalis from those that were idem (the same) or similis (similar) in other respects.
44
On the Historical Emergence of Basic Human Equality
despotism’ (Polybius 2011: 6.15–17). Here again, the idea was that of different or unequal parts of a polity being ‘equalized’ through ingenious constitutional design, ensuring that no one part of the polity could over-balance or dominate another one. In Roman, and especially republican, political thought, however, equality took a backseat to liberty as the governing ideal. As Malcolm Schofield (2021: 27–60) has shown, Roman writers invoked aequalitas primarily in describing how the liberty (libertas) and rights (iure) of the two chief social orders (ordines) of Rome—that is, the patrician and plebeian classes—had been ‘equalized’ through successive constitutional reforms. Livy (1922: 3.31.7), for example, described how successive institutional innovations produced ‘equal liberty’ by counterbalancing the political power of the plebs against the patricians, so that neither party might dominate the other. As in Athens, however, ‘elitist’ and ‘popular’ republicans disagreed on the relative political weight to be assigned to rich and poor citizens in order to achieve this balance.¹⁹ The latter argued that, in certain contexts, all adult male citizens (again, potential soldiers) resident in Rome should be counted as of indifferent weight (one man one vote), despite their many differences in wealth and social status. While the rules changed over time, certain popular assemblies (which organized male citizens on the basis of their ethnic, kinship, or military ties) did, in theory, give each resident a strictly equal vote within his sub-group; however, the complexities of Roman voting meant that outcomes were generally decided before every vote was cast, let alone counted.²⁰ On the other side, Cicero’s De Republica drew implicitly on Aristotle to argue the aristocratic line: ‘A particularly important kind of equality’ (aequabilitas) depended on differential rights and responsibilities being assigned to different parts of the polity as a means of balancing the whole (1928: I.69). In other words, there should be ‘something preeminent and regal in res publica, something [else] bestowed and assigned to the authority of its leading men, and [also] certain matters … kept for the judgment and will of the mass of the people’ (Cicero 1928: I.69).²¹ There was one domain, however, in which even elitists like Cicero regarded equality-as-indifference as the right principle in balancing the Roman body politic: the law. This brings us, at last, to the striking statement of human equality (omnes homines aequales sunt) preserved in Justinian’s Digest and
¹⁹ I borrow this distinction from Glover (1999). ²⁰ For aequalitas and Roman voting rights, see Arena (2012: 45–72). ²¹ For analysis, see Schofield (2021: 44–45).
Teresa M. Bejan
45
attributed to Ulpian in his commentary on the first-century jurist Sabinus. To understand its meaning, we must consider the passage in its entirety: As far as concerns the civil law, slaves are regarded as nothing [servi pro nullis habentur], not, however, in the natural law, because as far as concerns the natural law, all men are equal [quod ad ius naturale attinet, omnes homines aequales sunt].22
In context, this is clearly not the sort of egalitarian premise modern philosophers expect. Neither Ulpian nor Sabinus were claiming that those who were naturally ‘equal’ deserved to be treated ‘as equals’ legally, let alone socially or politically. Rather, they were highlighting the universality of human subjection to law, while also emphasizing individuals’ differential status and obligations with respect to different kinds of law. For instance, natural law obligated every human being to preserve his- or herself, but civil law permitted masters to beat their slaves and not vice versa. Here, we are firmly in the realm of equality-as-indifference. To say that omnes homines aequales sunt was to observe that, insofar as men and women belonged equally to one natural species (that is, they were equally human), they were also equal with respect to natural law, which weighed upon and obligated each indifferently. But Ulpian and Sabinus contrasted the naturally indifferent status of human beings as subject to natural law, both directly and indirectly, with the many distinctions of status respected in Rome according to the civil law of persons. These included not only the formal distinction between free men (liberi), freedmen (libertini), and slaves (servi), but also between those who were sui iuris—for example, Roman fathers (patres familiae) and those alieni iuris (for example, women, slaves, and children, including a Roman patriarch’s adult sons).²³ Here, perhaps, the distinction between natural and civil law is sufficient to explain why Romans regarded the natural equality of human beings as having few, if any, social or political consequences. But there was another, crucial respect in which jurists believed that the civil law did, in fact, treat some humans (that is, Roman citizens) ‘equally’ or indifferently—namely, through its codification (Schofield 2021: 44–45; see also Schiavone 2022). According to Cicero, the simple act of writing it down meant that the civil law bore equally on every citizen, much like death and other unavoidable features of the human condition. Each would, of course, still be assigned differential ²² My own translation, adapted from Watson (1985: 50.17.32). ²³ As Clifford Ando (2010: 194–195) explains in criticizing Skinner’s (2012) influential analysis of de statu hominis.
46
On the Historical Emergence of Basic Human Equality
rights and duties according to his or her personal status. Nevertheless, the law itself remain unchanged, thereby preserving the aequalitas of balanced judges, who (like the goddess Justitia) should decide every case impartially, unmoved by the power of the particular individuals involved.
4. Human Equality: Aequales and Pares We shall encounter this Roman ideal of aequalitas again, as a virtue of balanced judges holding powerful individuals to account, among the early Christians. But first we must introduce another concept informing Roman discussions of human equality: paritas, or parity. As a relational property, parity seems to have operated in Roman thought specifically in the domain of value. The Latin adjectives par or paris (like the Greek to ison) may have originated in trade to describe different objects that were sufficiently similar in relevant respect(s) to be a ‘pair’ or matching, hence subject to exchange.²⁴ But more generally, the phrase par esse, or ‘to be par’, conveyed a sense of something being ‘fit’ or ‘meet’. Thus things adjudged to be pares might go together, or be treated ‘on a par’; likewise, persons might be pares if they were ‘matching’ with respect to some shared experience or valuable trait, for instance honour or virtue.²⁵ For instance, a stock description of Roman citizen-soldiers described them as aequos et pares—that is, as equals and peers.²⁶ In classical and post-classical Latin, paritas thus emerged alongside aequalitas as an overlapping but distinct concept. The distinction is evident, for instance, in Cicero’s dialogue, De Senectute (‘Of Old Age’), which distinguished between those who were aequales in age and those were pares in consular rank and therefore socialized most easily together (1923: sec. 7). This discussion suggests that aequalitas remained in Latin, as in Greek, a largely quantitative concept, and paritas an evaluative one. Aequales were not necessarily pares, nor indeed vice versa. To see this, consider the Latin title primus inter pares applied to the senior member of the Roman Senate. His unequal status granted him the privilege of speaking first in debate, but this inequality did not threaten his fellow Senators’ standing as peers. (The primus inter pares title was subsequently assumed, less plausibly, by the Emperor Augustus and his successors.) ²⁴ See ‘pār, păris’ (1879). ²⁵ It is in this elevated, social sense that the Latin pares later entered into English, via medieval French, as the word ‘peer’. See ‘peer, n. and adj.’ (2022). ²⁶ See, for example, Cicero (1913: I.124).
Teresa M. Bejan
47
The difference between equality and parity may be unintuitive to modern English speakers, for whom the nouns ‘peer’ and ‘equal’ often function as synonyms to describe people who enjoy the same status or distinction. Because of this, the phrase primus inter pares is routinely translated into English as ‘first among equals’. But this obscures the persistence of the conceptual distinction highlighted by Cicero, according to which pares or peers were evaluated positively and thus enjoyed comparatively high and exclusive status relative to other people. And indeed, this sense persists in modern English, as for instance in Arendt’s description of Athenian isoi as a ‘body of peers’. Likewise, Roman men’s privileged status as aequos et pares reflected not only their indifferent status with respect to the civil law and their political rights, but also their distinctive (and indeed superior) virtue and value relative to the rest of humanity, including Roman women, slaves, foreigners, and an ever-expanding array of subject peoples. Under the republic and early empire, for instance, Roman citizens enjoyed a privileged status with respect to the law, according to which they were spared the corporal punishments (including crucifixion) applied to foreigners and slaves (Whitman 2005: 19–40). Evidently, the idea that human beings, as such, were aequales by nature existed happily alongside the idea that particular people were also disparata, hence fit for hierarchical relations of superiority and inferiority. Roman citizen-soldiers’ distinctive status as pares evidently entitled them to dominate those who were naturally their equals. Here, the conceptual distinction between equality and parity helps to resolve what might otherwise appear as a puzzle: namely, how the idea of human equality could emerge (and, indeed, gain currency) within a Roman Empire characterized by increasingly complex and highly formalized social hierarchies.²⁷ Indeed, as Darrin McMahon (2023, 131–132) points out, Roman jurists’ insistence that omnes homines aequales sunt went hand in hand with the expansion of citizenship beyond the city of Rome, first to its allies and then, with the Edict of Caracalla in 212 CE, to all free adult men within the Empire (Ch. 5). In this way, human beings’ indifference with respect to natural law paved the way for their universal subjection to the Roman civil law, as well, through conquest. But just as the notion that the emperor was merely primus among his Senatorial pares became strained past all plausibility, this expansion of citizenship fatally undermined any sense in which Rome’s increasingly socially, geographically, and ethnically distant citizens could be regarded as pares ²⁷ The basic republican distinction between the patrician and plebeian orders existed alongside finer, socioeconomic distinctions (e.g. between senators, consuls, equestrians, etc).
48
On the Historical Emergence of Basic Human Equality
as well as aequales. This tension is illustrated by the roughly contemporaneous introduction of a new socio-legal distinction between low-status citizens (humiliores) and high-status ones (honestiores), according to which the former became vulnerable to forms of public humiliation and corporal punishments limited previously to non-citizens (McMahon 2023, 132). Evidently, their equal status of subjection to law remained compatible with the disparity (hence hierarchical ordering) of different human beings and citizens. It is primarily as indifference, then, that we find human equality entering into Christianity, which (after the civil law) became Rome’s most important cultural export. While Waldron and other philosophers tend to focus on men and women’s creation imago Dei in Genesis as the theological foundation of BHE, an arguably more important proof text comes from the Acts of the Apostles, in which Peter tells a Roman centurion that ‘God is no respecter of persons’ (Acts 10:34, KJV ). The Apostle’s point seems to have been that one’s acceptability to the Christian God did not depend on one’s ethnicity nor one’s personal status according to the civil law, but on one’s spiritual worth. While human judges might be easily distracted by superficial differences, hence prone to partiality, divine aequalitas was superior. It regarded individual human beings as truly equal, treating their many differences of wealth, ethnicity, and social status as matters of indifference and subjecting all alike to God’s ultimate judgement, as the Roman law of persons did not.²⁸ Evidently, early Christians shared the Roman assumption that those who were naturally equal were not necessarily pares or peers with respect to righteousness. For example, the third-century Roman-Christian author Lactantius (celebrated during the Renaissance as the ‘Christian Cicero’) redeployed the aequos et pares idea to denote the community of Christians who, by accepting all sorts of people, including slaves and women, surpassed others as peers in humility and so enjoyed ‘a much higher rank of dignity in the judgment of God’ (Lactantius 2003: V.14–15). This seems in keeping with the only reference to man’s creation in God’s image in the New Testament: Jesus’s command that the Pharisees ‘render unto Caesar that which is Caesar’s; and unto God the things that are God’s’ (Matt. 22:19–21). There, Jesus likens the ‘image’ of God in man to the Emperor’s imprint on a Roman coin.²⁹ In contrast with the coins of a particular civitas (like the Athenian ‘owls’ or Roman denarii), all humans bore the same stamp and were equally subject to their Creator. Later Christian commentators would emphasize, however, ²⁸ For analysis, see Bejan (2021). Compare the early Christian criticisms of slavery presented in Garnsey (1995: 79–85). ²⁹ The fourth-century Latin Vulgate specifies this as a denarius or silver coin.
Teresa M. Bejan
49
that individuals’ superficial indifference did not necessarily entail their parity with respect to value, just as Caesar’s image might be impressed on coined metals of variable quality (Augustine 1847: 4.8; Aquinas, ‘Psalm 4’).³⁰ In the sixth century, Pope Gregory the Great would explain the connection between human equality and disparity with reference to a familiar image, the balanced scale. His commentary on the Book of Job declared, ‘Let [God] weigh me in an even balance’: For all of us men are equal by nature (omnes namque homines natura aequales sumus), but it has been added by a distributive arrangement, that we should appear as set over particular persons … but the very diversity, which has been added from defect, is rightly ordered by the judgments of God. (Gregory I 1845: 21.5–6, 21.14–15)
In Gregory’s description of God’s aequalitas, one can see the characteristically Roman combination of balance, indifference, and disparity: He to whom all are ‘equally’ (indifferently) subject would judge them ‘equally’ or impartially, and so balance the scales in turn. The appeal to ‘defect’ in this passage further suggests that the unjust ordering of human beings in this world had been a result of the Fall.³¹ But God’s judgement would overturn and replace it— not with parity of status, but with a new, just hierarchy reflecting men’s true merits and demerits in the next. So much for ‘the authoritarian doctrine of Original Sin’ blocking an inference from human equality to the injustice of hierarchical relations (cf. Anderson 2017: 13). The point of human equality for Roman Christians like Lactantius or Gregory, as for later medieval theologians, was generally that humans were not the same, let alone on a par, with respect to their spiritual worth. Nonetheless, their natural indifference as frail and finite creatures subject to their Creator provided the basis upon which God would ultimately compare and rightly order their souls. He alone would see through their personal differences in this world and balance the scales in the hereafter. But in the meantime, human judges must continue to do their best to preserve and correct the imperfect semblances of order that existed in the sublunary realm. The preceding analysis suggests that among early and medieval Christians, much like their Roman counterparts, human equality was not only not a normatively egalitarian principle, but often a normatively hierarchical one. ³⁰ The evolution of this metaphor may well reflect the increasing variety and instability of Roman coins, in contrast with Athenian ones. ³¹ See also Augustine (1960: 19.12, 19.15).
50
On the Historical Emergence of Basic Human Equality
Whereas modern moral and political philosophers assume that those who are ‘equal’ must therefore also be the same—hence ‘one another’s equals’ with respect to their intrinsic dignity or worth—premodern thinkers regarded human equality as compatible with, and fundamentally supportive of, the hierarchically ordered distinctions by which superiors rightly presided over inferiors both in heaven (for example, among angels) and on earth. As in ancient Greek analyses of isotes, then, Roman discussions of aequalitas continued to view social and political hierarchy as the norm and equality as the exception. Roman citizens might have been peers as well as equals (as per Cicero), as indeed might the small and dissident communities of early Christians (as per Lactantius). Still there was little sense that human beings, as such, might enjoy universal parity as well as equality.³² In affirming that ‘men were aequales in the state of innocence’, Aquinas argued forcefully that they must have been disparata, too, even before the Fall. For ‘the things which come from God are ordered … [and] as Augustine says: Order is the disposition of pares and dispares in such a way as to give each its proper place’ (Aquinas [n.d.]: Q96.A3.SC).³³ Hierarchy evidently suited the angels in their state of perfection. Surely it equally suited human beings in theirs (Q96.A4.SC). Of course, as Aquinas’s argument implicitly acknowledges, some Christians did believe that God’s respect for equality meant that the unjust hierarchies of this world would and should be overturned in this world, too, and not simply the next. While monastic life provided exceptional (and sexsegregated) sites of equality and parity for the godly to retreat from a fallen world for centuries, generations of millenarians imagined human equality as the basis of a great inversion here on earth, according to which the meek would inherit and the last would be first, in keeping with Christ’s Sermon on the Mount (See Cohn 2004). Still, even for these radicals, men and women’s equal or indifferent nature as human beings generally remained the basis for a new, inverted hierarchy recognizing the superior status and authority of ‘Saints’. Nonetheless, the fourteenth-century English Peasant’s Revolt offers an early glimpse of the process whereby BHE would eventually emerge in English political thought as an effectively egalitarian premise. In his famous ³² In his critique of slavery (written in Greek), Gregory of Nyssa presents humans as similarly superior ‘lords’ over the rest of Creation, and appeals to their Creation imago Dei as conferring upon them a worth beyond price. But neither, he suggests, preclude the appropriateness of relations of subordination between them on earth. See Garnsey (1996, 80-5). ³³ The Latin runs: ordo autem maxime videtur in disparitate consistere, dicit enim Augustinus, XIX de Civ. Dei, ordo est parium disparium que rerum sua cuique loca tribuens dispositio. Ergo in primo statu, qui decentissimus fuisset, disparitas inveniretur. He cites Augustine (1972: 21.13).
Teresa M. Bejan
51
sermon at Blackheath, the itinerant preacher John Ball asked, ‘Wh[e]n Adam dalf, and Eve span, Who was thane a gentilman?’ Although Ball himself spoke in English, the Benedictine chronicler Thomas Walsingham (1862: ii.32–33) glossed his sermon in Latin thusly: ‘[For] from the beginning all men by nature were created alike [pares], and our bondage or servitude came in by the unjust oppression of naughty men’.³⁴ Walsingham’s appeal to parity, instead of equality, in this passage reflects the former concept’s peculiar trajectory in early modern English, to which we now turn.
5. Equality-as-Parity By the time of Ball’s sermon, parity had evidently become synonymous with aristocracy in middle English, due to its association with the division of society into two classes, ‘Peers’ and ‘Commoners’. In this way, the conceptual distinction between equality and parity so important to Latin writers was eventually lost. Sir Edward Coke’s (1642) important seventeenth-century commentary on Magna Carta explained, for instance, that: Every of the Nobles is a Peer to each other, though they have several names of Dignity, as Dukes … Earles … and Barons; so … each Commoner is a Peer or Equall to another, though they be of several Degrees, as Knights, Esquires, Citizens, Gentlemen, Yeomen, and Burgesses. (19)
In Coke, we encounter a distinctive, early modern English conception of equality-as-parity according to which ‘peer’ and ‘equal’ could be treated as synonyms, and thus as equally plausible translations of the Latin pares. Elsewhere, I have shown in detail how this conception of equality-as-parity facilitated a radical reinterpretation of Christian and Roman law notions of natural equality in early modern England (Bejan 2022). The Digest’s assertion of human equality as a matter of indifferent subjection to law had remained a commonplace among medieval and early modern jurists, who emphasized its relevance to the distribution of natural liberty and rights. Accordingly, human equality did begin to have consequences for resistance theorists well before the eighteenth century as an argument for popular sovereignty (cf. Gosepath 2021). For instance, Cardinal Bellarmine argued in the late sixteenth century that men’s natural indifference meant that there was no ³⁴ For the Northern European reception of Ball’s maxim, see Resnikov (1937).
52
On the Historical Emergence of Basic Human Equality
criterion on the basis of which a just God would have elevated one as master of the rest, meaning that political authority (and only political authority) must have been conferred originally through popular consent. These arguments became popular among Protestants and Catholics keen to resist their sovereigns and replace them with more godly rulers. Still, they remained a far cry from demands for political equality, here and now, or a political voice for every human being, as such. And yet both claims were made—and spectacularly connected—in seventeenth-century England by the so-called ‘Levellers’. The year before celebrated Putney Debates of 1647, during which Rainborough insisted that ‘every man has a voice by right of nature’ (1998: 117–118), the Leveller leader John Lilburne (1998: 31) declared that: Every … individual man and woman, that ever breathed into the world … are, and were by nature all equal and alike in power, dignity, authority, and majesty, none having (by nature) any authority, dominion or majesteriall power, one over [or] above another.
Here, human equality appears not as a matter of indifference or subjection, but as a privileged and honorable status, one drawing deliberately on the intrinsic (and, indeed, superior) social worth claimed by English Peers. Lilburne’s associate, Richard Overton (1646: 3), made human equality’s newly aristocratic associations even more explicit: ‘By natural birth all men are equally and alike born to like propriety, liberty and freedom … even so are we to live, everyone equally and alike to enjoy his birthright and privilege’. As Commoners, Lilburne and Overton were ideally positioned to trade on the ambiguity of ‘Equall’ as a synonym for ‘Peer’ and so claim the privileges of English peerage for themselves. Their biblical language of ‘birthright’ likewise conflated the superior and inferior statuses characteristic of the Common Law. Even ‘the meanest man in England’, proclaimed one 1648 Leveller pamphlet, was ‘as much entitled and entailed … as the greatest subject’ (Lilburne 1648: 45). Therefore, all of the ‘Liberties, Franchises and Privileges’ previously reserved to Peers or citizens of London, including political voice, rightly belonged to every ‘freeborn’ English man—and, according to Lilburne at least—woman. Strikingly, a 1649 Leveller Women’s Petition claimed ‘an interest in Christ, equal unto men, as also a proportionable share in the freedoms of this Commonwealth’, including the right to petition Parliament (Crawford 2001: 210). These Leveller demands appear as paradigmatic examples of the process Waldron (2012b: 6) describes as ‘levelling up’ or the democratization of
Teresa M. Bejan
53
aristocratic dignity. Little wonder, then, that he and other modern egalitarians have cited Lilburne and Rainborough as historical exemplars. In Leveller hands, long-standing assertions of human equality began to issue directly in the recognizably egalitarian legal and political demands that philosophers expect, most notably for a radical expansion of the franchise.³⁵ Yet, as the preceding analysis makes clear, Levellers like Lilburne rendered human equality effective by first recasting it as a positive assertion of human parity, as the distinctive (and superior) worth and weight of human beings as such, rather than as their indifferent subjection to God. And indeed, the Levellers’ critics accused them of doing this directly. Cromwell (2022a: 484; 2022b: 72) for example, excoriated those who ‘drive at levelling and parity’ for destroying ‘the ranks and orders of men,—whereby England hath been known for hundreds of years’.
6. Conclusion In Leveller invocations of equality-as-parity, we encounter human equality in something like its modern form, as a normatively egalitarian assertion of the shared (high) status and intrinsic worth of human beings, as such. Accordingly, it may be fair, if still anachronistic, to describe their controversial seventeenth-century legal and political demands as ‘egalitarian’ ones. Yet even these fell short of modern expectations. For instance, the ‘proportionable’ share of political liberty claimed by Leveller women did not include their equal right to vote in Parliamentary elections (Crawford 2001: 210).³⁶ According to my historical analysis, however, the persistence of political inequality in Leveller thought was not a blind spot but rather an indication of the lasting power of equality-as-balance, specifically, as an ideal involving determinations of whom and how different kinds of people should be assigned political weight in a particular polity in order to balance the whole. Moreover, unlike modern BHE—which insists that every human being must regarded as the same with respect to their fundamental status and/or worth— Leveller equality-as-parity remained consistent with inequalities of status and dignity among the set of peers, as between barons and earls, men and women, old and young. By contrast, modern BHE insists that these and other differences are or should be regarded as matters of indifference—that ³⁵ Just how radical their proposed extension was, however, remains a vexed historical question. For an overview, see Foxley (2013). ³⁶ Cf. Anderson (2017: 16, 151).
54
On the Historical Emergence of Basic Human Equality
as peers, every human being must be regarded as the same or strictly equal with respect to their status and/or worth. My own, historically informed view is that modern BHE is a combination of two, historically distinct conceptions of equality—equality-as-parity and equality-as-indifference—and that these are in some tension. Consider Gerry Cohen’s (2013) posthumously published essay ‘On Regarding People as Equals’, which appeals to an eighteenth-century poem by Robert Burns: ‘The rank is but the guinea’s stamp, the Man’s the gowd [gold] for a’ that’ (193). Cohen links this to what he takes to be ‘the right principle’ expressed by BHE: ‘that nobody is anybody’s better’ (200).³⁷ Likewise, Burns’s poem suggests that the many embodied differences and social distinctions that divide human beings on earth are superficial, the ‘mere guinea’s stamp’. Like gold coins, we are made, deep down, of the same valuable stuff, and in strictly equal amounts. None is greater or lesser, none better or worse. Our claim to equality is simultaneously positive and privative. There is, as we have seen, a long history of using the metaphor of coinage to understand the ways in which human beings are and are not ‘equal’. In this chapter, I have sought to show that the subtle transformation of this metaphor over time reflected the relative priority of balance, indifference, and parity in making sense of equality as a relational property when applied to human beings in different contexts. To ignore this history is to flatten out important conceptual distinctions. Still, outside of the occasional historical quotations that serve as window-dressing for their normative arguments, modern philosophers continue to rely on an overly simplistic and misleading understanding of equality-as-sameness in their discussions of BHE, while trading on the concept’s geometric connotations of transcendent truth. Attending to the historical emergence of BHE reveals, however, that human equality neither is, nor was, a matter simply of sameness, nor indeed have arguments for political equality always proceeded in the way that modern egalitarians expect. Throughout this chapter, I have sought to show that the important question for philosophers and historians should thus be not when human equality was ‘invented’ or discovered, but rather how and why this idea became effectual as a premise of political theorizing in a particular place and time. My answer is that the idea of human equality became, at last, normatively egalitarian in early modern England through its successful conflation with the idea of parity—specifically, that of social parity among English Peers. Yet today, the distinctiveness of parity has been systematically ³⁷ For his own part, Cohen (2013: 198) expresses discomfort with Dworkinian ‘equal concern and respect’ on the grounds that ‘different degrees of respect are appropriate for different adults’. See also Green (2010).
Teresa M. Bejan
55
effaced through modern English translations, which render the Latin pares routinely as ‘equals’, as in the phrase primus inter pares. Recently, however, Ruth Chang (2002) has sought to revive the concept of parity in ethics. Echoing Aristotle, Chang argues that equality implies only three possible relations: things are always either ‘more than’, ‘less than’, or ‘exactly equal’ with respect to whatever characteristic is being measured. In the domain of value, then, this produces a trichotomy of ‘better than’, ‘worse than’, or ‘exactly equal’. For Chang, however, parity represents a fourth possibility, describing cases of evaluative ‘comparability’ in which two things are comparable, yet there exists no determinate answer to whether one is better or worse or both are equally good. (Her favoured example is that of Mozart and Michelangelo with respect to creativity.) Chang has thus far resisted applying her analysis to discussions of BHE, but as we’ve seen there is ample historical precedent for doing so. Those opposed might object that human beings, as such, are not incommensurable, and so Changian parity does not apply. If Chang’s comparisons are in the realm of apples and oranges (for example, music and sculpture as forms of creativity), surely basic equality is a matter of apples to apples.³⁸ Insofar as individuals are all ‘indifferent’ tokens of the same human type, then yes. But in discussions of BHE, we are generally concerned with the valuable characteristics we take to be determinative of human status or worth. Here, I agree with Waldron (2017) that BHE is best understood as the product not of one such characteristic (for example, moral agency, reason, or Rawls’s sense of justice), but of a constellation of characteristics that no two human beings will share in exactly equal amounts.³⁹ One need not insist on the infinite specialness of every individual, or indeed their worth beyond price, but only on their plurality with respect to the panoply of traits one associates with human worth. Given this, Changian parity would seem to capture well the relationship involved in BHE much better than equality-as-sameness.⁴⁰ Moreover, construing BHE as a parity relation would allow us to avoid the gradability objection pressed by critics like Pojman (1997) and Steinhoff (2015). In brief: it is theoretically implausible, as well as empirically false, to claim that every person is the same with respect to their innate abilities or characteristics. ³⁸ I thank Jay Wallace for pressing me on this point. ³⁹ While Waldron offers an incomplete list, Phillips (2021: 57) argues that we should give up on what Cohen called ‘the wild goose chase for defining characteristics’, instead remaining inclusive and perhaps even agnostic on what might count as a relevantly valuable trait. On my view, we should expect the putative bases of BHE to be contingent on historical context and shift over time. ⁴⁰ It might be fruitfully compared to Cupit’s (2000: 119) argument for BHE on the basis of individuals’ ‘incomparable worth’ and Kolodny’s (2022) arguments for the importance of ‘noninferiority’ as an egalitarian moral sentiment.
56
On the Historical Emergence of Basic Human Equality
Thus, if individuals differ in degree with regard to their valuable traits, they must also be either ‘more’ or ‘less’—that is, unequal—with respect to their worth or status, too. To insist otherwise is arbitrary. In response to this objection, Waldron (2017) has followed Rawls (1971: 444–445) in resorting to the technical language of range properties, presenting BHE as a binary property contingent on a scalar one.⁴¹ People are therefore ‘equals’, argues Waldron (2017: 117–120), because they possess some (valuable) trait or set of traits beyond a certain threshold, such that any difference of degree does not affect their shared status. Likewise, Carter (2011) suggests that the equal respect we are due as equals itself demands that individuals be spared the disrespect implicit in evaluating, let alone measuring outright, our relative agential capacities. The idea is that such assessments are themselves violations of BHE and its stipulation that every human being is of equal worth. Range properties are just one of several ways in which philosophers have sought to render the sources of BHE ‘nongradable’, in Vlastos’s (1997) phrase. Yet recovering parity as a distinct concept helps us recognize the problem as arising due to a doubling-up of ‘equality’ language, producing a tension in turn between our status as ‘equals’ and our evidently ‘unequal’ degree with respect to our defining characteristic(s). As Coke pointed out, however, people can clearly be peers—that is, pares with respect to status or worth—despite some differences of degree. Accordingly, substituting ‘peers’ for ‘equals’ in explicating BHE sidesteps the challenge posed by the gradability objection. Moreover, to say that human beings are on a par or ‘peers’ with respect to their fundamental status as human beings, or across the panoply of valuable traits we regard as sources of human worth, suggests that they ought to be treated as such—that is, as possessing a shared, high status entitled to positive, but not identical, respect.⁴² Bernard Williams (1973) once noted the ‘discomfort’ arising from the conceptual confusions and contradictory demands of equality but argued that these were ‘no greater … [than] with liberty, or any other noble and substantial political ideal’ (249). This chapter suggests, however, that the timeless, ⁴¹ Waldron (2017: 117–120) cites ‘being in Scotland’ as an example: if a place falls within a fixed range of coordinates, it counts equally as ‘being in Scotland’, despite its relative distance from the border. ⁴² Something like this thought, I believe, underlies Dworkin’s (1981: 185) influential distinction between people ‘equally’ and ‘as equals’. The former treats them as exactly the same—as though their differences were matter of indifference—while the latter treats them differentially, in a way that respects their differences while also preserving their shared status as peers deserving of respectful treatment. Interestingly, Rawls invokes parity directly in his discussion of BHE as a range property: ‘While individuals presumably have varying capacities for a sense of justice, this fact is not a reason for depriving those with a lesser capacity of the full protection of justice. Once a certain minimum is met, a person is entitled to equal liberty on a par with everyone else’ (Rawls 1971: 443. My emphasis).
Teresa M. Bejan
57
mathematical connotations of ‘equality-talk’ have created particularly stubborn blind spots in contemporary moral and political philosophy for which historical awareness provides an incomparable corrective. Specifically, the process whereby ancient ideals of equality-as-balance and -indifference gave way to an early modern conception of equality-as-parity sheds light on the peculiar structure of modern BHE as we continue to debate its basis and coherence today.⁴³
References ‘aequālĭtas, ātis, f. aequalis’. 1879. In A Latin Dictionary, edited by Charlton T. Lewis and Charles Short. Oxford: Clarendon Press. Available at Perseus Digital Library. http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04. 0059%3Aentry%3Daequalitas. ‘Egalitarian, adj. and n.’. 2022. In Oxford English Dictionary. Oxford: Oxford University Press. https://www.oed.com/view/Entry/59856. ‘Equality, n.’. 2019. Oxford English Dictionary. Oxford: Oxford University Press. https://www.oed.com/view/Entry/63702. ‘Extract from the Debates at the General Council of the Army, Putney. 29 October 1647’. 1998. In The English Levellers, edited by Andrew Sharp, 102–130. Cambridge: Cambridge University Press. ‘pār, păris’. 1879. In A Latin Dictionary, edited by Charlton T. Lewis and Charles Short. Oxford: Clarendon Press. Available at Perseus Digital Library. http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04. 0059%3Aentry%3Dpar. ‘părĭtas, ātis, f. par’. 1879. In A Latin Dictionary, edited by Charlton T. Lewis and Charles Short. Oxford: Clarendon Press. Available at Perseus Digital Library. http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04. 0059%3Aentry%3Dparitas. ‘peer, n. and adj’. 2022. Oxford English Dictionary. Oxford: Oxford University Press. https://www.oed.com/view/Entry/139725. Anderson, Elizabeth. 2017. Private Government: How Employers Rule Our Lives (and Why We Don’t Talk About It). Princeton, NJ: Princeton University Press.
⁴³ Thanks to the editors and anonymous readers for Oxford University Press, as well as other volume contributors, for their helpful comments. An earlier version of this chapter was presented to the ‘The Idea and Practice of Equality: Historical Perspectives’ conference hosted by Devin Vartija at the University of Utrecht in 2022. I am grateful to the other participants, especially Anne Phillips and Darrin McMahon, for their advice and encouragement. I am also grateful to Ming Kit Wong for editorial assistance in preparing the manuscript.
58
On the Historical Emergence of Basic Human Equality
Ando, Clifford. 2010. ‘“A Dwelling Beyond Violence”: On the Uses and Disadvantages of History for Contemporary Republicans’. History of Political Thought, 31: 183–220. Angle, Stephen C. et al. 2017. ‘Hierarchies Have a Place Even in Societies Built on Equality’. Aeon, March 22. https://aeon.co/essays/hierarchies-have-a-place-evenin-societies-built-on-equality. Aquinas, Thomas. n.d. ‘Psalm 4’. In Commentary on the Psalms, edited by Joseph Kenny and translated by Albert Marie Surmanski and Maria Veritas Marks. https://aquinas.cc/la/en/~Psalm.Ps4. Arena, Valentina. 2012. Libertas and the Practice of Politics in the Late Roman Republic. Cambridge: Cambridge University Press. Arendt, Hannah. 2006. On Revolution. London: Penguin Books. Aristotle. 1926. Nicomachean Ethics, translated by H. Rackham. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www. loebclassics.com/view/LCL073/1926/volume.xml. Aristotle. 1932. Politics, translated by H. Rackham. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www.loebclassics.com/ view/LCL264/1932/volume.xml. Aristotle. 1933. Metaphysics, vol. I, translated by Hugh Tredennick. Cambridge, MA: Harvard University Press. https://www.loebclassics.com/view/LCL271/ 1933/volume.xml. Arneson, Richard. 2015. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Augustine. 1847. Expositions on the Book of Psalms, vol. I. Oxford: John Henry Parker. Augustine. 1960. City of God, vol. VI, translated by William Chase Greene. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https:// www.loebclassics.com/view/LCL416/1960/volume.xml. Augustine. 1972. City of God, vol. VII, translated by William M. Green. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www. loebclassics.com/view/LCL417/1972/volume.xml. Barry, Brian. 1999. ‘Sustainability and Intergenerational Justice’. In Fairness and Futurity: Essays on Environmental Sustainability and Social Justice, edited by Andrew Dobson, 96–97. Oxford: Oxford University Press. Beard, Mary. 2014. ‘The Public Voice of Women’. London Review of Books, 20 March. https://www.lrb.co.uk/the-paper/v36/n06/mary-beard/the-public-voiceof-women. Bejan, Teresa M. 2019. ‘Mary Astell on Equality, Hierarchy, and Ambition’. Political Theory, 47: 781–808.
Teresa M. Bejan
59
Bejan, Teresa M. 2020. ‘Free Expression or Equal Speech’. Social Philosophy and Policy, 37: 153–169. Bejan, Teresa M. 2021. ‘No Respecter of Persons’. Journal of Biblical Literature, 140: 831–836. Bejan, Teresa M. 2022. ‘What Was the Point of Equality?’ American Journal of Political Science, 66: 604–616. Berman, Joshua A. 2011. Created Equal: How the Bible Broke with Ancient Political Thought, reprint edition. Oxford: Oxford University Press. Bidadanure, Juliana. 2021. Justice Across Ages: Treating Young and Old as Equals. Oxford: Oxford University Press. Carter, Ian. 2011. ‘Respect and the Basis of Equality’. Ethics, 121: 538–571. Chang, Ruth. 2002. ‘The Possibility of Parity’. Ethics, 112: 659–688. Christiano, Thomas. 2002. ‘Democracy as Equality’. In Democracy, edited by David Estlund, 31–50. Oxford: Wiley. Cicero. 1913. On Duties, translated by Walter Miller. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www.loebclassics.com/ view/LCL030/1913/volume.xml. Cicero. 1923. On Old Age. On Friendship. On Divination, translated by W. A. Falconer. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library https://www.loebclassics.com/view/LCL154/1923/volume.xml Cicero. 1928. On the Republic. On the Laws, translated by Clinton W. Keyes. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https:// www.loebclassics.com/view/LCL213/1928/volume.xml. Cohen, G. A. 2013. ‘On Regarding People as Equals’. In Finding Oneself in the Other, edited by M. Otsuka, 193–200. Princeton: Princeton University Press. Cohn, Norman. 2004. The Pursuit of the Millennium: Revolutionary Millenarians and Mystical Anarchists of the Middle Ages. London, Pimlico. Coke, Edward, 1642. The Second Part of the Institutes of the Lawes of England. London: M. Flesher and R. Young. Available at Google Books. https://www.google.co. uk/books/edition/The_Second_Part_of_the_Institutes_of_the/_1IzAAAAIAAJ. Crawford, Patricia. 2001. ‘The Poorest She: Women and Citizenship in Early Modern England’. In The Putney Debates of 1647: The Army, the Levellers, and the English State, edited by Michael Mendle, 199-220. Cambridge: Cambridge University Press. Cromwell, Oliver. 2022a. The Letters, Writings, and Speeches of Oliver Cromwell, vol. 1, edited by Andrew Barclay, Tim Wales, and John Morrill. Oxford: Oxford University Press. Cromwell, Oliver. 2022b. The Letters, Writings, and Speeches of Oliver Cromwell, vol. 3, edited by Joel Halcomb, Patrick Little, and David L. Smith. Oxford: Oxford University Press.
60
On the Historical Emergence of Basic Human Equality
Cupit, Geoffrey. 2000. ‘The Basis of Equality’. Philosophy, 75/291: 105–125. Dworkin, Ronald. 1978. Taking Rights Seriously. Cambridge, MA: Harvard University Press. Dworkin, Ronald. 1981. ‘What Is Equality? Part I: Equality of Welfare’. Philosophy & Public Affairs, 10: 185–246. Dworkin, Ronald. 1983. ‘Comment on Narveson: In Defense of Equality’. Social Philosophy & Policy, 1: 24–40. Dworkin, Ronald. 2000. Sovereign Virtue: The Theory and Practice of Equality. Cambridge, MA: Harvard University Press. Fleischacker, Samuel. 2004. A Short History of Distributive Justice. Cambridge, MA: Harvard University Press. Foxley, Rachel. 2013. The Levellers: Radical Political Thought in the English Revolution. Manchester: Manchester University Press. Frank, Jill. 2005. A Democracy of Distinction: Aristotle and the Work of Politics. Chicago, IL: University of Chicago Press. Garnsey, Peter, 1996. Ideas of Slavery from Aristotle to Augustine. Oxford: Oxford University Press. Glover, Samuel D. 1999. ‘The Putney Debates: Popular versus Elitist Republicanism’. Past & Present, 164: 47–80. Gosepath, Stefan. 2021. ‘Equality’. The Stanford Encyclopedia of Philosophy, Summer edition, edited by Edward N. Zalta. https://plato.stanford.edu/archives/sum2021/ entries/equality. Green, Leslie. 2010. ‘Two Worries About Respect for Persons’. Ethics, 120: 212–231. Gregory I. 1845. Morals on the Book of Job, vol. II. Oxford: John Henry Parker. Hoekstra, Kinch. 2013. ‘Hobbesian Equality’. In, Hobbes Today: Insights for the 21st Century, edited by S. A. Lloyd, 76–112. Cambridge: Cambridge University Press. Holland, Tom. 2019. Dominion: The Making of the Western Mind. London: Little Brown. Kaye, Joel. 2014. A History of Balance, 1250–1375: The Emergence of a New Model of Equilibrium and Its Impact on Thought. Cambridge: Cambridge University Press. Kirby, Nikolas. 2018. ‘Two Concepts of Basic Equality’. Res Publica, 24: 297–318. Kolodny, Niko. 2022. The Pecking Order: Social Hierarchy as a Philosophical Problem. Cambridge, MA: Harvard University Press. Kurke, Leslie. 1999. Coins, Bodies, Games, and Gold: The Politics of Meaning in Archaic Greece. Princeton, NJ: Princeton University Press. Kymlicka, Will. 1990. Contemporary Political Philosophy. Oxford: Clarendon Press. Lactantius. 2003. Divine Institutes, translated and edited by Anthony Bowen and Peter Garnsey. Liverpool: Liverpool University Press. Lilburne, John. 1648. The People’s Prerogative and Privileges. London. Available at Early English Books Online. http://name.umdl.umich.edu/A88231.0001.001.
Teresa M. Bejan
61
Lilburne, John. 1998. ‘Postscript to The Freeman’s Freedom Vindicated. 16 June 1646’. In The English Levellers, edited by Andrew Sharp, 31–32. Cambridge: Cambridge University Press. Livy. 1922. History of Rome, vol. II, translated by B. O. Foster. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www. loebclassics.com/view/LCL133/1922/volume.xml. Lombardini, John. 2013. ‘“Isonomia” and the Public Sphere in Democratic Athens’. History of Political Thought, 34: 393–420. Lucas, J. R. 1965. ‘Against Equality’. Philosophy, 40: 296–307. Lucas, J. R. 1977. ‘Against Equality Again’. Philosophy, 52: 255–280. McMahon, Darrin. 2023. Equality: The History of An Elusive Idea. New York: Basic Books. Ober, Josiah. 2012. ‘Democracy’s Dignity’. American Political Science Review, 106: 827–846. Ober, Josiah. 2017. Demopolis: Democracy before Liberalism in Theory and Practice. Cambridge: Cambridge University Press. Overton, Richard, 1646. An Arrow Against All Tyrants. London: Martin ClawClergy. Available at Early English Books Online. http://name.umdl.umich.edu/ A90228.0001.001. Pettit, Philip. 2014. Just Freedom: A Moral Compass for a Complex World. New York: W.W. Norton & Company. Pettit, Philip. 2017. ‘Hierarchy Is Either Strictly Constrained or It Is Indefensible’. Aeon, March 24. https://aeon.co/ideas/hierarchy-is-either-strictly-constrainedor-it-is-indefensible. Phillips, Anne. 2021. Unconditional Equals. Princeton, NJ: Princeton University Press. Plato. 1926. Laws, vol. I, translated by R. G. Bury. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www.loebclassics.com/ view/LCL187/1926/volume.xml. Plato. 2013. Republic, vol. I, edited and translated by Christopher EmlynJones and William Preddy. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www.loebclassics.com/view/LCL237/2013/ volume.xml. Pojman, Louis. 1997. ‘On Equal Human Worth: A Critique of Contemporary Egalitarianism’. In Equality: Selected Readings, edited by Louis P. Pojman and Robert Westmoreland, 282–299. Oxford: Oxford University Press. Polybius. 2011. The Histories, vol. III, translated by W. R. Paton and revised by F. W. Walbank and Christian Habicht. Cambridge, MA: Harvard University Press. Available at Loeb Classical Library. https://www.loebclassics.com/view/LCL138/ 2011/volume.xml.
62
On the Historical Emergence of Basic Human Equality
Rawls, John. 1971. A Theory of Justice. Cambridge, MA: Belknap Press of Harvard University Press. Resnikov, Sylvia. 1937. ‘The Cultural History of a Democratic Proverb’. Journal of English and Germanic Philology, 36: 391–405. Rosanvallon, Pierre. 2013. The Society of Equals. Cambridge, MA: Harvard University Press. Scanlon, T. M. 2018. Why Does Inequality Matter? Oxford: Oxford University Press. Schiavone, Aldo. 2022. The Pursuit of Equality in the West. Cambridge, MA: Harvard University Press. Schofield, Malcolm. 2021. Cicero: Political Philosophy. Oxford: Oxford University Press. Schwartzberg, Melissa. 2013. Counting the Many: The Origin and Limits of Supermajority Rule. Cambridge: Cambridge University Press. Siedentop, Larry. 2014. Inventing the Individual: The Origins of Western Liberalism. London: Allen Lane. Skinner, Quentin. 2012. Liberty before Liberalism. Cambridge: Cambridge University Press. Steinhoff, Uwe. 2015. ‘Introduction’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, xi–xviii. Oxford: Oxford University Press. Stuurman, Siep. 2017. The Invention of Humanity: Equality and Cultural Difference in World History. Cambridge: Cambridge University Press. Swift, Adam. 2006. Political Philosophy: A Beginners’ Guide for Students and Politicians. Cambridge: Polity Press. Vlastos, Gregory. 1947. ‘Equality and Justice in Early Greek Cosmologies’. Classical Philology, 42: 156–178. Vlastos, Gregory. 1997. ‘Justice and Equality’. In Equality: Selected Readings, edited by Louis P. Pojman and Robert Westmoreland, 120–136. Oxford: Oxford University Press. Waldron, Jeremy. 2002. God, Locke, and Equality: Christian Foundations in Locke’s Political Thought. Cambridge: Cambridge University Press. Waldron, Jeremy. 2012a. ‘The Image of God: Rights, Reason, Order’. In Christianity and Human Rights: An Introduction, edited by John Witte and Frank S. Alexander, 216–235. Cambridge: Cambridge University Press. Waldron, Jeremy. 2012b. Dignity, Rank, and Rights, edited by Meir Dan-Cohen. Oxford: Oxford University Press. Waldron, Jeremy. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: Harvard University Press. Walsingham, Thomas. 1862. Historia Anglicana, edited by H. T. Riley, 2 vols. London: Green Longman and Roberts Longman.
Teresa M. Bejan
63
Watson, Alan, translated and edited. 1985. The Digest of Justinian. 4 vols. Philadelphia, PA: University of Pennsylvania Press. Westen, Peter. 1982. ‘The Empty Idea of Equality’. Harvard Law Review, 95: 537–596. White, Nicholas P. 1971. ‘Aristotle on Sameness and Oneness’. The Philosophical Review, 80: 177–197. White, Stuart. 2007. Equality, Cambridge: Polity Press. Whitman, James Q. 2005. Harsh Justice: Criminal Punishment and the Widening Divide between America and Europe. Oxford: Oxford University Press. Williams, Bernard. 1973. Problems of the Self: Philosophical Papers, 1956–1972. Cambridge: Cambridge University Press.
2 Basic and Relational Equality Jeremy Waldron
1. A Question Basic equality is the idea that we humans are fundamentally one another’s equals. Relational equality is the idea that in society we should relate to one another as equals. What is the connection between these two ideas? Theorists of relational equality have defined and elaborated a powerful and attractive social ideal, which they present as an understanding of equality superior to distributive theories such as luck-egalitarianism.¹ In my book One Another’s Equals: The Basis of Human Equality,² I presented an understanding of what I called ‘basic equality’—others have called it ‘deep equality’ or ‘abstract equality’—which I also differentiated from luck-egalitarianism and other distributive theories.³ Does this common differentiation mean that basic equality and relational equality are the same? No. To mangle Euclid, things which are different from the same thing (distributive equality) are not necessarily equal to one another.
2. An Inadequate Answer I said something, but not nearly enough, about the difference between relational equality and basic equality in One Another’s Equals. In Chapter One of that book, referring, among other things, to an edited collection called Social Equality: Essays on What It Means to Be Equals,⁴ I said this:
¹ Relational equality is set out in Scheffler (2003: 5) and in Anderson (1999: 287). For the best-known version of luck-egalitarianism, see Dworkin (2002). ² Waldron (2017). ³ Waldron (2017: 2, 9–15). See also Dworkin (1983: 24). ⁴ Fourie, Schuppert, and Walliman-Helmer (2015).
Jeremy Waldron, Basic and Relational Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0003
66
Basic and Relational Equality
The first of these [relational equality] includes some important recent work on what it means for a society (or a family, a business, or a friendship) to be organized so that its members relate to one another as equals. This involves considering what are called relational (as opposed to distributive) ideas of equality. Relational equality is plainly closer to what I am calling basic equality than distributive equality is. But the two—basic equality and relational equality—donʼt seem to be regarded as identical. What is sometimes said is that we invoke the proposition that all people have equal moral worth to defend the claim that we should organize our society as a society of equals.5
And I continued: Mostly those who write about the relational conception spend their energy talking not about what relational equality is based on but about the structures and practices it requires. There has been interest, too, not only in highlighting the distinction between relational and distributive concerns but also in considering whether relational equality might itself make demands of a distributive kind. And the picture that sometimes seems to emerge is that we proceed from basic equality to relational equality and only thence to distributive equality. I am not sure about that; I believe that basic equality sometimes has distributive implications all by itself. Anyway, relational equality is certainly important. I say a little more about it in [Chapter] 2, for I believe it comprises some of the most exacting normative consequences of basic equality. In the meantime, let me reiterate that I am not equating basic equality with relational equality. Their exact connection needs to be explored.6
The cross-reference to Chapter Two of the book didn’t add much. All I did there was mention relational equality as one among several normative consequences of basic equality: ‘There are also what Samuel Scheffler and others call principles of relational equality, generated as specific normative consequences of the basic equality principle’.⁷ The relational idea deserved better than that. So, none of this discussion was adequate as an explication of the difference between basic and relational equality. I want to do better in this chapter. In One Another’s Equals I took for granted that relational equality is generated by basic equality; but in this chapter I want to consider whether there is sufficient difference between relational and basic equality to entitle us to treat one as an implication of the other. ⁵ Waldron (2017: 11–12). ⁶ Waldron (2017: 12). ⁷ Waldron (2017: 53).
Jeremy Waldron
67
3. My Aim in This Chapter I wish to emphasize that in considering whether and how relational equality differs from basic equality, my aim is definitely not to criticize the former (either for what it is or what it fails to do). I believe that relational equality is as important, normatively, as Anderson and Scheffler think it is. And I certainly don’t want to accuse them of having tried to disparage basic equality or anything like that. Nor, if they were guilty of that, would I want to return the favour. I would like to show that relational equality doesn’t really do the work of basic equality, so that discussion of it—though important in its own right— is no substitute for discussion of basic equality. And, as I said in the book, discussion focusing steadfastly on basic equality is long overdue.⁸ Mostly I want to get clear about the architecture in this area: how are these two conceptions related and how is their relatedness structured? With this as our topic, I fear that the discussion is going to be quite dry and structural.
4. Are Basic and Relational Equality Propositions of Different Kinds? I have a hunch that formulations of basic equality are propositions of a different kind from formulations of relational equality. The question asked by relational equality theorists (and their distributivist rivals) is something like: ‘What sort of policies should we adopt?’ ‘What overall approach should we take to social relations?’ The question asked by the theorist of basic equality is: ‘What, ultimately, are we like?’ ‘Are we one another’s equals?’ These foundational questions might make relational egalitarians impatient. Relational accounts of equality, like distributivist theories, are normative or normative in character (telling us what we ought to do or how things ought to be organized), while basic equality makes a more abstract metaphysical claim (maintaining that we are—as a matter of fact—one another’s equals).⁹ Relational and distributivist theorists may say, ‘Never mind the metaphysics. What should we do?’ Certainly, something feels ‘ontological’ in the basic equality claim. An inquiry into basic equality asks ‘What sort of creatures are we? Are we one ⁸ Waldron (2017: 10–12). ⁹ A word about this loose terminology. Sometimes the term ‘normative’ is distinguished from ‘prescriptive’ and sometimes too from ‘evaluative’. I am aware of those distinctions (see Waldron 2017: 42–43). But I don’t think they matter much here; and so I shall use the word ‘normative’ in a way that could also cover ‘prescriptive’.
68
Basic and Relational Equality
another’s equals?’ So, I suspect that something along the lines of a fact/value contrast applies here. On the other hand, it doesn’t apply straightforwardly. For we shall soon see that basic equality can (and must) be read also as a normative position. I shall explain this further in Sections 12 to 15 of this chapter. That seems to put basic equality in the same category—the normative category—as claims of relational and distributive equality. But in fact things are more complicated than this. Propositions about the basic equality of all humans seem to have a foot in both camps. If they are normative principles, they are not just normative principles. And if they comprise a descriptive/ontological element, it is an element alloyed with their normative commitment.
5. This Is Not about Moral Realism That we humans are one another’s equals appears to represent a claim about reality. Now, I don’t mean ‘a claim about reality’ in the sense that a moral realist might call any normative proposition a claim about (moral) reality. I don’t want to rule that out, but actually I really don’t want to get into the issue of moral realism at all.¹⁰ Suffice to say that if ‘Humans are one another’s equals’ is a claim about moral reality in this sense, so are all the propositions that I have been contrasting with it in this discussion, like those propounded by relational and distributive egalitarians. Whatever the truth of moral realism, ‘Humans are one another’s equals’ seems more fact-ish, more descriptive than that. That is what I want to explore.
6. Factual Similarities and Differences The proponent of basic equality often says that racists and sexists are wrong about the facts. Men are not so different from women, blacks are not so different from whites, as to refute the claim that all of them are one another’s equals. There are differences of course, as there are between any one human person and another, but the differences can be understood as falling within an intelligible range of variation indicating the common possession of what John Rawls called a range property.¹¹ Or the differences that exist between one person and another can be seen as relevant to some kinds of moral assessment and not others: so, for example, there are differences of merit relevant to the application of certain principles even though there are no differences ¹⁰ Waldron (2017: 46). ¹¹ Waldron (2017: 117–118). See Rawls (1999: 444).
Jeremy Waldron
69
that affect fundamental worth (relevant to the attribution of human rights etc.). These conceptions, which an adequate theory of basic equality has no choice but to develop and elaborate, help us understand what the moral world is like.¹² They are not just expressive of certain prescriptions or commitments of principle.
7. Supervenience Perhaps what makes propositions of basic equality seem normative is their supervenience upon certain indubitably factual claims.¹³ I am not sure about this. Supervenience is not necessarily a mark of the normative. Some factual statements are supervenient on other factual statements: claims about mental properties are said to be supervenient on claims about physical or neural properties, for example. Still the supervenience of basic equality on descriptive propositions is worth considering in our discussion. We say: ‘A ll human persons are one another’s equals because they all have the potential for moral reason’ (or something along those lines). Even if the ‘A ll human persons are one another’s equals’ part is ultimately a normative statement, it usually comes packaged together with a factual assertion like ‘They all have the potential for moral reason’, and we consider and debate the whole package. What is more, the packaging seems pretty tight. What basic equality is supposed to supervene upon is not just something that the basic egalitarian might eventually get around to discussing (or not). It is something he is expected to discuss more or less as soon as he announces his position. I think this is a contrast with other forms of egalitarianism, like relational equality and luck-egalitarianism. Discussion of those theories tends to remain at the normative level; it does not explore the ontological underpinnings (except to the extent that luck-egalitarianism raises questions about the nature of choice and agency in distributive matters).
8. Egalitarian Commitments without Supervenience? So: supervenience seems to matter more in the case of basic equality. Or does it? Here’s a possible counter-example. In my book I noted a number of basic equality principles whose proponents claimed to eschew this sort ¹² Waldron (2017: 157–166). ¹³ Please forgive a little looseness in my use of supervenience. Sometimes I will talk as though certain values supervene upon certain facts; mostly I shall talk as though certain value-propositions supervene upon certain descriptive propositions. I think the latter is the more correct formulation.
70
Basic and Relational Equality
of supervenience. In her 1947 essay ‘Natural Rights’,¹⁴ Margaret MacDonald based her egalitarian view on a radically decisionist conception of value utterances. She said: ‘To assert that all men are of equal worth is not to state a fact but to choose a side. It announces: this is where I stand’. She acknowledged that somebody might ask, ‘Yes, but why do you choose to stand there? What is it about humans that makes this a sensible place to stand?’ But she responded, defiantly, ‘I affirm that no natural characteristic constitutes a reason for the assertion that all human beings are of equal worth’. The idea (as I understand it) is that some of us simply commit ourselves or find ourselves committed to a moral outlook organized around basic equality. We may try to attract others to it, but we do not do that by pointing to any property or fact that is going to compel other people to line up with us in the same way. I spent some time in One Another’s Equals criticizing positions of this kind.¹⁵ I won’t repeat that here. Suffice to say (as I said in Section 7) that— even if there are some outliers—many if not most basic equality positions are treated by proponents and opponents alike as a pretty tight package asserting a supervenience of basic equality on certain factual properties that the putative equals are supposed to possess.
9. What Does Relational Equality Supervene Upon? If supervenience upon factual properties applies to a principle of basic equality, it probably applies in one form or another to principles of relational equality as well (and so it doesn’t really distinguish the one kind of principle from the other). The relational principle ‘Social arrangements should be organized so that they embody and express equal respect between persons’ makes sense on account of certain features that human persons have in common in virtue of which this is an appropriate way to treat them. A full account of relational equality would include a factual account that makes the normative principles of relational equality intelligible. It would try to answer questions like: ‘What is it about human persons that makes this principle viable or intelligible or desirable?’ Admittedly, theorists of relational equality have not spent a whole lot of time on this aspect of the matter. I don’t think this is because they adopt a sort of Margaret McDonald attitude to their principle, as discussed above in ¹⁴ MacDonald (1984: 21). (This paper was originally published in Proceedings of the Aristotelian Society, 47 [1947]: 225.) ¹⁵ Waldron (2017: 57–61).
Jeremy Waldron
71
Section 8. (Perhaps I am wrong about that. Can we imagine a version of relational equality which is simply expressive of a determination to treat people as if they were basically one another’s equals without any commitment to this being the way they actually are?) Alternatively, it may be thought that relational equality supervenes upon the same facts that basic equality supervenes upon (with no ‘as if ’ about it). I think it is an interesting question whether this is so, and it raises an issue which of course we will have to discuss (Sections 16 to 18), which is whether basic equality and relational equality are normatively more or less the same—in which case it would be no surprise if the two principles supervene upon the same facts.
10. Supervenience and Shapelessness One other point about supervenience. I have spoken as though there were a clear division between normative statements and descriptive statements, each intelligible on their own terms, so that a given normative statement might be treated as supervenient upon a descriptive proposition, with the latter being independently intelligible qua descriptive characterization. But this may not be so. The descriptive characterization on which a moral principle is supervenient may be intelligible only in its relation to the moral principle. Apart from that relation, it might seem, as John McDowell suggests, shapeless.¹⁶ I discussed this briefly in One Another’s Equals.¹⁷ An example might be our use of the constructive form of range-property to delineate the characteristic(s) that basic equality is supposed to supervene upon. The discernment of a range-property doesn’t really make sense except in relation to a normative agenda of some kind. So even if basic equality inevitably presents itself as a package comprising not just a normative position but a factual characterization on which the normative position is supervenient, still the whole package—both parts—may be permeated through and through by the normative orientation that basic equality conveys.
11. A Summary So Far I have spent some time considering the possibility that basic equality might be distinguished from relational equality as factual claims are distinguished from normative ones. That possibility doesn’t really hold up. The closest we ¹⁶ McDowell (1981: 144). ¹⁷ Waldron (2017: 64–65).
72
Basic and Relational Equality
come to it is some sense that the principle of basic equality hugs its factual basis to itself in a way that is closer than the relation between relational equality and its factual basis. I don’t know why this would be. But discussions of basic equality almost always open up questions about the truth and content of the descriptive claims that basic equality is supposed to be supervenient upon, whereas discussions of relational equality seem able to proceed purely at the normative level, without considering any factual underpinnings. This may be just an informal division of labour, not particularly theorized or well thought out. Perhaps it’s just a function of what most relational egalitarians happen to be interested in.
12. Basic Equality as a Normative Position Let us consider now the view that basic equality is indeed an unabashedly normative position, in which case it is in the same general category as relational equality. If this is so, then it looks as if we have to say that either (i) there are important normative differences between basic equality and relational equality or (ii) they are, in the final analysis, the same normative ideal. I take it, by the way, that there is no disagreement that a proposition can be normative notwithstanding its (grammatically) indicative formulation. We may say that we are one another’s equals, but the conveyed meaning of what we say—and I mean quite apart from what it supervenes upon—may be normative.
13. Immediate Normative Implications of Basic Equality I said in One Another’s Equals that basic equality was a normative position, albeit a very abstract one.¹⁸ I said basic equality had normative implications. It requires that all persons be given the benefit of whatever principles of social justice we come up with. It places limits on the way we make consequentialist calculations (if indeed there is room for such calculations in our normative system): ‘Everyone to count for one, no one for more than one’. It may generate certain fundamental human rights. Basic equality is sometimes expressed in a principle that promises to all persons equal concern and respect. And it may well require in social relations what relational equality is said to require. (That is one of the things we have to figure out: as normative ¹⁸ Waldron (2017: 46).
Jeremy Waldron
73
ideals, how far do the normative implications of basic equality and relational equality overlap?) Since basic equality has these far-reaching normative implications, it will hardly do to say that it is a descriptive statement as opposed to a normative one. Some of the normative implications of basic equality are, as it were, direct and immediate. This is true, I think, of most of the implications outlined in the previous paragraph. I don’t think you understand basic equality if you don’t understand these direct implications of it. Others are more indirect or at any rate require considerable argument. The link between basic equality and luck-egalitarianism (such as it is) is of that kind. We argue for luck-egalitarianism on the basis of basic equality—as indeed Dworkin did¹⁹— but it can hardly be said that a person doesn’t understand basic equality if he doesn’t understand this implication of it. (Indeed, the argument to luckegalitarianism may fail, which is what relational egalitarians have claimed.) In One Another’s Equals, I used the contrast between ‘deep’ and ‘surface-level’ positions to convey this difference.²⁰
14. Basic Equality and Human Dignity Some versions of basic equality are framed in terms of equal human dignity. I say some versions: in One Another’s Equals, I said that attributions of human dignity are usually associated with the conviction that humans rank higher than the other animals, whereas basic equality as such is neutral on this, insisting only that there are no fundamental divisions among humans analogous to those commonly believed to hold as between humans and animals.²¹ I suppose there are ways of looking at dignity that make it a feature—albeit a metaphysical rather than a natural feature—of the beings who have it. Just as rights may be seen as things ‘held’ by persons, so dignity might be seen as a sort of magic amulet or badge or mark of God’s favour. But these are crude understandings. It is better to see both rights and dignity as normative positions that apply to individuals (in virtue of certain descriptive features that they possess) rather than just descriptions of ‘queer’ non-natural features that somehow have normative valence.²² ¹⁹ For example, Dworkin (2002: 12–13). ²⁰ Waldron (2017: 10). ²¹ Waldron (2017: 3–4, 30–35). ²² For this notion of ‘queer’ features, see Mackie (1991: 38).
74
Basic and Relational Equality
15. Moral Status But even on this more sophisticated understanding, attributions of human dignity are not straightforward prescriptions. Sometimes dignity is the content of a right, as in the norm in Common Article III of the Geneva Conventions that any outrage upon the personal dignity of a detainee is prohibited.²³ Sometimes human dignity is said to be the foundation of rights: we are told in the preamble to the International Covenant on Civil and Political Rights that the rights contained in the Covenant ‘derive from the inherent dignity of the human person’.²⁴ Elsewhere I have argued that ‘human dignity’ is best regarded as a status term: human dignity is a status that comprises fundamental human rights rather than being the foundation of rights or just the content of some of them.²⁵ It is a form of moral standing that goes to fundamental issues of respect. It is, so to speak, a heading under which are arrayed a whole set of normative principles and demands. But, like all status terms, ‘human dignity’ operates not just as an abbreviation of the items listed under it. It presents the principles and demands arrayed under it as making sense, together, in virtue of the kind of beings to whom the status in question is attributed— their important characteristics and capabilities and their high standing in the world of sentient creatures. As I said in Section 14, not all claims of basic equality are claims about human dignity (though all claims of human dignity are claims about basic equality). But more generally, basic equality also might be seen as a status claim. Human persons are to be understood as having the same basic moral status. Being of equal worth—something which is based on their possession of (broadly speaking) equal capacities—they are to be accorded equal respect and dealt with as though their lives and their living of their lives matter equally. Basic equality holds that they are morally considerable in the same way. That’s the package of commitment and justification conveyed in assertions of basic equality. The package is a little more complex than the one mentioned ²³ Geneva Conventions 1949 on treatment of prisoners, detainees, and others placed hors de combat (e.g. by injury, shipwreck etc.): ‘III (1) … the following acts are and shall remain prohibited at any time and in any place whatsoever with respect to the above-mentioned persons: (a) violence to life and person, in particular murder of all kinds, mutilation, cruel treatment and torture; (b) taking of hostages; (c) outrages upon personal dignity, in particular humiliating and degrading treatment; … ’. ²⁴ International Covenant on Civil and Political Rights (1966), Preamble: ‘The States Parties to the present Covenant, [c]onsidering that … recognition of the inherent dignity and of the equal and inalienable rights of all members of the human family is the foundation of freedom, justice and peace in the world, [r]ecognizing that these rights derive from the inherent dignity of the human person … ’. ²⁵ See Waldron (2012: 57–61) and Waldron (2015).
Jeremy Waldron
75
in Sections 7 and 11. It is not just normative content packaged tightly with a set of descriptive characterizations on which the prescriptions are supervenient. It is an array of normative positions, supervening upon descriptive characterization, organized as a unity and as the basis of a certain distinctive form of equal standing or considerability in the world.
16. Doesnʼt Relational Equality Do Roughly the Same Work? If you take away the descriptive part of the package, relational equality may not seem to be very different from basic equality. Some or many of the normative things that relational theorists say are quite close to the sorts of normative things that basic equality theorists say. They too talk about equal worth; they too talk about basic moral standing. Or if they don’t say exactly what basic equality theorists say, they seem to be in the same ballpark, competing to occupy roughly the same space.
17. Concept and Conception To explore this further I want to introduce yet another distinction. Will Kymlicka, following Ronald Dworkin, invites us to consider the difference between abstract equality—something that all egalitarian theories have in common—and particular theories which make rival claims about what that abstract equality requires in the organization of a society.²⁶ There is a sort of a ‘plateau of agreement’ on these matters, which gives us what Dworkin would call the concept of equality²⁷; and then there are conceptions of equality, of which luck-egalitarianism/distributivism is one. If that’s our picture, should we regard relational equality as a different and opposed conception? I must confess that, after years of using the concept/conception distinction, I am not sure how it plays out here. (a) On one account, a conception is a (controversial) theoretical working-out of the direct and indirect implications of a given abstract idea in moral and political argument: so luckegalitarianism would be a conception of equality (one which other egalitarians think represents faulty reasoning). (b) On another account, a conception is a way of presenting and understanding the abstract idea itself: so distributivism, a view that approaches equality more or less entirely in terms of ‘who ²⁶ Kymlicka (2002: 3–5). ²⁷ See Dworkin (1986: 90–96).
76
Basic and Relational Equality
gets what’, would be a conception of equality. Relational equality might be thought of as a conception of equality in either of these two senses.
18. Differences of Emphasis as Between Basic and Relational Equality As I understand it, relational equality is presented by its proponents as a pretty comprehensive theory, covering the implications of the ideal of equality for just about all aspects of moral, social, and political life. Others may see it, however, as a partial ideal, concerned as it mostly is with the implications of equality for relations between persons. ‘Equality’, says one relational theorist (Samuel Scheffler), ‘is understood as a social and political ideal that governs the relations in which people stand to one another’.²⁸ He goes on to say that ‘the core of the value is a normative conception of human relations’.²⁹ But maybe not everything morally important is about human relations. For example, Scheffler’s formulation seems to neglect or play down the implications of the equality ideal for the situation of each person compared to others, whether that is a matter of actual social or personal relations between them or not. If X is punished much more severely for a given offence than Y is, then there ought to be some sort of egalitarian concern about the disparity, whatever the impact is on the relationship between X and Y (if indeed they are even known to one another). Basic equality needs to concern itself with the foundations of concerns about comparative justice in addition to the specific concerns about the quality of people’s relations with one another, which are the domain of relational equality. I don’t intend this as a criticism of relational equality. The specific socialrelations concerns of relational equality are important, and it may be that they need to be given more prominence and perhaps even constrain the way we handle equality’s distributive implications. I bring all this up simply to show that the relation between relational and basic equality may be (roughly) that of part to whole.³⁰ One doesn’t need to be a luck-egalitarian or even a distributivist to make this point. Someone interested in equal human rights could make it also: equality of some rights matters even when they have little or nothing to do with the nature of social relations. ²⁸ Scheffler (2003: 31). ²⁹ Scheffler (2003: 31). ³⁰ Just as political equality is not a full account of the equality ideal though it is still important as a part of it, so relational equality may be very important even though it does not offer a full account of equality.
Jeremy Waldron
77
I have the impression that some relational egalitarians would not accept this. Sometimes what they say suggests that they are trying to get us to reconceive equality altogether so that the quality of relations between persons really is its main concern. This certainly seems to be the implication of Elizabeth Anderson’s view that ‘the proper positive aim’ of equality is ‘to create a community in which people stand in relations of equality to others’.³¹ It seems to be the implication too of Scheffler’s view that equality is ‘a moral ideal governing the relations in which people stand to one another’.³² If this is indeed the aim, then relational equality is a distinctive conception of equality and not just an exploration of some of its more important implications.
19. Basic Equality as a Ground for Relational Equality But some of the other things that are said by relational egalitarians convey the thought that one might also articulate the foundational idea of equality in (initially) non-relational terms. So, for example, Scheffler also says the following about equality: ‘A s a moral ideal, it asserts that all people are of equal worth, and that there are some claims that people are entitled to make on one another simply by virtue of their status as persons’.³³ The idea seems to be that one needs to get hold of that first and then (if one wants to) one can build up a relational account on that basis.³⁴ Similarly, Scheffler suggests, the onus is on luck-egalitarians to show how their conception is anchored ‘in some version of that [basic] ideal, or in some other comparably general understanding of equality as a moral value or normative ideal’.³⁵ On this account, which is the account I favour, the phrase ‘basic equality’ describes the foundation or anchor point here: the idea of equal worth or human dignity or basic status and considerability. The relational egalitarian acknowledges that but doesn’t say a whole lot about it. Which is okay because there is a division of labour, there are many tasks to perform, and the relational egalitarian has other fish to fry. ³¹ Anderson (1999: 289) (my emphasis). I am assuming that ‘relations of equality’ here means something like the egalitarian character of real social relations, not just the arithmetical relation of equality that may obtain between persons who have nothing to do with each other. ³² Scheffler (2003: 21). ³³ Scheffler (2003: 22). ³⁴ I think this is what Scheffler means when he says (2003: 33) that ‘equality is most compelling when it is understood as a social and political ideal that includes but goes beyond the proposition that all people have equal moral worth’. ³⁵ Scheffler (2003: 23).
78
Basic and Relational Equality
20. Conclusion So that is where I would like to leave it. Relational equality, in my view, may be distinguished from basic equality in three main ways: (i) Basic equality is more general than relational equality. Relational equality is concerned specifically with the quality of personal, social, and political relations, whereas basic equality is concerned (though in a foundational way) with every aspect of practice on which the ideal of equality has a bearing. (ii) Basic equality is more foundational than relational equality. It concerns our understanding of the grounds or anchor points of every plausible egalitarian claim, particularly propositions about equal worth and moral considerability. (iii) To the extent that relational equality concerns itself with any of the foundational content of basic equality, it does so in a way that is less preoccupied than basic equality is with factual underpinnings, on which the normative grounds of basic equality are said to be supervenient. This is a difference of interest or preoccupation. It is important nonetheless.
References Anderson, Elizabeth. 1999. ‘What Is the Point of Equality?’. Ethics, 109: 287–337. Dworkin, Ronald. 1983. ‘Comment on Narveson: In Defense of Equality’. Social Philosophy and Policy, 1: 24–40. Dworkin, Ronald. 1986. Law’s Empire. Cambridge: Harvard University Press. Dworkin, Ronald. 2002. Sovereign Virtue: The Theory and Practice of Equality. Cambridge: Harvard University Press. Fourie, Carina, Schuppert, Fabian, and Walliman-Helmer, Ivo, eds. 2015. Social Equality: Essays on What It Means to Be Equals. Oxford: Oxford University Press. International Covenant on Civil and Political Rights (1966). U.N. General Assembly Resolution 2200A (XXI), 16 December 1966. Kymlicka, Will. 2002. Contemporary Political Philosophy: An Introduction, 2nd ed. Oxford: Oxford University Press. MacDonald, Margaret. 1984. ‘Natural Rights’. Reprinted in Theories of Rights, edited by Jeremy Waldron, 21–40. Oxford: Oxford University Press. Mackie, J. L. 1991. Ethics: Inventing Right and Wrong. Harmondsworth: Penguin Books.
Jeremy Waldron
79
McDowell, John. 1981. ‘Non-Cognitivism and Rule-Following’. In Wittgenstein: To Follow a Rule, edited by Steven Holtzman and Christopher Leich, 141–162. Routledge. Rawls, John. 1999. A Theory of Justice, new ed. Cambridge: Harvard University Press. Scheffler, Samuel. 2003. ‘What is Egalitarianism’. Philosophy and Public Affairs, 31: 5–39. Waldron, Jeremy. 2012. Dignity, Rank and Rights. Oxford: Oxford University Press. Waldron, Jeremy. 2015. ‘Is Dignity the Foundation of Human Rights?’. In The Philosophical Foundations of Human Rights, edited by S. Matthew Liao, Massimo Renzo, and Rowan Cruft, 117–137. Oxford: Oxford University Press. Waldron, Jeremy. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge: Harvard University Press.
3 Should People Who Are Moral Equals Relate as Social Equals? Should People Who Are Not Moral Equals Relate as Social Unequals? Kasper Lippert-Rasmussen
1. Introduction In what is probably the most prominent piece of work in the booming literature on relational egalitarianism, Elizabeth Anderson (1999: 312) writes: Inegalitarianism asserted the justice or necessity of basing social order on a hierarchy of human beings, ranked according to intrinsic worth. Inequality referred not so much to distributions of goods as to relations between superior and inferior persons. Those of superior rank were thought entitled to inflict violence on inferiors, to exclude or segregate them from social life, to treat them with contempt, to force them to obey, work without reciprocation, and abandon their own cultures. These are what Iris Young has identified as the faces of oppression: marginalization, status hierarchy, domination, exploitation, and cultural imperialism. Such unequal social relations generate, and were thought to justify, inequalities in the distribution of freedoms, resources, and welfare. This is the core of inegalitarian ideologies of racism, sexism, nationalism, caste, class, and eugenics … Egalitarian political movements oppose such hierarchies. They assert the equal moral worth of persons … all competent adults are equally moral agents: everyone equally has the power to develop and exercise moral responsibility, to cooperate with others according to principles of justice, to shape and fulfil a conception of their good.
This passage contains three ideas. First, all human beings—or all persons, at least—are moral equals, and thus at a fundamental level they are owed, in a sense that doubtless needs to be explained, the same moral concern. Let us Kasper Lippert-Rasmussen, Should People Who Are Moral Equals Relate as Social Equals? Should People Who Are Not Moral Equals Relate as Social Unequals?. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0004
82
Should People Who Are Moral Equals Relate as Social Equals?
call this idea basic moral equality. Different theorists can and do give different accounts of what makes all humans moral equals. As the excerpt indicates, Anderson takes the Rawlsian view that we are moral equals because we have the same fundamental moral powers.¹ Second, the quoted passage contains the idea of a non-hierarchical social order, where no one relates to others as a superior to inferiors and vice versa. For example, no one relates to others as if she has a higher social standing than others in virtue of her race, gender, or nationality. As David Miller (1998: 31), quoting Michael Walzer and with unapologetic androcentrism, puts it: relational egalitarians want a ‘society of misters’, not a ‘society of misters, Sirs, His Highnesses, and Her Majesty etc’. I will refer to this ideal of a non-hierarchical social order as the anti-hierarchical ideal. Third, the excerpt expresses the idea that what justifies the antihierarchical ideal is the basic moral equality of people. For it is very naturally read as saying that egalitarians oppose hierarchies of race, sex, nationality, caste, class, or people with ‘good’ versus ‘bad’ genes because they ‘assert the equal moral worth of persons’, and as saying that antiegalitarians favour hierarchical social relations because they believe people have unequal moral worth. Call this claim about the justificatory relevance of basic moral (in)equality to the (anti-)hierarchical ideal the claim of justificatory relevance (cf. Bengtson and Lippert-Rasmussen, 2023; Waldron, this volume: 66). The same three elements can be found in the work of other relational egalitarians. Samuel Scheffler (2003: 21–22), for example, writes: Equality, as it is more commonly understood, is not, in the first instance, a distributive ideal, and its aim is not to compensate for misfortune. It is, instead, a moral ideal governing the relations in which people stand to one another. Instead of focusing attention on the differing contingencies of each personʼs traits, abilities, and other circumstances, this ideal abstracts from the undeniable differences among people. It claims that human relations must be conducted on the basis of an assumption that everyoneʼs life is equally important, and that all members of a society have equal standing. As Anderson insists, in defending a version of this ideal, equality so understood is opposed not to luck but to oppression, to heritable hierarchies of social status, to ideas of caste, to class privilege and the rigid stratification of classes, and to the undemocratic distribution of power.
¹ The two distinguishing features of the moral persons in question are that they are ‘capable of having (and are assumed to have) a conception of their good’ and that they are ‘capable of having (and are assumed to acquire) a sense of justice’ (Rawls 1971: 505; see Arneson, this volume).
Kasper Lippert-Rasmussen
83
Here are the three elements: the idea of basic moral equality (‘everyone’s life is equally important’); the anti-hierarchical ideal (‘all members of a society have equal standing’); and the claim of justificatory relevance (it is because ‘everyone’s life is equally important’ that all members of a society must have equal standing and that hierarchies are unacceptable).², ³ Despite the centrality of the idea of basic equality, the anti-hierarchical ideal, and the claim of justificatory relevance, relational egalitarians have not scrutinized them sufficiently closely. In particular, they have failed to distinguish clearly between the ideas of basic moral equality and equal social standing. Nor have they explained why, exactly, our moral equality rules out a hierarchical social order, or at least hierarchical social orders of the kind to which Anderson refers. No doubt, the idea of basic equality and the anti-hierarchical ideal are connected. Thus our moral equality may somehow imply that we should relate as social equals—at least in some ways and in some contexts. Also, if the moral status of people varied significantly and uniformly with, say, their race, it would be more difficult, at least, to justify a racially non-hierarchical society. However, none of this amounts to a strong logical connection. In my chapter, I pursue three aims. First, I propose analyses of what it is to be moral equals and of what it is to be social equals. Second, drawing on these, I probe the notion that our being moral equals justifies the moral requirement that we relate as social equals. Third, I ask whether there are ways of our not being moral equals such that the requirement that we relate
² Admittedly, Scheffler describes the idea that ‘everyone’s life is equally important’ as ‘an assumption’. However, it is unclear why he does not characterize it as an assertion instead. First, one cannot ground any normative demand, for example, a demand about how human relations should be conducted, on a mere assumption. Second, shortly after the quoted passage he writes: ‘A s a moral ideal, [the ideal of equality] asserts that all people are of equal worth and that there are some claims that people are entitled to make on one another simply by virtue of their status as persons. As a social ideal, it holds that a human society must be conceived of as a cooperative arrangement among equals, each of whom enjoys the same social standing. As a political ideal, it highlights the claims that citizens are entitled to make on one another by virtue of their status as citizens, without any need for a moralized accounting of the details of their particular circumstances’ (Scheffler 2003: 22). Strictly, here, it is the ideal of equality that asserts that ‘all people are of equal worth’, but since it is this understanding of the ideal of equality that Scheffler defends, I take it that he asserts this too. Third, Scheffler sees a tight relation between basic equality and the social ideal of equality: ‘When we say, for example, that a friendship or a marriage should be a relationship of equals, we do not mean merely that the participants are of equal moral worth but also that their relationship should have a certain structure and character’ (2003: 33). But if part of what it means for two parties to relate socially as equals is indeed that they are moral equals, then basic equality cannot justify the ideal of social equality, since it is a component of the justificandum. Things are different if social equality simply requires that the parties to the relation are thought of as moral equals (whether or not they are so). ³ In another passage giving clear expression to the idea of justificatory relevance, Scheffler (2003: 38) characterizes his preferred form of egalitarianism as one ‘that begins from the question of what relationships among equals are like and goes on from there to consider what kinds of social and political institutions are appropriate to a society of equals’.
84
Should People Who Are Moral Equals Relate as Social Equals?
as social equals can be justified even if, in one of these ways, we are not moral equals.⁴ I defend an affirmative answer to this question, suggesting that antiinegalitarian views that are neutral on whether all persons are moral equals provide a stronger justification for the main forms of social equality that relational egalitarians care about than moral equality-based justifications do. If this defence is successful, then it is important. For one thing, it means that, contra the many theorists who assume otherwise, our commitment to social equality is robust across different claims about moral status. But it is also important because, despite the popularity of the view that all persons are moral equals, the philosophical literature on moral status is inconclusive in this area: some philosophers reject the very idea of moral status (Floris and Kirby, this volume: 14–16; Lippert-Rasmussen 2022(a); Sangiovanni, this volume; Singer 2011; 1990); others, who accept it, for equally powerful reasons, are doubtful whether all humans possess that status—one that is in some sense higher than that of non-humans (for example, Arneson 1999; 2015; Arneson, this volume: 305–310; Husi 2017; McMahan 2002; Sher, this volume: 287–289; for a reply, Parr and Slavny 2019).
2. Moral and Social Equality It is often asserted that all human beings, or at least those of them who are persons, are moral equals. Richard Arneson expresses the view as follows: All humans have an equal basic moral status. They possess the same fundamental rights, and the comparable interests of each person should count the same in calculations that determine social policy. Neither supposed racial differences, nor skin color, sex, sexual orientation, ethnicity, intelligence, or any other differences among humans negate their fundamental equal worth and dignity. (Arneson 1999: 103; cp. Arneson, this volume: 305–306)5
It is important here that people have the same basic moral status, since that is consistent with two people having a different non-basic moral status. Suppose ⁴ Sangiovanni poses roughly the same questions in his contribution to this volume, and, as far as I can tell, we give roughly the same answers though for different reasons (which is not to say that I reject Sangiovanni’s reasons). As Sangiovanni (this volume: 119; see also Floris and Kirby, this volume: 9) puts it: ‘it is not sufficient to cite equal moral status in defense of equal social status, and it is not necessary either’. ⁵ The second sentence spells out one way in which all humans can have ‘equal basic moral status’. It is consistent with there being other ways. This means that it, the second sentence, does not capture what it is to have ‘equal basic moral status’. Like Arneson’s piece, much of the literature on ‘equal basic moral status’ does not say what ‘equal basic moral status’ is. Rather, like Arneson, the writers I have in mind mention ways in which equal basic moral status is instantiated, or simply focus on what grounds equal basic moral status. Below I provide a definition of this kind.
Kasper Lippert-Rasmussen
85
I am about to invent a novel vaccine that is effective against any possible flu pandemic and would prevent large-scale suffering. For some reason, I can invent this vaccine only if my interests are given priority over the similar interests of others in the calculations that determine social policy, and no one else will invent the vaccine if I do not. In this case, it might be morally justified to give my interests greater non-basic weight than those of other people. But that is consistent with my interests not having any greater basic moral weight than the interests of others—it is because everyone’s interests in avoiding disease count and, at a fundamental level, count equally, other things such as age being equal, that, under the circumstances, it is justified to give nonfundamental priority to my interests. At the input side—the factors that go into the function that determines what one morally ought to do—my interest in the vaccine being developed counts for the same as the equally strong interests of others, even if, at the output side, morality gives greater weight to my interests because of how doing so happens to benefit everyone (see Sangiovanni, this volume: 113–114; Sher, this volume: 295–303). Another important feature of Arneson’s characterization is that it is intended to be ecumenical across two markedly different moral theories, both of which are supposed to ascribe equal basic moral status to people. Specifically, it is meant to be such that both consequentialists (for example, utilitarians who assert that ‘the comparable interests of each person should count the same’) and rights-based non-consequentialists (who assert that ‘A ll humans … possess the same fundamental rights’) subscribe to basic moral equality (cf. Sangiovanni, this volume: 103–110).⁶ Though Arneson’s characterization spans the consequentialist/nonconsequentialist divide, there are theories with which it cannot be combined. Presumably, virtue ethicists might subscribe to basic moral equality in part, or in whole, because they think that what, at a basic level, counts as and grounds virtue for any human being is the same and that, at a fundamental level, the perfection of each person matters equally no matter who they are. Similarly, theorists, for example in the contractarian tradition, who think that (the withholding of ) consent has the same normative consequences across all humans subscribe to basic moral equality independently of the weight they give to different people’s comparable interests, and whether or not they think that all humans have the same basic moral rights (cf. Kirby 2018). These points suggest that while Arneson’s characterization fits two prominent strands of moral thinking nicely, it fails to capture the very notion of ⁶ Typical consequentialists can agree that all humans have the same fundamental moral rights (that is, none). Rights-focused non-consequentialists typically agree that interests are relevant to questions about what should be done. Those who deny that interests are relevant at all to what should, morally, be done can still agree that ‘the comparable interests of each person should count the same’ (that is, not at all).
86
Should People Who Are Moral Equals Relate as Social Equals?
moral equality: that is to say, it fails to capture what it means for all people to have equal basic moral status.⁷ Hence, in its stead I propose that the following proposition captures what basic moral equality is. (Note that this is distinct from what counts, within specific moral theories, as moral equality, or what makes all humans moral equals.) ʻAll humans are moral equalsʼ means that the same basic moral considerations apply to all humans.8
This definition should be congenial even to virtue ethicists, for it allows them to affirm (or deny) that virtue, and the moral significance of being virtuous, does not vary across people in accordance with, say, class, gender, race, or caste. It should also be acceptable to moral theorists preoccupied with normative powers, since it allows them to claim that individuals’ moral powers, and their moral significance, do not vary across people in such a way that the consent of some people (patriarchs) counts for more than the consent of others (women). Let us now move on to the notion of social equality. Obviously, there are various accounts on offer. In what follows I shall basically rely on a social norms-focused account offered recently by Han van Wietmarschen:⁹ a social position A is ʻhigher thanʼ or ʻaboveʼ social position B if and only if, for the participants in the relevant social network, when they display the norm-required complexes of attitude and behaviour they thereby and to that extent value the occupants of A more than the occupants of B. (van Wietmarschen 2021: 6)
Assuming that if, and only if, A is no higher than B (and vice versa) then A and B are equal social positions, it follows that A and B are equal social ⁷ Arneson’s (1999: 105) main purpose in the work in question is to demonstrate the intractability of what he calls ‘Singer’s problem’. For that purpose it is not important to provide a completely general definition of basic moral equality. ⁸ This is consistent with their not being equal in terms of what grounds the basic moral considerations that apply to them. One reason why is that, as a logical possibility, the same basic moral considerations might apply to differential individuals even though the reason why they do so is due to these individuals having different capacities etc. that, however, grounds the applicability of the same basic, moral considerations. ⁹ Even if it turns out that van Wietmarschen’s account does not capture everything that we mean by social hierarchy—say, it does not capture the presence of asymmetric relations of de facto authority or power not involving differential valuing of persons and that is part of what we mean by social hierarchy (van Wietmarschen 2021: 8–10; see also footnote 11)—it surely captures a core aspect of it. Moreover, some of the main differences between van Wietmarschen’s account of social hierarchy and other accounts have no significant bearing on my arguments in Sections 3 and 4. For instance, people can have de facto authority or power over others who are their moral equals and, thus, even if we adopt a de facto authority— or a power-focused account of social hierarchy—we should still accept that moral equals can relate as social equals.
Kasper Lippert-Rasmussen
87
positions if, and only if, it is not true, of the participants in the relevant social network, that when they display the norm-required complexes of attitude and behaviour they thereby and to that extent value the occupants of A more than they value the occupants of B.¹⁰ Or, to put this point in a simpler way: a social position A is equal to social position B if, and only if, when participants in the relevant social network display the norm-required complexes of attitude and behaviour they thereby and to that extent value the occupants of A and B equally.11
As an illustration of non-hierarchical social ordering, consider a team in an operating theatre with medical doctors each of whom performs a distinctive, insufficient, but necessary and equally skill-requiring task as part of an operation, where, accordingly, the norm-required complexes of attitude (for example, admiration) and behaviour (for example, praising) associated with all the positions are relevantly similar (though not necessarily identical, since, for example, the grounds of admiration might differ because the relevant doctors are supposed to occupy different, but equal, social positions). By a ‘social network’ van Wietmarschen means a set of social relations between a number of individuals who are socially related. To explain the notion of a relevant social network, van Wietmarschen offers the example of a hospital ward: It is often possible to … identify structural similarities between the sets of social relationships in which different participants in the network stand to the other participants. On a hospital ward, all of the nurses may relate in similar ways to one another, and to the other participants in the network—medical assistants, residents, attending physicians, and so on. The term ʻsocial positionʼ refers to such structurally similar locations in social networks. (van Wietmarschen 2021: 2)
To see what participants ‘valuing A more than B’ involves, consider van Wietmarschen’s example of a prestige hierarchy among professional tennis players: ¹⁰ The assumption is not an idle one. A and B could be incommensurable social positions (see Section 4). ¹¹ The focus on the extent to which the ‘occupants of A and B’ are valued is probably too narrow (recall footnote 9). Suppose participants esteem etc. occupants of A and B equally but are more inclined to display the norm-required complex of attitude and behaviour consisting of accepting an order and executing it in response to an order issued by an occupant of A than in response to an order issued by an occupant of B. Plausibly, A is hierarchically superior to B, even though it is unclear that the former is valued more than the latter. I set aside this complication as it makes no difference to my argument below that moral and social status can diverge—one can have social authority over another without having moral authority over them, as is true of men and women in patriarchal societies.
88
Should People Who Are Moral Equals Relate as Social Equals?
My view would say that such a hierarchy is constituted by a set of social norms asking the participating players to respond to one another in various ways. Most importantly, participants are expected to display different patterns of admiration, emulation, praise, attention, and so on, towards different people with whom they interact. When participants respond to different players with different levels of praise and admiration, they thereby and to that extent value them more or less as tennis players. Players are located in higher and lower positions in the hierarchy depending on how highly they are socially required to be esteemed in this sense. (van Wietmarschen 2021: 6)
Having fleshed out the notions of basic moral equality and social equality, I want to stress that the two notions are distinct: people can be moral equals even if they are not social equals, and they can be social equals even if they are not moral equals. Perhaps the first of these possibilities makes more sense initially. We often criticize the way people—say, nobles and serfs, men and woman, freemen and slaves—are, or were once, obliged to endure social inequality precisely because they are all moral equals. This is possible because the properties in virtue of which people are moral (un)equals are different from the properties in virtue of which they are social (un)equals. Admittedly, it is formally possible for people to be moral equals in virtue of being social equals (or, more oddly, in virtue of being social unequals!), and therefore it could be held that, provided people are not social equals, they are not moral equals either. However, as far as I am aware no one has taken this position—one reason for this being, presumably, that by doing so one would rule out the possibility of criticizing an existing social order in the light of facts about people’s moral equality. Here Anderson’s Rawlsian view on the matter is quite common. In essence it is that all people are moral equals in virtue of their capacity to form a sense of justice and a conception of the good. These are not properties in virtue of which, typically, people are, or are not, social equals—whether on van Wietmarschen’s account or the accounts offered by others.¹², ¹³ ¹² This is a contingent truth. We can imagine a group of people living in a strict egalitarian community in which there are no differential ‘patterns of admiration, emulation, praise, attention, and so on’. These people are all admired, emulated, praised, attended to, and so on, solely on their basis of moral personhood. Such a social world would be very different from ours not just in terms of how people interact, but also because the properties that ground moral status (according to Rawlsians) would also be the properties grounding social standing (in van Wietmarschen’s sense). This possibility—that the same properties ground moral and social standing—does not show that the two forms of standing are identical. After all, the properties at issue ground moral standing regardless of people’s response to them but social standing only because people respond to them in a particular way. ¹³ As Rawls (1971: 505, n. 30) puts it: ‘The existence of these attributes [in virtue of which moral persons, unlike non-human animals, are entitled to equal justice] depend solely on certain natural attributes the
Kasper Lippert-Rasmussen
89
It is also possible for people to relate as social equals on van Wietmarschen’s account even if they are not moral equals. Suppose that the critics of the Rawlsian notion of a range property are right. The capacities that underpin our moral personality are indeed what ground our moral status. However, people all of whom have a moral personality in Rawls’s sense vary in terms of the properties that their moral personality supervenes upon. Pace Rawls, this results in their having unequal moral status—for example, those with a capacity for a more intricate sense of justice and a more coherent conception of the good have higher moral status than those with a cruder sense of justice and a simple, relatively unreflectively endorsed conception of the good. Surely these people could form a social, egalitarian community where the normrequired complexes of attitude and behaviour do not vary across different social positions. For example, there need not be any differences across people in terms of the degree to which they are socially required to be admired, and so forth.¹⁴ This concludes my exposition of the difference between the notion of moral equality and the notion of social equality.
3. How Does Moral Equality Justify Social Equality? Relational egalitarians see an intimate connection between the fact that all people are moral equals and the fact that justice requires people to relate as social equals.¹⁵ In this section, I consider how tightly connected these two facts are. I will do this by asking two questions. First, does basic moral equality justify moral requirements other than the requirement to relate socially (also) as equals? Second, does basic moral equality justify the requirement that we relate as social equals? Because my answer to the first question is affirmative and my answer to the second is negative, I take the view that the two facts above are more loosely connected than is suggested by Anderson’s opening passage (which, as I have indicated, is not unrepresentative of how philosophers generally view the matter). I set out a further reason for taking this view in Section 4. presence of which can be ascertained by natural reason … The existence of these attributes … is established independently of social conventions and legal norms’. ¹⁴ It is natural to assume that if people are moral unequals in this sense they will also be social unequals. However, as I discuss in Section 4, there is no simple entailment here, and perhaps people should (morally) relate as social equals even if they are not moral equals. ¹⁵ Or at least the fact that justice requires that there are no social hierarchies, or at least no social hierarchies based on ‘inegalitarian ideologies of racism, sexism, nationalism, caste, class, and eugenics’. The third of these requirements can be satisfied even if none of the other two are. The second can be satisfied even if the requirement that we all relate as equals is not. However, the satisfaction of this requirement means that the second and the third are satisfied as well.
90
Should People Who Are Moral Equals Relate as Social Equals?
Let me start with my view that insofar as basic moral equality justifies the claim that people should relate socially as equals, it justifies other requirements than that as well. To see the force of this view, consider a purified case of: the medieval raid, a common form of warfare in which aristocratic knights—on horseback and fully armed—would pillage large areas of farm land, burning crops and villages, and attacking unarmed and defenceless peasants … Those peasants would have lords of their own, and would stand in hierarchical relationships to those lords, but they would confront a foreign knight not as someone owed loyalty or deference, but simply as someone with the capacity to apply overwhelming force. (van Wietmarschen 2021: 9)
For van Wietmarschen’s purposes, the important feature is that the case illustrates domination without social hierarchy. For my purposes, the important thing is that medieval raids, in a purified variant, would be cases of harm and domination not just without social hierarchy, but without social relations at all.¹⁶ By ‘purified’ I mean the following. No doubt, historically medieval raids did involve social relations of some sort between raiding knights and raided peasants. For example, the raiding knights would not just kill peasants. They would also humiliate them (and sometimes spare them). And the raided peasants would not just try to resist or flee the brute force of the armoured knights. They would also plead for their lives or even try to reason with the raiding knights, and so on. However, in a purified medieval raid of the sort I am envisaging the knights raid a peasant village in the same way that sailors hunted Mauritius’s population of dodos to their extinction, and the raided peasants respond to the raiding knights in the way they would to a pack of raiding wolves. Neither party engages in any attempt to communicate with the other. Here is what I believe the case of the purified medieval raid shows. It is possible to act unjustly towards others even if one is not socially related to them—in my, conceptually speaking, purified raid the knights treat the ¹⁶ I realize it is tricky to state the necessary and sufficient conditions under which two persons (or animals) are socially related. If you are inclined to think that there is a social relation between the raiding knights and the peasants in the purified example, tweak it further so that there is no visual contact between the parties (for example, the knights, now technologically advanced, drop bombs on the peasants from a height of 10 kilometres and then land to collect their possessions). Surely, there are no social relations between rights-violating bomber-knights and peasants.
Kasper Lippert-Rasmussen
91
peasants unjustly, but they do not relate to them socially.¹⁷ Moreover, if the requirement of justice to relate socially as equals is justified by moral equality, then so are the requirements of justice that the knights violate in the purified medieval raid. Or at least if they are not, then some duties of justice that are very stringent and more stringent than most—‘Don’t kill innocent people’—are not grounded in moral equality, in which case the appeal to moral equality plays a less central role to an account of what justice requires than that ascribed to it by many relational egalitarians. Hence, if relational egalitarians ground the requirement of justice to relate as social equals in people being moral equals, then they should also accept that there are additional requirements of justice that do not pertain to people’s social relations. By its own lights, relational egalitarianism is an incomplete theory of justice.¹⁸ I now turn to my second question: Does basic moral equality justify the claim that we should relate as equals? An important initial distinction that needs to be made here is between the claim that moral equality justifies social equality necessarily and the claim that it does so contingently only. Much here hangs on, first, what we take moral equality to be and, second, how we think moral equality enters the justification of social equality. To see this, let us briefly review Richard Hare’s (1979) discussion of slavery, a paradigm state of social inequality.¹⁹ ¹⁷ It might be suggested that it is impossible to act unjustly towards someone to whom one cannot relate socially, and that this view is not defeated by the case of the purified medieval raid because even if there are no social relations between knights and peasants, there could be. I reject this move, since, in my view, the knights would act unjustly towards the peasants even if we specify that for some reason the parties to the raid would fail in establishing social relations even if they tried (for example, as a result of the presence of a powerful prince who is opposed to such relations and, overseeing the raids, is ready to intervene should there be any signs of such relations being about to form). ¹⁸ Recall that according to Scheffler (2003: 21–22, quoted in the introduction) equality is a ‘moral ideal governing the relations in which people stand to one another’, where by ‘relations’ presumably he means social relations, since otherwise it would be unclear in which way his understanding of the moral ideal of equality is different from that of distributive egalitarians. (In a less specific sense of ‘relations’ they too are concerned with relations in which people stand to one another, that is, relations of distributive inequality.) An alternative interpretation of what the purified medieval raid shows says that, plausibly, the knights do not regard the peasants as equals—after all, they do not see themselves as violating the peasants’ rights by killing them. Since justice requires that we relate to each other as equals, and relating as equals requires regarding others as equals, the purified medieval raid shows that, unlike what central relational egalitarians have claimed, the requirement of relational equality governs causal interactions between people who are not socially related. Another alternative interpretation expands the notion of social relations and says that if someone can violate someone else’s rights, then ipso facto they are socially related. In my view, the notion of being socially related that is required for this attempt to rescue the relational egalitarian claim that justice pertains to social relations only, however, is implausibly thin and not in line with how relational egalitarians otherwise think about social relations (recall footnote 16). Specifically, it implies that the notion of rights violation comes to enjoy a certain explanatory primacy over the notion of social relations when it comes to spelling out the requirements of relational egalitarianism. ¹⁹ ‘The slave is so called first of all because he occupies a certain place in society, lacking certain rights and privileges secured by the law to others, and subject to certain liabilities from which others are free. And secondly, he is the slave of another person or body (which might be the state itself ’ (Hare 1979: 105). This description of a slave most naturally concerns the Hohfeldian legal incidents that slaves, unlike citizens, do not enjoy and not the weight given to slaves’ interests ‘in calculations that determine social policy’, to refer
92
Should People Who Are Moral Equals Relate as Social Equals?
Hare is a preference utilitarian. Hence, he affirms moral equality in the sense that he thinks that the satisfaction (and the non-satisfaction) of preferences counts morally in strict proportion to nothing but the strength of the preference—and, in particular, regardless of whose preferences they are. Moreover, while he takes preference utilitarianism to imply that, in any realistic circumstances, slavery is not justified (because it invariably fails to maximize the sum of preference satisfaction), he also concedes that in logically possible circumstances it would be wrong to abolish slavery (Hare 1979: 115).²⁰ Could Hare be a relational egalitarian? Well, he could, given the truth of his assumption about the overwhelmingly likely consequences of slavery in the world as it happens to be, as long as all it takes to be a relational egalitarian is to object to social hierarchies in the world as it happens to be. If it takes more than that—if the relational egalitarian is required to hold that social hierarchies in any world are pro tanto wrong—then he is not. Those who insist there is a tight connection between moral and social equality typically would not be satisfied with the idea that moral equality contingently justifies social equality, even if it does so relatively robustly across a significant range of empirical circumstances. Hence, it is the latter, ‘any world’ view that has the greatest interest here. Does anyone subscribe to the view that social hierarchies among moral equals are in one respect unjust—perhaps even almost always unjust all things considered—whatever the empirical circumstances? Yes. Elizabeth Anderson takes this view. On her view, what makes an act, practice, etc. wrong is the fact that it expresses an attitude to people that is morally wrong, where ‘“[e]xpression” refers to the ways that an action or a statement (or any other vehicle of expression) manifests a state of mind’ (Anderson and Pildes 2000: 1506). Arguably, slavery expresses the attitude that people are not moral equals. Again, arguably, because they are in fact moral equals slavery is wrong on an expressivist theory of wrongness. Moreover, since it is necessarily the case that people are moral equals, and since what slavery expresses does not vary across social worlds, it is arguable that expressivists
to Arneson’s characterization of moral equality cited above. Admittedly, a lack of rights and powers tends to go hand in hand with a lower weight being assigned to one’s interest by social planners. But hierarchies of social value or esteem are analytically distinct from hierarchies of power. I thank the editors of this volume for emphasizing this point. ²⁰ Hare accepts that this is a highly counterintuitive implication of his preference-utilitarian theory, but for reasons that need not detain us here he does not consider this an objection to the theory. In fact, he considers it a weakness of the ‘intuitionism’ the relevant relational egalitarian position involves that ‘it has lost contact with the actual world with which the intuition it relies on were designed to cope’ (Hare 1979: 118).
Kasper Lippert-Rasmussen
93
like Anderson should submit that the social hierarchy which slavery consists in is non-contingently wrong. This expressivist, moral equality-based defence of the non-contingent wrongness of slavery has two weaknesses. First, it is not clear that slavery always expresses the attitude towards people that they are moral unequals. Second, even when it does, this might not be the reason slavery is wrong. In support of the first worry, we can consider a society of utilitarians who find themselves in the sort of highly counterfactual circumstances under which slavery is morally justified on the grounds offered by Hare. Let us suppose they are all persuaded by the arguments in Hare’s paper on slavery and consensually implement slavery for that reason. Under the highly contrived conditions that obtain surely the slavery does not express the attitude that the slaves and the slave holders’ states of mind include a denial of moral equality. On the contrary, under these conditions slavery expresses a deep commitment to the idea that the equally weighty interests of all count equally, morally speaking, and, thus, that no one has higher intrinsic worth than anyone else. One might respond that slavery still expresses an attitude to people that is morally wrong—for example, the attitude that people are simply containers of welfare. However, that—even if it is true—does not rescue the relational egalitarian objection to slavery (that is, the objection that slavery expresses the attitude that people are moral unequals). Alternatively, one might tweak the concept of slavery. One might suggest that the contrived situation I have described is not really slavery, because the relevant institution of ownership of other people arose consensually against the backdrop of a unanimous conviction that this was morally required. This is strikingly unlike actual (historical and present-day) slavery, which arises through brute force and oppression, and which involves institutions of a kind that do indeed noncontingently express the attitude that people are moral unequals. While this may be right, it amounts to a revision of our conception of slavery, as it is not uncommon for philosophers to discuss the wrongness of voluntary slavery. It is also a partial retreat, since it requires us to concede the possibility that not all forms of ownership of other people are unjust after all. This is not what one would expect relational egalitarians to tell us.²¹ ²¹ One can also imagine a society where people believe that they ought to take turns being slaves, and where the reason people believe everyone should be a slave for the same, limited amount of time is that they believe people are moral equals. Here taking-turns slavery seems to express the belief in basic equality, in one way at least. It might be responded here that even though my arguments might apply to some forms of relational egalitarianism, other forms are immune to them. Specifically, relational egalitarian views according to which it is a basic moral fact that it is unjust for some people to be under the power of others (whether their being so derives from an act of consent in a situation of equal moral powers) are immune to my arguments. In response, however, I could simply concede this point. In this chapter, I address forms
94
Should People Who Are Moral Equals Relate as Social Equals?
To my mind, the second challenge to an expressivist basis for relational egalitarianism (or, for that matter, any other moral theory) is even stronger. Essentially, this says that by explaining the wrongness of social hierarchies such as slavery in terms of the fact that they express a denial of moral equality, we get the right result for the wrong reasons. Yes, slavery is wrong. But is that because it expresses an objectionable attitude? Surely, first and foremost, it is wrong because it affects slaves in a more direct way: it denies them freedom and almost always harms them physically and mentally.²² My next question is whether relational egalitarians object to all forms of social hierarchy or only some. Plainly, social hierarchies are nearomnipresent in our social world. Hence, even if we managed to eliminate all of the hierarchies based on ‘inegalitarian ideologies of racism, sexism, nationalism, caste, class, and eugenics’ (Anderson 1999: 312), lots of social hierarchies of the kind characterized in the previous section might remain. These might include social hierarchies between professors and students, judges and defendants, colonels and lieutenants, cool kids and average kids, and, perhaps most intriguingly of all, children and their parents (Scheffler 2003).²³ Call this (with a bit of licence) the omnipresence problem. The problem, of course, is not just that there are a lot of social hierarchies, but that while some of them seem clearly morally wrong, others do not (Wolff 2019). Basically, two responses to the omnipresence problem seem to be the most promising. In the first, relational egalitarians would restrict their opposition to social hierarchy. For example, they might say that while some forms of social hierarchy, like those mentioned by Anderson, are unjust, others are not. It would then be necessary, of course, to explain what distinguishes wrongful from innocent social hierarchies—but perhaps this is a challenge that anyone, relational egalitarian or not, faces. Second, relational egalitarians might persist in opposing all social hierarchies, but explain that, in virtue of this or that feature, different social hierarchies are wrong to differing degrees. of relational egalitarianism that see duties of equality as grounded in some kind of basic equality between persons. ²² A weaker view, which might not involve this mistake, simply says that expressivist concerns capture one of several wrong-making features of slavery. ²³ Children are an especially interesting case for the following reason. Clearly, they are not, and justifiably so, the social equals of adults, including their parents. Inter alia parents have the legal authority to make a lot of decisions on behalf of their children. Accordingly, relational egalitarianism based on basic moral equality seems to face a trilemma at this point. That is, the following claims, making up an inconsistent set, seem to be true: (1) If people are moral equals, it is unjust if they do not relate as social equals. (2) Children and adults are moral equals. (3) It is not unjust that children and adults do not relate as social equals. We must give up at least one of these.
Kasper Lippert-Rasmussen
95
They might want to say all the allegedly innocent hierarchies are in one respect wrong, though in different degrees, just as they have redeeming features in varying degrees. I worry that some will refuse to concede that, say, the hierarchical relation between parents and children has any wrong-making features at all. But even if one or the other (or both) of these lines of argument proves successful, the fact remains—and this is the main claim of the present section—that moral equality is consistent with social hierarchy.
4. Justifying Social Equality in the Absence of Moral Equality I now turn to the question whether we could be morally permitted or even morally required to relate as social equals even if we are not moral equals (or even if we lack an adequate reason to believe we are morally equals). In answering this question, it is important to start by saying that there are quite different ways of denying that people have equal basic moral status.²⁴ I want to distinguish between two dimensions in which the denials can differ. The first contrast is between (a) denials of moral equality that correlate with, or are explained in terms of, the sorts of difference that Anderson refers to in her list of the hierarchical social relations that relational egalitarians oppose (racial, gender, caste, etc. differences), and (b) denials that do not involve the claim that socially salient groups differ in terms of basic moral status.²⁵ The racist views of Hastings Rashdall, to which Jeremy Waldron (2017: 31) devotes a chapter in his recent book on moral equality, fall into category (a). The view that Mother Theresa and Albert Einstein have higher basic moral status than Richard Arneson—a view which Arneson briefly entertains for purely illustrative purposes (I believe)—falls into category (b),²⁶ since the denial of basic moral equality in this case involves the claim that exceptional individuals scattered across all socially salient groups, for ²⁴ Elsewhere (Lippert-Rasmussen 2022(a)) I have argued that utilitarians contest this claim in a certain sense because they reject the notion that any individual—human or not—has a moral status in a nonepiphenomenal sense: in other words, they deny that an individual’s moral status is one factor among others that determine what she can do to others and what others can do to her, morally speaking. Utilitarians deny this because they think that all that matters is how our actions affect total wellbeing (that of all individuals). For present purposes, I can set aside the distinction between basic moral equal status in an epiphenomenal sense (on which we all have equal status because none of us has any) and a nonepiphenomenal sense (on which we all have the same status and our possession of that equal status is part of what determines what an individual can do to others and what others can do to her, morally speaking). ²⁵ Wolff (2019: 14–16) discusses a very similar distinction between hierarchies of individuals and hierarchies of groups. ²⁶ Admittedly, exceptional individuals do form a group—the group of exceptional individuals. However, while this group is not the sort of hierarchically superior group that relational egalitarians typically have in mind, there are ways in which intellectually superior individuals can be treated morally which
96
Should People Who Are Moral Equals Relate as Social Equals?
example, men and women, have higher moral status than others. Call the dimension in which we contrast denials of type (a) with denials of type (b) the (socially salient) group dimension.²⁷ The second dimension in which denials of basic equality can differ is in how great the differences in moral status between people are claimed to be. On (c) an elitist view they are huge. On a particularly objectionable elitist view, it counts negatively, morally speaking, that a certain action satisfies the interests of individuals with the lowest moral status. Perhaps some Nazis took this view of the moral significance of the interests of Jews. On a less but still hugely offensive elitist view, there are people whose interests just do not count morally, or count for very little—no more, perhaps, than the interests of animals did in the ethical outlooks of most people in the nineteenth century. We can distinguish elitist denials of equal basic moral status from (d) nonelitist ones. Non-elitists deny that all human beings have equal basic moral status, but they also deny some human beings have a much higher moral status than others. I will consider two non-elitist views shortly. I will call this second dimension the elitism dimension. I also want to use two cross-cutting terms. The first applies to group-based, elitist denials of basic equal moral status. I call these extremist denials of basic equality. The second applies to non-group-based, non-elitist denials of basic equality, which I call moderate denials of basic equality. My contention is that, typically, when philosophers ask themselves if relating as social equals can be justified if we are not moral equals, they have in mind extremist denials of basic equality. Accordingly, it is unsurprising that they take deniers of basic equality to support social hierarchy. After all, if, morally speaking, the interests of members of certain socially salient groups count negatively, their interests in being related to as social equals count against doing so, morally speaking. However, if the alternative to basic equality is moderate—some form of non-group-based, non-elitist denial of basic equality—it becomes unclear why we cannot not deny basic equality and yet affirm social equality (or something very close to it). The specific terms of the moderate denial of basic equality are crucial here. Below I introduce two such alternatives. I do so with the aim of defending the condition that if the alternative to moral equality is moderate, rejecting basic would involve valuing them as superior people in van Wietmarschen’s sense (see Lippert-Rasmussen 2022(b); Lippert-Rasmussen 2023). ²⁷ I am not suggesting that exceptional or simply superior individuals, for example, superior in terms of agential capacities or in terms of moral desert, could not form a socially salient group. However, as a matter of fact they do not, but if they did, affirming their moral superiority on grounds of this group membership would then turn to a type (a) denial of moral equality.
Kasper Lippert-Rasmussen
97
equality does not mean rejection of social equality as relational egalitarians describe it. For that purpose, I can remain neutral on whether we should prefer one of the alternative views to basic equality. The first alternative view is: The incommensurability view: For any pair of human beings, X and Y, it is: (1) false that X and Y have equal moral status; (2) and false that one of X and Y has higher moral status than the other (see Bejan in this volume: esp. 55–57).²⁸ To appreciate the motivation for this view, on which moral statuses among human beings are incommensurable, consider two masterpieces scoring equally on all relevant evaluative dimensions. It may seem that the two paintings have the same aesthetic value. Suppose, however, that we improve the quality of one of them marginally in one dimension. Plausibly, that would not make this masterpiece better than the other. This, so the reasoning goes, shows that it is wrong to say that they are exact equals in their aesthetic value, for if they were, then, if one were to become slightly better than it is presently, it would then be better than the other. In this sense, the aesthetic values, or statuses, of the two paintings are incommensurable. We might say something similar about the moral status of human beings. In fact, this is not just something we might say—it is something people do say, or at any rate hint at, when they submit, as they often do, that human worth, or dignity, is in some sense enjoyed in full and always by all. Consider two people who score equally on all those parameters which ground moral status (or the range property which grounds moral status). Plausibly, their moral status must be the same. Suppose then that we boost the score of one of these individuals marginally in one of the parameters. Intuitively, this would not make the moral status of this individual a little higher than that of the other—‘Why should we flip a coin biased slightly in your favour (rather than an unbiased one) because your agential capacities are slightly greater than mine to decide who gets the last loaf of broad on our rescue boat? I too have kids, a life to live, and a subjective perspective of my own’.²⁹ As with the aesthetic value of paintings, this may be taken to suggest that the moral statuses of different individuals are incommensurable. Would denial of basic equality on grounds of the incommensurability of moral status oblige us to affirm social hierarchy? I do not think so. For one thing, the incommensurability view certainly justifies none of the social hierarches Anderson describes as those relational egalitarians oppose, that is, ²⁸ Cf. Raz (1986: 342). ²⁹ This kind of reasoning is part of the motivation for Rawls’s range property view.
98
Should People Who Are Moral Equals Relate as Social Equals?
hierarchies based on ‘inegalitarian ideologies of racism, sexism, nationalism, caste, class, and eugenics’. On the incommensurability view, all human beings, whatever their gender or race, possess incommensurable moral status—a status enjoyed, as we might say, in full and always by all. A second consideration is this. One of Anderson’s central positive characterizations of social equality describes a society in which each accepts ‘the obligation to justify their actions by principles acceptable to the other’ and in which all ‘take mutual consultation, reciprocation, and recognition for granted’ (Anderson 1999: 313–314). Since no incommensurable is morally superior to another, and since the incommensurability view is incompatible with a no-need-forjustification-to-those-below hierarchy of moral statuses, I see no reason why ‘the obligation to justify [one’s] actions by principles acceptable to the other’ might not be consistent with, and perhaps even derive justificatory support from, the incommensurability view. Consider next: The sufficientarian view: For any pair of human beings, X and Y, it is true that whether or not one of them has a higher moral standing than the other, both have sufficient moral status. An individualʼs ʻpossession of sufficient moral standingʼ means that there is, in some sense, a sufficient number of sufficiently important things that one cannot do, morally speaking, to that individual and which this individual is permitted to do.30
Variants of the sufficientarian view of moral status include strong versions in which there is a quite extensive set of significant things one cannot do to an individual with the moral status of a sufficient, even if, when all individuals’ interests are satisfied to a very high degree (etc.) it is morally permissible to give less moral weight to that individual’s interests (etc.) than one gives to the interests of individuals with higher moral status.³¹ For a wide range of sufficientarian views, the stronger the sufficientarian view is, the closer it approximates to the moral equality view. Would acceptance of the sufficientarian view of moral status oblige us to affirm social hierarchy? As with the incommensurability view, the answer is no.³² My reasoning here more or less repeats that just set out in ³⁰ One complicating factor, which I ignore here, is the possibility of trade-offs between the number of forbidden acts and the significance of these acts. ³¹ I need ‘etc.’ because sufficientarians about moral status might not think that interests are all that matters to the moral status of an individual (for example, her will might matter too). ³² As was also the case with the incommensurability view, in suggesting this I am not seeking to promote sufficientarianism. Indeed elsewhere I have argued against a sufficientarian view of relational justice, and
Kasper Lippert-Rasmussen
99
connection with the incommensurability view. Like the incommensurability view, strong sufficientarian views cannot be combined with (at least) the standard hierarchical moral status views often mentioned as alternatives to moral equality. They cannot be so combined, because they deny that higher moral status is correlated with membership of any sort of socially salient group. Specifically, they are inconsistent with the sort of social hierarchies deemed unobjectionable by Anderson’s inegalitarians, for these imply that the members of certain socially salient groups, such as women, or people of a certain race, have a lower moral status than that of sufficiency. Similarly, in virtue of its anti-elitism (especially if the bar of sufficiency is set high), a strong sufficientarian view will be compatible with Anderson’s characterization of a society of social equals. It is possible that one of the things we are not morally permitted to do to a sufficient (or indeed to someone with higher moral status than that of a sufficient) is reject ‘the obligation to justify [our] actions by principles acceptable to’ them, or dismiss the demand for ‘mutual consultation, reciprocation, and recognition’ (Anderson 1999: 313–314). Perhaps sufficients have the right to be given sufficient justification of what is done to them by others which appeals to principles that are acceptable to them. In that case, the sufficientarian view potentially justifies the requirements of Anderson’s democratic equality. I conclude that moderate denials of basic equality—for example, denials based on the incommensurability view and the sufficientarian view—do not oblige us give up on justifying social equality. Moral equality may be something we wish to commit to for other reasons, but we do not need it to defeat justifications of the sort of social hierarchies to which Anderson’s relational egalitarian objects. Hence, what follows normatively from rejecting moral equality is less than is normally assumed. This has been missed because denials of moral equality tend to be considered either with no explicit alternative in mind, or against the background of an extremist form of denial (which makes the equality look very attractive).
some of my arguments there apply mutatis mutandis to a sufficientarian view of moral status, too (LippertRasmussen 2020). Nor am I suggesting that Anderson will have no objections to the sufficientarian view. I am merely contending that it resonates with her most salient characterizations of a society of equals. I would add though that I suspect that whether social hierarchies that a demanding sufficientarian view might justify between above and below threshold individuals (and between different above-threshold individuals scoring differently on the relevant moral status grounding dimensions) are unjustifiable is something that people will have less clear intuitions about than about the sort of hierarchies that Anderson addresses.
100
Should People Who Are Moral Equals Relate as Social Equals?
Of course, the second of these intellectual habits is not terribly surprising. Historically speaking, the alternative to moral equality has very often been an extreme denial of moral equality. However, that does not undermine my present critique of the idea of justificatory relevance (defined in Section 1). Note that I have not argued, nor need I do so for the purpose of the present critique, that people are not equals, or that people are either incommensurables or sufficients. The view I have defended is that even if we are not moral equals, it need not follow that we cannot justify social equality. To make the case for this conditional I do not need to affirm that we are incommensurables or sufficients.
5. Conclusion My main aim in this chapter has been to explore the relationship between the basic moral equality that most moral theorists subscribe to and social equality of the sort advocated by relational egalitarians. Some leading defenders of relational egalitarianism seem to think that the two are tightly connected. For two reasons, I disagree. First, it is unclear that basic moral equality justifies social equality. Perhaps this is not novel. After all, a number of theorists have argued that people are moral equals and yet embraced hierarchical relations of various sorts, under some circumstances at least (Hare’s view of slavery in normal and then remote counterfactual circumstances is a case in point). Second, social equality can be robustly justified even if we reject the idea of moral equality. While this view might be at odds with the way some relational egalitarians have thought of their own view, they should welcome it. It untethers our commitment to social equality from the outcome of the philosophical debate about basic moral equality, and at present it is far from clear that there is any real prospect of a knock-down argument for basic moral equality being provided by those who subscribe to it.³³
³³ I thank Giacomo Floris, Nikolas Kirby, and an anonymous Oxford University Press reviewer for very constructive and insightful written comments of an earlier version of this chapter. Also, I thank Andreas Albertsen, Simone Degn, Søren Flinch Midtgaard, Ida Nørregaard, Tore Vincent Olsen, Astrid Oredsson, Viki Pedersen, and Jens Tyssedal for discussion of the chapter. Finally, I am grateful to the Danish National Research Foundation for financial support in relation to work on this chapter (DNRF144).
Kasper Lippert-Rasmussen
101
References Anderson, E. 1999. ‘What is the Point of Equality?’ Ethics, 109/2: 287–337. Anderson, Elizabeth, and Pildes, Richard H. 2000. ‘Expressive Theories of Law’. University of Pennsylvania Law Review, 148: 1503–1575. Arneson, R. J. 1999. ‘What, If Anything, Renders All Humans Morally Equal?’. In Peter Singer and His Critics, edited by Dale Jamieson, 103–128. Oxford: Blackwell. Arneson, R. J. 2015. ‘Basic Equality: Neither Acceptable Nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Bengtson, A., and Lippert-Rasmussen, K. 2023. ‘Relational Egalitarianism and Moral Unequals’. Journal of Political Philosophy, online first: 1–24. https://onlinelibrary. wiley.com/doi/pdfdirect/10.1111/jopp.12299 Hare, Richard. 1979. ‘What Is Wrong with Slavery?’. Philosophy & Public Affairs, 8/2: 103–121. Husi, S. 2017. ‘Why We (Almost Certainly) Are Not Moral Equals’. Journal of Ethics, 21: 375–401. Kirby, N. 2018. ‘Two Concepts of Basic Equality’. Res Publica, 24: 297–318. Lippert-Rasmussen, Kasper. 2020. ‘Relational Sufficientarianism and Frankfurt’s Objections to Equality’. Journal of Ethics, 25/1: 81–106. Lippert-Rasmussen, Kasper. 2022(a). ‘What Is It For Us To Be Moral Equals? And Does It Matter Much If We’re Not?’ The Journal of Contemporary Legal Issues, 23/2: 307–330. Lippert-Rasmussen, Kasper. 2022(b). ‘Discrimination and Moral Equality of Persons: Why We Can Oppose the Former for Egalitarian Reasons and Nevertheless Reject the Latter’. Ethical Theory and Moral Practice. https://doi.org/10.1007/ s10677-022-10343-3. Lippert-Rasmussen, Kasper. 2023. ‘Is Discrimination Wrong Because It Is Undeserved?’. Inquiry, online first: 1–28. https://www.tandfonline.com/doi/abs/10. 1080/0020174X.2023.2186947 McMahan, J. 2002. The Ethics of Killing. Oxford: Oxford University Press. Miller, David. 1998. ‘Equality and Justice’. In Ideals of Equality, edited by A. Mason, 21–36. Oxford: Blackwell. Parr, T., and Slavny, A. 2019. ‘Rescuing Basic Equality’. Pacific Philosophical Quarterly, 100/3: 837–857. Rawls, J. 1971. A Theory of Justice. Oxford: Oxford University Press. Raz, Joseph. 1986. The Morality of Freedom. Oxford: Clarendon Press. Scheffler, S. 2003. ‘What Is Egalitarianism?’ Philosophy & Public Affairs, 31/1: 5–39. Singer, P. 1990. Animal Liberation, 2nd ed. New York: Random House.
102
Should People Who Are Moral Equals Relate as Social Equals?
Singer, P. 2011. Practical Ethics, 3rd ed. Cambridge: Cambridge University Press. Waldron, J. 2017. One Another’s Equals. Cambridge, MA: Harvard University Press. van Wietmarschen H. 2021. ‘What Is Social Hierarchy?’ Noûs online first: 1–20. https://doi.org/10.1111/nous.12387. Wolff, J. 2019. ‘Equality and Hierarchy’. Proceedings of the Aristotelian Society, 119/1: 1–23.
4 Is There a Thing Called Moral Equality? (And Does It Matter if There Isnʼt?) Andrea Sangiovanni
In Philosophical Explanations, Nozick writes: ‘When there seems to be no conceivable alternative, yet the one view lends us to a philosophical impasse, suspect entrapment by a particular picture or model of how things must be’.¹ And so it is, I shall argue, with a common picture or model of moral equality. To make things clearer, I will recommend abandoning all talk of moral equality. Nothing will be lost, and quite a few things gained, if we do so. In Section 1, I contrast non-consequentialist and consequentialist interpretations of moral equality, and explain how (a) they diverge so significantly as to call into question whether they share a commitment to the same thing, and (b) the reference to equality in each interpretation is redundant. In Section 2, I suggest that we ought to distinguish more clearly two different roles that appeal to moral equality has played in social and political debate: its role in animal ethics, and its role in assessing the permissibility of hierarchies in social status. In Section 3, I explore in more depth how arguments in animal ethics regarding moral considerability and moral status can come apart from arguments regarding social equality, and trace some implications for relational egalitarianism. Section 4 concludes by recommending that we abandon talk of moral equality.
1. Do Consequentialism and Non-Consequentialism Accept the Same Principle of Moral Equality? Everybody seems to know what they are talking about when they talk about moral equality. And, as many philosophers have said, everyone seems to converge in affirming it.² It is meant to be a benchmark commitment of ¹ Nozick (1983: 99). ² See, for example, Dworkin (1977: 272–273); Kymlicka (2002: 4). Andrea Sangiovanni, Is There a Thing Called Moral Equality? (And Does It Matter if There Isn’t?). In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0005
104
Is There a Thing Called Moral Equality?
liberalism. Here is a somewhat abstract formulation: there is something that all are equal in the possession of that generates significant and basic moral duties owed to beings that possess it—duties of equal respect (mostly in non-consequentialist accounts) or equal consideration of welfare (mostly in consequentialist ones). But what is this ‘something’? Who is within the scope of the ‘all’? What is the content of the duties that flow from that ‘something’ (whatever ‘it’ is)? And in what sense are the duties ‘equal’? I will argue that once we give a value to each of these variables, not only does the convergence seem to fade, but so does the sense that we are really even speaking about the same thing.³ And, even worse, the egalitarian character of the commitment seems to fade, too. We aren’t left with much. I will conclude that we need to think about the phenomena classed under the label ‘moral equality’ in a different way than we have been wont to do. Suppose we specify ‘rights’ as the object of equality, such that anyone who affirms moral equality affirms that all possess equal basic rights, where by ‘basic’ we mean that they are rights that are not derived from other rights or duties, and that give rise to duties but are not themselves grounded in some pre-existing duty or set of duties. There are a number of worries here. First, note that this immediately excludes any sort of welfare consequentialist who doesn’t believe that rights are basic in the required sense. For the welfare consequentialist, if we should treat everyone as possessing the same set of (moral, legal, or conventional) rights, then this is because assigning this set to everyone is most likely to make the outcomes better in terms of welfare. The significance of rights is derivative. In what sense, then, do the welfare consequentialist and the rights-based non-consequentialist converge on a single commitment to moral equality? It seems more appropriate to say that they hold very different basic moral commitments, rather than to say that they hold a single commitment to moral equality which is just interpreted in different ways. This is even more evident when there is also disagreement about who is within the scope of the ‘all’. Suppose, as is common, that the rights-based non-consequentialist believes that all and only persons can have basic rights, while the welfarist consequentialist believes that all and only the interests of sentient beings matter morally in deciding which outcome is best. What is then the point of saying they converge in something as abstract as ‘for all x and y, x is equal to y when and because they are equal in the possession
³ Cf. Kirby (2018), who argues that there are two concepts of basic equality, one of which applies to value, and the other to authority.
Andrea Sangiovanni
105
of some duty-grounding z’ when they differ so fundamentally both on the values to assign to z and on the domain of x and y? Second, to say that all are equal in rights is equivalent to saying that they have the same rights. But within the scope of the ‘all’ we ought, surely, to include all women and men. But now look at the rights in the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW), which includes the right to special protection during pregnancy in types of work proved to be harmful to women.⁴ This is not a right possessed by men. An obvious fix is to point out that the rights in the CEDAW are not basic; rather, they are rights derived from more basic rights—possessed by both men and women—to bodily integrity or perhaps health. It seems clear that what is driving this move is a prior commitment to equality among all persons: we move upwards in the scale of abstraction until we have rights that all persons can be said to have to the same extent. If we don’t, we move higher still. But then we wonder: what is the more fundamental, morally relevant property, shared by all persons, in virtue of which all persons have the same rights? Rights-based non-consequentialists usually point to the capacity for normative agency or moral personality, each of which are, in turn, grounded in the presence of further psychological capacities (for example, capacities for thought, reflection, choice, and principle-guided action).⁵ But here two problems emerge. What is it about the capacity for normative agency or moral personality that makes it a ground for basic rights and their corresponding direct duties, such that all who have the capacity have the same basic rights? Suppose, as is common, that one responds that normative agency or moral personality gives its possessors a worth or value that demands respect—demands that its possessors be treated not only as counting morally in their own right and for their own sake, but also as counting equally (that is, every such demand, where circumstances are similar, should be treated as creating equally stringent and weighty claims as every other such demand). The first problem is familiar: if the demand to be treated with equal concern and respect is grounded in a worth or value possessed in virtue of the capacity for moral personality/normative agency, then, if the capacity varies, the worth should, too. And if the worth varies, then the demands must not, then, be equally stringent and weighty: those with higher capacity have higher worth; their claims should therefore be treated as more weighty and stringent than those of others with lesser capacities. ⁴ See https://www.ohchr.org/en/instruments-mechanisms/instruments/convention-elimination-allforms-discrimination-against-women. ⁵ See, for example, Griffin (2008); Rawls (1999).
106
Is There a Thing Called Moral Equality?
The usual response is either to argue that moral personality is a range property, and so does not vary within the range, or to look for some other morally relevant property, possessed by all persons, that doesn’t vary. But then one wonders: why shouldn’t we take the capacity for moral personality/normative agency to vary along with the psychological properties on which it, in turn, is grounded? And why a threshold above which moral personality/normative agency doesn’t vary and below which it does? I don’t want to pursue this particular dialectic any further, because it has been pursued already many times. Suppose the back and forth ends up in a stalemate, or, for the sake of argument, suppose that the defender of moral personality/normative agency wins. The second problem (much less familiar than the first) is this. The defender of the moral personality/normative agency argument concedes that were there to be no independent reason to believe that moral personality/normative agency is a morally salient range (or binary) property, then we would be moral unequals. The commitment to the claim that all persons have equal rights in virtue of their personhood therefore presupposes, and follows from, a higher-level commitment to the principle ‘respect beings in accordance with the worth of their capacity for moral personality/normative agency’. Note that this higher-level principle is not egalitarian: it doesn’t demand or require equality in any form. All else equal, there would be no reason, for example, to believe that a world in which persons turned out to have such capacities to different degrees, and hence to possess different rights, was any more unjust or unwelcome than a world in which they didn’t. There would be no pressure, in this case, to equalize capacities. The equality in the equal rights claim is therefore entirely redundant.⁶ The trouble is that the advocate of the moral personality argument does seem committed to the egalitarianism implicit in assertions of moral equality. What drives the search, after all, for a morally relevant, natural property that does not vary, and that can ground a claim to equal treatment, is our commitment to equality rather than a commitment to respecting worth wherever it lies. As in the case of women and men, were the properties discovered to vary, we would keep looking until we found ones that didn’t. If this is right, then we have a prior and independent commitment to equality that is not explained by the usual arguments for moral equality. One might object that I have mischaracterized arguments for moral equality.⁷ Such arguments seek to explain moral equality in terms of the ⁶ Raz (1986: Ch. 9) and Westen (1982) make a similar argument. ⁷ Thanks to Nik Kirby for the objection.
Andrea Sangiovanni
107
psychological properties on which it is grounded. They do not seek, that is, to justify a commitment to moral equality, and they take no view on whether the particular distribution in status that results from an application of the grounding principles is a good or bad thing. What matters is that moral status should track some status-grounding, normatively relevant property, and the objector is neutral on how that particular inquiry should turn out. If it turns out that there is no morally relevant range or binary property on which moral status might be grounded, then the objector would be happy to welcome inequalities in status among people with (presumably) different psychological capacities. But should we be? So far in the debate, there has been much more scepticism regarding the existence of relevant grounding binary/range properties than there has been confidence that one has been, or will be, ‘found’. And yet most really do want to resist the conclusion that would follow. This is evidence that, in the wider reflective equilibrium that is required to establish the truth of any particular grounding claim, people are more strongly committed to equality than they are to the grounding claims. One rightly senses that accepting inequality in status among human beings (I discuss animals below) leads to some dark paths, and one resists the conclusion, and continues the search for something that might justify the commitment to equality; one then tends to despair or go quiet or change the subject when the search runs dry. As I will suggest below, maybe the search for a grounding property is a fool’s errand, and we would do better to try to characterize and explain our prior, stronger, more basic commitment to equality in other ways.⁸ Similar arguments apply if we take a different starting point. Suppose we are welfare consequentialists of a standard kind. We believe that we ought to act so as to bring about the outcomes that are best, where what is best is entirely a function of the wellbeing of sentient creatures. We are committed to basic equality because we believe that the comparable interests of all sentient creatures ought to have the same weight in our deliberations. Any one (morally considerable) being’s interests ought to count for as much as any other’s comparable interests. If one could alleviate a similar sized headache of a mouse or of a human being but not both (and there are no other goods whose pursuit is inhibited as a result of the headache, and both the mouse and the human being will forget the headache after it is gone), one ought to flip a coin (or equivalent). Each is to count for one and no more than one. Once again: it is very unclear how the welfarist consequentialist has anything much in common with the equal-rights non-consequentialist. What is ⁸ For a similar complaint, see also Phillips (2021); Sangiovanni (2017).
108
Is There a Thing Called Moral Equality?
the point of saying that they converge on something called moral equality, or have different interpretations of a determinate idea, given the extent to which they disagree? With respect to the variation objection, I believe that welfare consequentialists do better (though this is controversial). The reason usually given for why the interests of sentient beings matter in their own right and for their own sake is that only beings of that kind have a welfare in the first place.⁹ While, of course, plants can also flourish, wither, and die, and so things can go better or worse for them, there is nothing that it is like to be a plant, no point of view from which the plant’s flourishing matters to it. For sentient beings, on the other hand, there is.¹⁰ There are two reasons why this kind of view doesn’t fall prey to the variation objection. First, the significance of sentience does not travel via an account of a being’s worth. It is not as if more sentience grounds more significance. Second, the significance of sentience is limited to establishing the idea that sentient beings’ interests matter in their own right and for their own sake. How much or in what further ways they matter is left open, and requires further moral discussion.¹¹ For example, it may be that (as most welfare consequentialists believe), it is morally worse, on average, to kill a person than a mouse. This is not because persons have greater worth or value or sentience than mice, but merely because persons stand to lose many more goods in dying than a mouse. They have then a greater interest in continuing to live than mice, and it is for this reason that killing the person is worse. The commitment to the moral importance of sentience also leaves open whether other, interest-independent factors might make a difference to what we ought to do all things considered. For example, if one is a prioritarian, then the fact that one being is at a lower level of wellbeing than another gives one reasons to give additional weight to the interests of that being in determining what to do. This is, again, not because those who are at a lower level of wellbeing are worth more, or because they have, say, a greater degree of sentience. The reason is that there is, or so the prioritarian argues, more (impersonal) moral value in aiding those who are worse off, all else equal.
⁹ See, for example, Singer (1999: 328–332). ¹⁰ It is not relevant, for our purposes, whether sentience requires consciousness, or whether sentience is sufficient for what I will call basic moral status. Some will want to say that a being that merely has phenomenal experience of the world but does not have desires or intentions, and is not therefore capable of agency, does not have enough for moral standing. Others will want to say that mere agency (even without phenomenal experience) is sufficient. I leave these possibilities aside. In our world, sentience and agency, for all we know, go together. See Kagan (2019: 23ff ) for further discussion. ¹¹ This is why it is also possible to fit the claim that only the interests of beings with a subjective point of view matter morally into non-consequentialist accounts. See, for example, Regan (2004 [1985]); Sangiovanni (2017: Ch. 1); Sher (2014: Ch. 1); Williams (2005). On this reading, the argument for what we might call basic moral status is not yet an argument for moral equality.
Andrea Sangiovanni
109
However, this open-endedness comes at a price. As in the case of the equalrights non-consequentialist, we wonder: in what sense is the commitment to treating all comparable interests equally really an egalitarian commitment? Is the reference to equality redundant (just as it was for the equal-rights nonconsequentialist)? Note that there is no pressure in the commitment to make or keep interests equal. The interests come as they are. What really matters, for the welfare consequentialist, is whether they are comparable. Whether they are equal is neither here nor there; saying we ought to treat comparable interests equally is just to say that we ought to treat interests that are in no morally relevant way different the same. That is hardly a substantive commitment. There is, therefore, nothing lost in dropping any reference to equality, let alone moral equality for the welfare consequentialist, too.¹² I have limited myself to two very broad-brush sketches of consequentialist and non-consequentialist approaches to moral equality. One might wonder whether the account generalizes to other actual and possible accounts. I believe it does. In the non-consequentialist case, the problem is what we might call the grounding strategy—the attempt to find some morally relevant binary/range property, itself grounded in some set of psychological capacities, that explains how and why we are moral equals. I don’t see how this kind of strategy can yield a substantively egalitarian conclusion, given that it rests, ultimately, on the thought that we are equals because we are equal in the possession of the grounding properties, which implies that if we are not equal in those properties, then we are not equal in moral status. In the consequentialist case, the problem arises when we realize that welfare consequentialists do not value the interests of every being equally tout court. They only value the comparable interests of every sentient being equally. But saying that we ought to treat every sentient being’s comparable interests as mattering equally is no different to saying that we ought to treat sentient beings’ interests the same except when there is a morally relevant difference between them. Everything now rides on what makes interests comparable. This kind of strategy then leaves entirely open whether and when we ought to treat others as equals in some more substantive sense, and in what way. As we will see in the next section, the conclusion we should draw is not that we are, in fact, moral unequals; rather, we need to reject both interpretations of moral equality and think of the underlying commitment to something called ‘moral equality’ in different terms. ¹² The only exception would be an egalitarian welfare consequentialist, who believes that the outcomes are best when wellbeing is equal. But this kind of egalitarianism is not basic but distributive. The egalitarian welfare consequentialist also has the formidable task of explaining why this pattern in welfare is preferable to others. See Temkin (2003).
110
Is There a Thing Called Moral Equality?
2. Abandoning Moral Equality So far I have argued that it is very unclear what, in the usual formulations, a commitment to moral equality is a commitment to. On either an equalrights non-consequentialist formulation or a welfarist consequentialist one, the appeal to equality looks redundant. And, even if it isn’t redundant, it doesn’t look like a commitment to the same idea given how differently each class of views understands the scope, content, and meaning of its different components. Is it sensible, then, to abandon the enterprise of identifying a theory of moral equality, and just to carry on in normative ethics and political philosophy without it?¹³ In short, yes. Why might we be reticent to abandon inquiry into moral equality? We ought to answer this question by asking a different one: what roles has appeal to moral equality played in social and political discussion over the past few centuries? It seems clear that it has played at least two roles. First, it has played a role in the following major developments of the modern era: the democratic revolutions of the eighteenth century, the revolutions of 1848, the abolition of slavery, the patchwork expansion of the franchise across all liberal democracies, labour movements (especially those associated with socialism), the development of human rights in the wake of the Second World War, decolonization in the 1960s, Civil Rights and Women’s Liberation, the wave of democratizations in the 1980s and 1990s (including the end of Apartheid in South Africa), and recent social movements such as Occupy, #MeToo, and Black Lives Matter. In each of these cases, the focus of concern has been on the wrongfulness of treating as inferior. The role played by calls for equality has been, then, to undermine the structures that serve to enable the oppression of the ‘inferior’ by the ‘superior’—structures, that is, constituted by (objectionable) social status hierarchy. Paradigmatic instances of wrongful treatment that flow from, and are justified in terms of, social status hierarchy include caste societies; slavery; sexual harassment and assault; segregation and apartheid; political persecution and exclusion; invidious forms of discrimination; and demeaning forms of paternalism. The second role has become prominent much more recently. Since at least the 1970s (with, among others, the publication of Singer’s Animal Liberation 1995 [1975]), inquiry into moral equality has been associated with animal ethics.¹⁴ Singer (like Bentham before him) wondered: might the same arguments usually marshalled in favour of the conclusion that all comparable ¹³ For moral equality scepticism, see Husi (2017); Kekes (1988); Pojman (1992); Steinhoff (2014). ¹⁴ Singer (1989, especially Ch. 1). See also Singer (1989).
Andrea Sangiovanni
111
human interests matter equally be extended to (nonhuman) animals? Compare, for example, the interests of a human being with a headache with the comparable interests of a dog with a headache. And suppose we can alleviate the headache of one but not the other. And suppose further that the alleviation of the headache in both cases will improve the wellbeing of the dog and the human to the same extent (that is, the human being’s headache prevents the realization of the same amount of good as the dog’s). Why should we favour the interests of the human being in this case rather than, say, flip a coin? Singer anticipated that many will say: the human being has greater rationality, or something similar, that raises him up in the order of nature, and gives him a higher status than the dog. This is what justifies giving the interests of the human greater weight. Singer asked us then to imagine we offered the same argument on behalf of racism or sexism. And to the response that women or, say, black people don’t, on average, have lower rationality than men or white people, he challenged: and what if they did? Would that change our judgement? Singer answered: It is an implication of this principle of equality [the equal consideration of interests] that our concern for others ought not to depend on what they are like, or what abilities they possess—although precisely what this concern requires us to do may vary according to the characteristics of those affected by what we do. It is on this basis that the case against racism and the case against sexism must both ultimately rest; and it is in accordance with this principle that speciesism is also to be condemned. If possessing a higher degree of intelligence does not entitle one human to use another for his own ends, how can it entitle humans to exploit nonhumans?15
The only thing that should matter morally when comparing the interests of two different beings is the nature of the interest itself, rather than the nature of the being possessing it. This is why, according to Singer, racism and sexism are wrong: because both racism and sexism require us to take into account the race or sex of an individual rather than merely the strength and scope of their interests, they violate the principle that we ought to give all interests equal consideration. If Singer is right about our objections to racism and sexism, then there is no reason to treat the comparable interests of animals any differently to the comparable interests of black and white people (or men and women).¹⁶ But is it really true that what best explains our considered judgement regarding the wrongness of social status inequality between men and women, ¹⁵ Singer (1989: 151). ¹⁶ See also Singer (1995 [1975]: 36ff ).
112
Is There a Thing Called Moral Equality?
black and white people, is Singer’s version of the equal consideration of interests (which commits one to utilitarianism)?¹⁷ If it is, then one might think that any account of the wrongness of social status hierarchy and the wrongness of treating the comparable interests of animals and human beings differently must stand or fall together. But, I want to argue, we should reject the idea that what best explains our rejection of social status inequality between men and women, black and white people, is the principle of equal consideration of interests (as Singer construes it). This will clear up space for the idea that, in fact, arguments for and against various forms of moral status in animal ethics need only have a very loose connection to arguments for and against sexism, racism, and other objectionable kinds of social status hierarchy. The first step is to question whether the best arguments against racism (or sexism) must ultimately rest on the idea that comparable interests must receive equal consideration. The idea seems to do, on one interpretation, too little work and, on the other, too much. Compare two university policies. The first gives special consideration to applications from whites; the second gives special consideration to applications from blacks. And take two applicants who are identical in every way—test scores, socio-economic background, gender, and so on—except that one is black and one is white. What should the bare idea of equal consideration of comparable interests say in comparing, from a moral point of view, the two university policies? The idea does too little if it permits the following response. The interests of the two applicants are not comparable: given historical injustice, etc., there is a morally relevant difference between the weight we ought to give to the white applicant’s interests and the black student’s interests. The idea does too little for the reasons given in Section 1: on this interpretation, all the idea says is that we ought to treat the interests of two individuals the same unless there are good moral reasons to treat them differently. The idea is too weak to support our scepticism of racially based status hierarchy, since it leaves entirely open what kinds of differences between interests are ‘morally relevant’, and so leaves open exactly what our commitment to moral equality is supposed to aid us in explaining. The idea of equal consideration does too much if it commits one to the idea that the only thing that should enter our moral deliberations are interests, and the only thing that ought to determine the weight of an interest in our deliberations is its strength, duration, and scope. On this interpretation, the interests of the white and black student are equal in strength, duration, and scope (that is, we can easily assume that admission would contribute to ¹⁷ Singer (1993).
Andrea Sangiovanni
113
how well their lives end up going to the same degree). On this interpretation, we can only permissibly treat the white and black student differently by consulting the overall balance of interests affected by the policy across the rest of the society. If the policy of racial preference turns out to promote overall welfare, then we should adopt it; if it doesn’t, we should reject it. But this would be to purchase a substantive answer at the price of losing its ecumenical reach and plausibility: why should only interests count morally, and why must the weight of an interest turn only on their strength, duration, and scope? Only a utilitarian would agree (not even a prioritarian—who believes that how badly off someone is also matters [over and above the strength, duration, and scope of their interests] in moral judgement—could join forces here). The second step in the counter-argument is to suggest how arguments against status hierarchy and in animal ethics can easily come apart.¹⁸ Take an argument, like Shelly Kagan’s, for hierarchy in moral status between (most) human beings and (most) animals. If any argument within animal ethics is to have unpalatable implications for social status hierarchy, this would be one. For Kagan, we ought to treat beings with higher psychological capacities, beings who only have the potential for developing those capacities, and beings that do not, but could have had, such capacities as having a higher moral status, all else equal, than individuals lacking each of these items.¹⁹ This will be the case for most (though not all) human beings when compared with most (though not all) animals, and it will also be the case between some animals (say apes) and others (say fish). What Kagan means by ‘status’ is not what I meant by ‘social status’ above when discussing our objections to social inequality. Social status is a socially conferred set of advantages (such as influence, power, prestige, and authority) granted to individuals on the basis of some socially salient characteristic that marks them out as especially competent in some domain or otherwise socially valued.²⁰ An example might be the social status accorded to doctors as opposed to hairdressers. A status in Kagan’s sense is broader; it is a moral rather than social status. A being with a higher moral status will have a ‘normative profile’—a set of features governing how it ought to be treated—that is more encompassing, demanding, or comprehensive than another.²¹ An example: a being with a higher moral status might have, all else equal, entitlements to be protected or saved from harm that are more stringent or weighty than a being with a lower status. More generally, the interests of a being with higher moral status ¹⁸ Cf. Floris (2019). ¹⁹ Kagan (2019: Ch. 5). ²⁰ See, for example, Ridgeway (2019). ²¹ Kagan (2019: 8–9).
114
Is There a Thing Called Moral Equality?
can be given greater weight in our deliberations just in virtue of the fact that such beings have higher psychological capacities independently of how those capacities affect the strength, quality, psychological connectedness, and character of their interests. This is something that philosophers like Singer—and, for very different reasons, Tom Regan,²² Christine Korsgaard,²³ and James Rachels²⁴—deny. And it might seem like just the kind of argument that would fall prey to Singer’s analogy to racism and sexism. But does it? This is not the place to assess whether the argument succeeds on its own terms. Rather, I want to ask whether it has (a) the resources to resist Singer’s move from the rejection of social status hierarchy to the principle of equal consideration of interests, which would also imply a rejection of Kagan’s moral status hierarchy, and (b) implausible implications regarding objectionable instances of social status hierarchy. Does an argument for hierarchy in moral status, that is, necessarily translate into an argument for hierarchy in social status? And, conversely, do objections to social status hierarchy depend on arguments for equality of moral status? Or do the two address different concerns? Kagan could respond in the following way. First, he could point out that his argument has an entirely different structure to the racist and sexist arguments as Singer construes them. The racist and sexist say that membership in a social group gives one a higher moral status; Kagan says that it is an individual’s properties (including modal and potential properties) that matter. But, more importantly, even if it were true that, say, all men have higher psychological capacities when compared with (but comparable interests) all women, or all whites have higher capacities when compared with all blacks, Kagan could point out that the racist and sexist have very different ideas about the normative profiles that go along with having a lower moral status. Many (though, as we will see below, not all) racists and sexists believe that things like demeaning paternalism, slavery, segregation, public discrimination, punitive legislation, and so on are justifiable as a consequence of lower moral status. Kagan denies that any of those things must follow from possession of a lower moral status (indeed, he believes that, for example, the equivalent of slavery for animals with much lower psychological capacities is unjustified). It all depends on what the normative profile appropriate to an individual is, which, in turn, depends both on features of their psychological capacities more generally (including whether they are merely modal ²² Regan (2004 [1985]). ²³ Korsgaard (2018), who derives the conclusion from an argument about the incomparability of goodness-for across different species. See Ch. 4. ²⁴ Rachels (1990).
Andrea Sangiovanni
115
or potential capacities), but also on how those capacities bear on their interests. He denies, for example, that the differences between non-disabled adult human beings are significant enough to warrant radically different moral statuses. At most, he claims, those differences are tiny (compared with differences with animals), and, because of insurmountable epistemic hurdles in identifying them accurately, he also claims that we do better to disregard them when deciding what to do.²⁵ The conclusion we should draw is that even defences of hierarchy in moral status (in Kagan’s sense of a normative profile) need not have the implications we usually assume they do for objectionable social status hierarchies. Indeed, an account of why and when social status hierarchy is objectionable can be grounded in entirely different considerations. Social equality could be defended, inter alia, on the basis of independent harms that social inequality does, on the way it impinges on the autonomy of those with lower status, or on the non-instrumental value of social equality (the fact, say, that it realizes a form of unity and reciprocity among those involved that is unavailable in relations based on hierarchy). At no point need it rely on an underlying commitment to moral status equality in Kagan’s sense, where a moral status is a normative profile of things one can, cannot, and must do, with respect to a particular (morally considerable) being. Indeed, in the reverse direction, many widely dispersed and entrenched forms of social status hierarchy, including racially based social status hierarchies, don’t depend on beliefs in differential moral status. Racism and sexism need not require belief in the lower moral status of women and members of targeted minorities; as is well known, they can be based on (conscious or unconscious) prejudice, false belief about traits of character, dispositions, etc., or animus, and be structurally embedded in institutions; together, they can still serve to reproduce persistent and objectionable social status hierarchy. (This is undoubtedly true of most contemporary forms of sexism and racism.) If this is right, then it looks like an argument for equality in moral status may be too weak to find fault with current forms of racism and sexism without further argument. Hierarchies in moral status, therefore, need have no consequences for the legitimacy of social status hierarchy, and arguments for social equality need not be grounded in an account of equal moral status. To be sure, accounts like Kagan’s don’t sit well with our egalitarian intuitions. It doesn’t sound right to say that, for example, the claims of the more intelligent are more weighty in certain, even restricted, contexts (even by tiny amounts, and even if we can’t discern the differences reliably) than ²⁵ Kagan (2019: 288ff ).
116
Is There a Thing Called Moral Equality?
those of the less intelligent, merely because the latter lack intelligence and independently of any effect on their interests. (It is no surprise that Kagan desperately tries to counter this implication of his view.²⁶) This reticence is explained, in part, by the fact that were such differences in moral status, albeit small, to be realized in societies like ours, they would take on important symbolic and social meaning. It is very likely, given the importance of social status for us (which is based, recall, on assessments of presumed competence and value to the group) that differentials of Kaganesque moral status—based as they are on, among other things, intelligence—would generate differences in social status, with potentially important downstream effects for the structure of society. But this further consequence is contingent: it is possible to imagine a Kaganesque society that didn’t allow differences in moral status to translate into differences in social status. Furthermore, it is also available to someone like Kagan to say that, if recognizing differences in moral status did have objectionable downstream consequences for status hierarchy, then the harm done by the latter should outweigh the moral value of recognizing the former. (Note that, although he does not, this kind of response could be used to support Kagan’s claim that [minor] differences in moral status among nondisabled human beings should be disregarded.) And these possibilities are all that is required to secure the conclusion that inquiries into Kaganesque moral status can be divorced from considerations about what makes, and doesn’t make, social status hierarchy legitimate. None of this is to deny that social status hierarchies, especially the objectionable ones, can and have been based on arguments for differential moral status. But they need not be. This conclusion can be reinforced by noting the different sense of ‘equality’ at stake in assessing moral and social status hierarchies. Recall that, in Section 1, I argued that framing debates about moral equality in terms of equality of rights and interests was misleading. Claims about ‘equality’ in that context are redundant. Not so with respect to debates on equality in social status. Social status is essentially comparative in a way that moral status is not. Ascriptions of differential social status always imply a higher and a lower, a superior and an inferior. To subscribe to social egalitarianism is to subscribe to the view that there should not be superiors and inferiors. But ascriptions of differential (Kaganesque) moral status need not: someone may just have a different moral status—or normative profile—than another, without one being higher or lower than the other. When Kagan discusses ‘higher’ and ‘lower’ statuses, this is for ease of discussion, and is meant to track intuitive differences in the comprehensiveness and ²⁶ See, for example, Kagan (2019: Ch. 6).
Andrea Sangiovanni
117
demandingness of those normative profiles that track (on his view) psychological properties.²⁷ When Kagan describes two individuals as having equal moral status, then, all he means is that they have the same basic normative profile; the higher-level principle on which such judgements depend is something like ‘moral judgments should track differences not just in interests, but also in other interest-independent properties, such as intelligence’. It is not, absent further argument: ‘make differences in intelligence or welfare equal’. I return to this below. So far I have argued that moral status equality is not necessary for social status equality. But might it be sufficient? Might an argument for moral equality be enough to reject the bad forms of social status inequality but permit the good ones?²⁸ It is unclear how it could be. Why and how does an argument for, say, equal moral worth tell us which patterns of social status hierarchy are legitimate, and which ones aren’t? Pointing to worth-grounded claim-rights against the bad kinds of social status hierarchy, and no rights against the permissible kinds, risks begging the question. Why and how does ‘respect’ require one thing in one case, and not the other? Why and how does the equal possession of normative agency/moral personality distinguish between, for example, invidious and permissible discrimination? When does treating comparable interests equally prohibit social status inequality and when does it permit it? To answer these questions requires a richer account of the wrong involved in treating others as inferior. We require, that is, a further stretch of moral reflection. Is the wrongful kind of social status inequality wrongful because of its violation of the autonomy of those on the lower end of the hierarchy? Is it wrong because of the harm it does? Is it wrong because it is unfair in some way? Is it wrong when and because it makes outcomes worse (how?)? Establishing that two beings have the same set of rights—rights of equal stringency and weight—doesn’t tell us much about social status hierarchy until we know what those rights are. And establishing that we ought to treat the comparable interests of two beings the same doesn’t tell us much about social status hierarchy until we know when and why interests are comparable. This conclusion shouldn’t surprise us: if what I argued in Section 1 is correct, arguments regarding basic moral status are not egalitarian in any meaningful sense. One might object that arguments in favour of moral equality, when pursued in the context of social status hierarchy, are intended to counter familiar arguments for racism, sexism, feudalism, and so on that are based on claims ²⁷ Kagan (2019: 56). ²⁸ I thank Giacomo Floris for discussion on this point.
118
Is There a Thing Called Moral Equality?
regarding the moral inferiority of their targets. There are two problems with this kind of defence of a more traditional view of the relation between moral and social equality. First, arguing for moral equality would at most serve to undermine one ideological basis for sexism, racism, feudalism, and so on. But it would not explain what is wrong with the sexism, racism, feudalism, and so on, itself. Leaving aside the variation objection, there would still be a gap between the argument for moral equality (whatever it is) and the affirmation of social status equality. Second, and closely related, arguments in favour of moral equality would leave entirely unaddressed what is wrong with forms of racism, sexism, feudalism, and so on that are not based on beliefs in inferiority in moral status—where everyone engaged in reproducing these practices already affirms such equality. At this point in the argument, pointing to moral equality is like pointing to a black box. But how can one defend an account of social equality if not by reference to some prior, orthodox account of moral equality (for example, one that seeks to ground it in one or more morally salient properties possessed by all in equal measure). As we have already seen, however, social equality can be defended on other grounds. One could, for example, defend social equality on the basis of its non-instrumental value—on how, say, it realizes forms of unity, reciprocity, or solidarity. Or one could defend it on instrumental grounds—on how, say, social inequality harms those who are considered low status, violates their autonomy, or makes impossible cooperatively productive, non-coercive practices. The argument also has implications for the growing literature on relational egalitarianism.²⁹ Relational egalitarians believe that what matters, when assessing social justice, is not only the pattern in the distribution of goods or welfare, but also the quality of social relationships. And they believe that social relationships are just when citizens treat one another as equals. In turn, the demand to treat one another as equals, it is often argued, derives from a more fundamental commitment to moral equality.³⁰ If I am right, however, then defences of social equality need more than a reference to moral equality. Relational egalitarians need to explain, that is, what makes things like oppression, marginalization, social exclusion, and so on wrong (or bad) that neither begs the question (by building the wrongness into the definition of oppression, marginalization, etc.) nor merely adverts to the way each violates the respect owed to moral equals. The latter invocation, I have argued, is nothing but a restatement of the commitment to social equality ²⁹ See, for example, E. Anderson (1999); Miller (1997); Scheffler (2003); Schemmel (2021); Wolff (1998). ³⁰ See, for example, E. Anderson (1999: 313); Scheffler (2003: 21–24); Schemmel (2021: 55–56). But see Lippert-Rasmussen (2018: 61ff ).
Andrea Sangiovanni
119
(or a non-egalitarian step in an argument in animal ethics), not an explanation of what makes social equality valuable or required. This lacuna is especially clear when relational egalitarians explain when and why hierarchies of social status and esteem—not all of which are objectionable—violate the basic egalitarian commitment, and when they don’t.
3. Social Status Hierarchy and Animal Ethics I have argued that the underlying commitments that go under the name ‘moral equality’, such as it is, play two different roles that ought to be separated more clearly than they have been. In the first role, calls for moral equality have served to challenge deep and pervasive social status hierarchies (including slavery, caste systems, invidious discrimination, demeaning paternalism, and so on). In the second role, calls for moral equality have been used to extend moral concern to certain animals that meet various criteria (such as sentience). I have argued that addressing these separately is more fruitful: it is not sufficient to cite equal moral status in defence of equal social status, and it is not necessary either. And addressing the question of moral status—of what normative profile goes along with different interests, modalities, potentials, levels of wellbeing, types of sentience, and so on—isn’t best understood as a question about equality at all. Equality comes in, when it comes in, only non-comparatively, only when we want to say that two beings have the same normative profile—where a profile is just a list of the things one can, cannot, and must do, with respect to that being. Regarding this last point, even the disagreement between Singer and Kagan on the importance to give to interest-independent factors about an individual is not best understood, I believe, as an argument between a defender of moral equality (Singer) and a denier (Kagan). Rather, it is best understood as an argument about how different kinds of characteristics ought to matter in moral deliberation. Does it matter to an individual’s normative profile what capacities that individual—say, a child—could have had but now doesn’t (and will never) have? Singer: no. Kagan: yes.³¹ Does it matter, in compiling an individual’s normative profile, how badly off they are (independently of the strength, quality, and scope of their interests)? Singer: no. Prioritarian: ³¹ Others point to the importance not of individual modal capacities, but of the normative relevance of the characteristic life cycle and form of the species, such that individuals who lack the capacities that enable them to participate in this cycle and form, but that are members of the species, should still be treated as if they had them (or at least as if they had the potential to develop them). For the latter, see Kumar (2008); Nussbaum (2009); Thompson (2008). For criticism, see McMahan (2002: 146–154; 2008). For the thought that such beings are internally directed to the development of such species-typical capacities (even where they lack any, even genetic, potential to develop them), and this is why we ought to treat them as if they had them, see Lee and George (2008); Waldron (2017).
120
Is There a Thing Called Moral Equality?
yes. Assuming two individuals have exactly the same interests in receiving some treatment, or will forego the same extent of future good in being killed, should the fact that one has more sophisticated cognitive capacities matter in compiling its normative profile? Singer: no. Kagan (and many others³²): yes. Does agency matter independently of sentience in determining whether a being ought to be treated as mattering in its own right and for its own sake? Singer: no. Kagan (and some others³³): yes. All of these questions can be raised (and answered) without ever invoking equality, let alone something called ‘moral equality’, at all. Things are different when assessing social status hierarchies. First, recall that, unlike moral status, social status is, as I have already mentioned, essentially comparative: if someone has a high social status, someone else must have a low one. Social status differentials assign some people to a role as superior and others to a role as inferior. When social status is equal, this is because no party is deemed higher or lower in the esteem, privileges, authority, and influence characteristics of social status than any other. Some sociologists argue that the whole point of status is to pick out smaller groups of individuals, and to reward them for their competence, in the expectation that they will benefit the wider group.³⁴ Social status is also reflected, reinforced, and ossified by institutions. Differentials in social status based on race and gender, for example, are embedded, reflected, and reproduced in patterns of discrimination, segregation, and interaction across a society. And so are social status differentials based on wealth and physical appearance. Second, there is no social status hierarchy between (most) animals and humans. For there to be a social status hierarchy, there must be a common recognition of who is superior and who is inferior; social status is conferred not possessed.³⁵ In any human society, there is a remarkable awareness among its inhabitants who has higher and who has lower status. Even in situations where such assignments are contested, there is common knowledge about who is recognized as superior and who inferior and on what basis.³⁶ Because animals do not participate, except in special cases,³⁷ in the complex
³² See, for example, McMahan (2002). ³³ See, for example, Korsgaard (2018). ³⁴ C. Anderson and Willer (2014); Ridgeway (2019: Ch. 2). ³⁵ Ridgeway (2014); Ridgeway (2019: Ch. 4). ³⁶ C. Anderson et al. (2006). ³⁷ The special cases may include, for example, animals that understand social status hierarchy, and can participate in such relationships with human beings. Examples include various species of monkey and perhaps dogs as well. In these cases, the ways in which we treat such animals as inferior may well trigger the same kinds of moral concern as similar treatment among human beings. Whether it does or not will depend on what ultimately justifies our judgements regarding when and why social status hierarchy is wrong.
Andrea Sangiovanni
121
social interactions that determine and reproduce social status, they cannot be participants in social status hierarchy. To be sure, animals can be subject to human authority, influence, and so on, but, because such influence and authority is not conferred by both animals and humans on the basis of commonly known social norms determining presumed competence on the basis of typed characteristics (race, wealth, appearance, gender, and so on), it is not a form of social status. Someone assigned as a rabbit breeder, for example, is assigned that authoritative role by other human beings; the rabbits do not participate in the relations of social authorization and reproduction that are central to social status hierarchies, and so are not participants in them. Human slavery, by contrast, essentially requires, under extreme coercion, the active social participation of slaves in their own subjugation; a slave is aware of their place in the society (and aware of the owner’s knowledge of their knowledge), just as the owner is. This common knowledge is used by the owner to accomplish the social annihilation and stigmatization of the slave, not merely to mirror it.³⁸
4. Conclusion Is the argument presented above a form of scepticism about moral equality? Yes and no. No: I have not argued that we are moral unequals. I have, rather, argued that this is the wrong question to ask in the first place. Yes: I have suggested that we abandon talk of moral equality. Arguments commonly thought to be about moral equality elide important differences between moral and social status; we do better to distinguish more clearly between the two and to develop arguments about them without ever referring to something called ‘moral equality’. It is useful to summarize the arguments presented for these claims. Arguments regarding moral status or considerability are not usefully characterized as about equality. They are, rather, about the normative profile of different kinds of beings, including what makes a being morally considerable in the first place. They are arguments within, broadly, animal ethics. Philosophers, like Singer, who believe that only information about interests ought to matter in moral deliberation reject the idea that, even where interests are comparable, animals have different claims on us than other human beings in virtue of interest-independent features (such as intelligence). This does not, however, make them, I conclude, egalitarians in any relevant sense. ³⁸ Patterson (1982).
122
Is There a Thing Called Moral Equality?
Arguments for social status egalitarianism prescribe equality in ways that arguments about moral status do not. The social status egalitarian says that there should be no higher and lower rank between us when, say, voting, or between the sexes. Those who, like Singer, believe that there are no relevant differences between (comparable) interests or in dignity do not prescribe equality. For those like Singer, interests come as they are; our moral judgements should be responsive to relevant differences rather than seeking to eliminate them. Similarly, for those emphasize, say, the equal rightsgrounding worth or dignity of human beings, the argument takes the basis of worth, say, rationality, or intelligence, as it is; there is no prior prescription to eliminate differences in rationality or intelligence. And even those who are committed to hierarchy in moral status (like Kagan) need not be committed to forms of social status hierarchy. For these reasons, it is misleading to argue that equality in social status is grounded in equality of moral status. Doing so pre-empts substantive understanding of what makes social status hierarchies problematic (when they are problematic). Consider, for example, that not all forms of status hierarchy are objectionable in the same way as social status hierarchy between the sexes or among races. Social status awareness (and hence awareness of difference), as I mentioned above, is present in almost any human interaction, whether within a university, workplace, sporting team, or in a shop. Some might be tempted to say that the difference between objectionable and unobjectionable forms of status hierarchy is that the former are based on ‘arbitrary’ characteristics or on false beliefs (for example, about the distribution of intelligence among races or sexes). But what makes a characteristic ‘arbitrary’? And what if the beliefs weren’t false: would that make sexism or racism legitimate? Answering questions like this are beyond the scope of this chapter. It is enough if we have shown that they require much further argument than appeal to something called ‘moral equality’.
References Anderson, C., Srivastava, S., Beer, J., Spataro, S., and Chatman J. 2006. ‘Knowing Your Place: Self-Perceptions of Status in Face-to-Face Groups’. Journal of Personality and Social Psychology, 91: 1094–1110. Anderson, C., and Willer, R. 2014. ‘Do Status Hierarchies Benefit Groups? A Bounded Functionalist Account of Status’. In The Psychology of Social Status, edited by Joey Cheng, Jessica Tracy, and Cameron Anderson, 47–70. New York: Springer. Anderson, E. 1999. ‘What Is the Point of Equality?’ Ethics, 109: 287–337.
Andrea Sangiovanni
123
Dworkin, R. 1977. Taking Rights Seriously. Cambridge, MA: Harvard University Press. Floris, G. 2019. On the Basis of Moral Equality: A Rejection of the Relation-First Approach’. Ethical Theory and Moral Practice, 22: 237–250. Griffin, J. 2008. On Human Rights. Oxford: Oxford University Press. Husi, S. 2017. ‘Why We (Almost Certainly) Are Not Moral Equals’. Journal of Ethics, 21: 375–401. Kagan, S. 2019. How to Count Animals, More or Less. Oxford: Oxford University Press. Kekes, J. 1988. ‘Human Worth and Moral Merit’. Public Affairs Quarterly, 2: 53–68. Kirby, N. 2018. ‘Two Concepts of Basic Equality’. Res Publica, 24: 297–318. Korsgaard, C. 2018. Fellow Creatures: Our Obligations to the Other Animals. Oxford: Oxford University Press. Kumar, R. 2008. ‘Permissible Killing and the Irrelevance of Being Human’. Journal of Ethics, 12: 57–80. Kymlicka, W. 2002. Contemporary Political Philosophy. Oxford: Oxford University Press. Lee, P., and George, R. P. 2008. ‘The Nature and Basis of Human Dignity’. Ratio Juris, 21: 173–193. Lippert-Rasmussen, K. 2018. Relational Egalitarianism: Living as Equals. Cambridge, UK: Cambridge University Press. McMahan, J. 2002. The Ethics of Killing. Oxford: Oxford University Press. McMahan, J. 2008. ‘Challenges to Human Equality’. Journal of Ethics, 12: 81–104. Miller, D. 1997. ‘Equality and Justice’. Ratio, 10: 222–237. Nozick, R. 1983. Philosophical Explanations. Cambridge, UK: Harvard University Press. Nussbaum, M. 2009. Frontiers of Justice: Disability, Nationality, Species Membership. Cambridge, MA: Harvard University Press. Patterson, O. 1982. Slavery and Social Death. Cambridge, MA: Harvard University Press. Phillips, A. 2021. Unconditional Equals. Princeton: Princeton University Press. Pojman, L. 1992. ‘Are Human Rights Based on Equal Human Worth?’. Philosophy and Phenomenological Research, 52: 605–622. Rachels, J. 1990. Created from Animals: The Moral Implications of Darwinism. Oxford: Oxford University Press. Rawls, J. 1999. A Theory of Justice. Cambridge: Harvard University Press. Raz, J. 1986. The Morality of Freedom. Oxford: Clarendon Press. Regan, T. 2004 [1985]. The Case for Animal Rights. Berkeley: University of California Press.
124
Is There a Thing Called Moral Equality?
Ridgeway, C. 2014. ‘Why Status Matters for Inequality’. American Sociological Review, 79: 1–16. Ridgeway, C. 2019. Status: Why Is It Everywhere? Why Does It Matter? New York: Russell Sage Foundation. Sangiovanni, A. 2017. Humanity without Dignity: Moral Equality, Respect, and Human Rights. Cambridge: Harvard University Press. Scheffler, S. 2003. ‘What Is Egalitarianism?’. Philosophy & Public Affairs, 31: 5–39. Schemmel, C. 2021. Justice and Egalitarian Relations. Oxford: Oxford University Press. Sher, G. 2014. Equality for Inegalitarians. Cambridge, UK: Cambridge University Press. Singer, P. 1989. ‘A ll Animals Are Equal’. In Animal Rights and Human Obligations, edited by Tom Regan and Peter Singer, 148–162. New Jersey: Prentice-Hall. Singer, P. 1993. Practical Ethics, 2nd ed. Cambridge, UK: Cambridge University Press. Singer, P. 1995 [1975]. Animal Liberation. London: Penguin. Singer, P. 1999. ‘Response’. In Singer and His Critics, edited by Dale Jamieson, 269–336. London: Blackwell. Steinhoff, U. 2014. ‘Against Equal Respect and Concern, Equal Rights, and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 142–173. Oxford: Oxford University Press. Temkin, L. 2003. ‘Egalitarianism Defended’. Ethics, 113: 764–782. Thompson, M. 2008. Life and Action. Cambridge, MA: Harvard University Press. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: Harvard University Press. Westen, P. 1982. ‘The Empty Idea of Equality’. Harvard Law Review, 95: 537–596. Williams, B. 2005. ‘The Idea of Equality’. In In the Beginning Was the Deed: Realism and Moralism in Political Argument, edited G. Hawthorn, 97–115. Princeton: Princeton University Press. Wolff, J. 1998. ‘Fairness, Respect, and the Egalitarian Ethos’. Philosophy & Public Affairs, 27: 97–122.
PART II
JUSTIFICATION Grounding and Scepticism
5 Basic Equality and the Contexts of Opacity Respect Ian Carter
1. Introduction: Equality and Opacity In previous work, I have argued that our basic equality is grounded in a kind of respect for persons which I have called ‘opacity respect’ (Carter 2011; 2013; 2018; 2019). Basic equality is a normatively significant kind of descriptive equality of a given set of individuals. The basic equality of those individuals expresses the sense in which they are all equal. Assuming a set of individuals to be basically equal, it is appropriate to treat those individuals as equals, where ‘treating as equals’ is understood either as a kind of ‘distributive equality’—for example, giving people equal resources or opportunities or voting power—or as a kind of ‘relational equality’—for example, showing mutual respect or promoting equality of some kind of social status. I shall assume here that basic equality is a necessary part of the normative grounds for distributive or relational equality, consisting as it does in an equality of certain entitlement-grounding properties. As I see it, if the relevant individuals were not basically equal, then distributive or relational equality would be arbitrary or irrational. In contrast to those who have courageously claimed that we should treat people as equals even if there is nothing about them that is equal (Phillips 2021), I believe that a commitment to ‘treating as equals’ cannot itself be the bottom line in normative terms. But how is basic equality itself to be explained? This is a difficult problem, because human beings are generally so different from one another in both their achievements and their potential. Treat equal cases equally, and unequal cases unequally. Why then treat people equally? In addressing this problem, my starting point was John Rawls’s claim that what makes us equal, in such a way as to ground certain liberal egalitarian entitlements, is our equal possession of a range property: the fact of having a
Ian Carter, Basic Equality and the Contexts of Opacity Respect. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0006
128
Basic Equality and the Contexts of Opacity Respect
certain set of basic agential capacities within a certain range, from a minimum threshold upwards (Rawls 1971: sec. 77). In virtue of our equal possession of this range property, we enjoy an equal status as ‘moral persons’. For Rawls, the relevant agential capacities, which vary from person to person, consist in the ‘two moral powers’ of having a capacity for a conception of the good and having the capacity for a sense of justice. Alternative accounts could appeal to different, or more specific, agential capacities. The range property supervenes on the relevant capacities. A range property is binary—it is either possessed or not possessed—rather than existing in degrees. Therefore, the set of individuals who possess a particular range property are, in terms of possession of that property, equal to one another, even though the underlying capacities are themselves unequal. Rawls needs, but does not clearly supply, an adequate justification for focusing on the range property, which is binary, rather than focusing directly on the scalar, unequally possessed properties on which that range property supervenes.¹ If we focus on the binary property, we will consider people to be equal; if we focus on the more basic scalar properties, we will consider them to be unequal. Why not focus on the more basic scalar properties? Simply pointing to the range property seems to amount to little more than stating that we ought to treat people as basically equal even though they are really basically unequal. I suggested that what gives salience to the range property is a deeper sense of respect for agency. Where an individual is perceived as meeting a minimum threshold of agential capacities, there is a way of showing respect for that individual that involves adopting an external perspective in one’s interactions with them, a perspective that involves maintaining a distance and taking their capacities as given, rather than ‘looking inside’ them and acting on assessments of the very capacities on which their moral personality is taken to supervene. I call this attitude ‘opacity respect’. To show opacity respect is to engage in a kind of ‘evaluative abstinence’. I used this term not to indicate a refusal to evaluate per se, but, more specifically, to indicate a refusal to allow certain evaluative inputs into our practical reasoning about how to treat people. Our perception of another individual as meeting the minimum threshold is what makes this evaluative abstinence morally appropriate. From this particular kind of respect it follows that the particular levels of people’s agential capacities, above the threshold, are excluded from our
¹ Rawls does provide a justification, based on the contingency of the variable capacities, but most commentators have judged it to be inadequate. For a defence, see Andersson (2022). For a more general critique of contingency-based justifications, see Floris (2023).
Ian Carter
129
reasoning about how we ought to treat them. And from this it follows that people’s possession of the range property is morally salient. Because this account of basic equality proceeds from an unequally possessed property (agential capacities) to a relation (opacity respect) to another property that is equally possessed (the range property), it is not easily classifiable as adopting either a ‘property-first approach’ or a ‘relation-first approach’ (on this distinction, see Floris 2019b). Nevertheless, my approach to basic equality counts as a property-first approach inasmuch as it assumes that a morally relevant equally possessed property is required in order to normatively ground any kind of ‘relating as equals’, ‘not treating others as inferior’, and so on. Critics of my account of basic equality have presented a number of challenges.² Some dislike it because they find that its distributive implications do not chime with their favoured egalitarian prescriptions.³ Others have challenged its reliance on a particular threshold.⁴ Here, I will focus on a more direct challenge to the effect that opacity respect is not always an appropriate interpersonal moral attitude and that it therefore fails to explain the appropriateness of treating people as equals. I will defend my account by distinguishing between different ‘contexts of reasoning’ in which opacity respect might or might not be an appropriate moral attitude. Opacity respect is not an appropriate interpersonal attitude in all contexts of reasoning, but neither is basic equality a morally relevant feature of all such contexts. Moreover, those contexts in which opacity respect is appropriate tend also to be those in which our moral reasoning appeals to people’s basic equality. By making some relevant distinctions between the relevant contexts of practical reasoning, I hope to show how critiques of opacity respect as a moral attitude have been over-simplistic and have failed to invalidate its role in explaining basic equality. I will begin, in Section 2, by introducing a preliminary distinction between different ‘levels’ of basic equality. Next, in Section 3, I will summarize the general justification for opacity respect. In Section 4 I will introduce the idea of a relevant relation between individuals on the basis of which opacity respect might or might not be appropriate. Certain such relations can be said to constitute ‘the contexts of opacity respect’. In Sections 5, 6 and 7, I will examine
² See, in particular, Arneson (2015); Beillard (2013); Charvet (2013); Christiano (2015); Floris (2019a); Lippert Rasmussen (2016a; 2016b); Miklosi (2022); Sangiovanni (2017); Sher (2015); Steinhoff (2015); Wollner (2015); Waldron (2017); Williams (2020). ³ See, for example, Wollner (2015), and Arneson as discussed in footnote 8, below. ⁴ See Carter (2021) for some discussion.
130
Basic Equality and the Contexts of Opacity Respect
two ways of responding to the contrasting demands of evaluation and of evaluative abstinence in various contexts.
2. Levels of Basic Equality ‘Basic equality’ has often been assumed to mean equality of the most basic kind of status, sometimes called ‘equal moral status’ or ‘equal worth’. I do not conceive of basic equality in this narrow way. Basic equality does not necessarily mean ‘equality of basic moral status’ (although it can consist in an equality of basic moral status), for the way in which the adjective ‘basic’ qualifies the noun ‘equality’ is different from the way in which it qualifies the noun ‘status’ (Carter and Page 2023: sec. 5). There are many examples of equality of entitlement-grounding properties other than those that might ground equality of the most basic kind of status, and an equality of status can count as ‘basic’ equality even when the status is not itself basic. A status counts as basic if there is no other status among its grounding properties (Carter and Page 2023: sec. 2). Depending on the substantive normative theory adopted, an example of a non-basic status might be ‘citizenship’, whereas an example of basic status might be ‘moral personhood’. I take basic equality to be equality of any kind of entitlement-grounding property on the basis of which it is appropriate to confer a kind of status, where the resultant equality of two or more people in terms of that status does not itself derive from the mere fact of their occupying the same level in some broader status hierarchy (Carter and Page 2023: secs. 4, 5). On this view, the equal status of moral agents can certainly qualify as a kind of basic equality (in this case, basic equality of basic status), but so too can the equality of citizens that statist egalitarians assume to ground egalitarian entitlements only within a state, as long as ‘citizenship’ is understood as a binary property and the equality of citizens derives from some appropriate way of relating as contributors to the functioning of a particular society (Sangiovanni 2013: 61). And there are also yet more local kinds of basic equality, such as the equality of the members of a private club (who, for example, are entitled to equal voting power when choosing their president), or the equality of siblings (who, for example, might have equal inheritance rights, as is the case in some legal systems). The latter are all instances of basic equality (of non-basic status). In such instances, the equality in question is basic, not because the status, of which there is equality, is itself basic, but because the equality (of that status) does not itself consist in the mere fact of occupying the same level in a broader hierarchy of statuses. In other words, equality is basic if its value is not derived
Ian Carter
131
from the value of a more general kind of proportional justice relating people’s different statuses to their respective grounding properties. For example, the equality of two sergeants in a particular army—the equal standing on the basis of which they are allocated an equal set of rights and privileges within that army—is not a case of basic equality, because it derives from the positions those two individuals occupy within a broader status hierarchy. The equality of the two sergeants is therefore a kind of non-basic equality of entitlement-grounding properties. Not all of the various levels of basic equality need necessarily depend on opacity respect, but I think they are likely to do so wherever they depend on the moral salience of a range property supervening on the possession of certain kinds of agential capacities. Being a member of a private club might depend simply on being nominated as such. If being nominated is a binary property but not a range property, this might look like a case of equality that does not depend on the common possession of a range property. Yet even here, some further reason is needed in order to explain the equal voting power of all the members of the club when electing their president, as opposed to basing voting power on, say, the age or level of competence of each member. In cases such as this, my suggestion is that the equal status of the relevant set of individuals—in the present case, their status as voters— depends on the moral salience of a range property, and that the attitude of opacity respect explains that salience, together with the irrelevance of the underlying scalar differences in age or voter competence. The nature of the relevant capacities can vary, depending on the particular kind of status in question (Carter 2013). Although I cannot go into detail here, some examples may help to clarify this point. In the case of the equality of the citizens within a democratic society, we consider people as equally in possession of a range property that we might call ‘being a political agent’, which supervenes on the capacity for strategic political action, the capacity to assess government policies, and so on. If we assign citizens equal voting power, it is because we refuse to consider relevant, in this context, their particular levels of competence in such matters. This refusal is based on a kind of respect for the individual conceived as a political agent, a kind of respect that is opposed to assessing the very capacities that are supposed to constitute that individual so conceived. One point of such an assessment might be to aim at producing better political decisions on the part of the hapless demos. Thus, opacity respect at the political level rules out the various ‘epistocratic’ proposals that have recently emerged in democratic theory. Similarly, members of an academic college consider themselves equals: ‘collegiality’, in this sense, involves certain academics refusing, in certain contexts, to consider
132
Basic Equality and the Contexts of Opacity Respect
the levels of the intellectual capacities of fellow members of their particular academic community beyond a threshold necessary to become a member (in this case, normally a fairly high intellectual threshold). It is for this reason that many academics have seen the ‘research assessment exercises’ carried out by university funding authorities as undermining, if only indirectly, their ‘collegiality’. The fact that there are many levels of basic equality, each with a particular reference set of capacities upon which a particular range property supervenes, already goes some way towards explaining how we can consistently see opacity respect as being appropriate in some contexts but not in others.
3. The Substantive Grounds for Opacity Respect I will not try to defend the relatively uncontroversial claim that agential capacities call for the moral attitude of respect. I will simply assume that ‘valuable’ does not only mean ‘to be promoted’ (Scanlon, 1998: Ch. 2), that moral agents have value in the alternative sense of being due a kind of ‘recognition respect’ (Darwall 1977; 2006), and that this recognition respect, understood both as an appropriate attitude and as an appropriate form of behaviour, necessarily involves moral constraints on the ways we may view them and the ways we may promote their good (including through the promotion of the agential capacities that constitute them). If you are not moved by such deontological notions then you will not be interested in the arguments I am presenting here—unless, that is, you are also sceptical about the ideal of equality, in which case you might still understand those arguments as clarifying the source of an erroneous moral perspective. I will, however, try to say briefly why I think that recognition respect, when applied to persons, can plausibly be seen as including the requirement of opacity. In defence of the claim that recognition respect includes a requirement of opacity, and that ‘opacity respect’ is an essential part of ‘recognition respect for persons’, I have invoked the Strawsonian distinction between the ‘participatory’ perspective and the ‘objective’ or ‘scientific’ perspective (Carter 2011: 559; see also Bennett 2018). I believe this distinction helps to explain basic equality at various levels. The relevance of the distinction is at its clearest where we are theorizing about basic moral status, because at this basic level we consider persons in abstraction from their various social roles. But it is still relevant at ‘higher’ levels insofar as people’s social roles depend on their being particular kinds of agent—as, for example, in the case mentioned
Ian Carter
133
above, where the role of citizen depends on the individual qualifying as a ‘political agent’. According to Strawson, viewing someone from the participatory perspective means taking their agency as given and reacting in ways that demonstrate respect for that agency. It means feeling, and displaying, positive or negative reactive attitudes. To view someone from the objective or scientific perspective is instead to view them as a thing to be ‘managed or handled or cured or trained’ (Strawson 1974: 9). The two perspectives are in mutual tension, and I believe that adoption of the attitude of opacity respect should be seen as a part of the ‘art of separation’ that affirms the participatory perspective against encroachments by the scientific perspective that undermine it. It is no coincidence, I suggest, that the stand against such encroachments is a peculiar feature of modernity, just as the ideals of moral and political equality are peculiar features of modernity. As Christopher Bennett has written, the idea implicit in opacity respect is not to reject science and rationalism, with all the progress and opportunities they have brought us in the modern age, but to reject ‘a too-simple form of rationalism that applies the same model of knowledge to all areas of human and non-human life’ (Bennett 2018: 255). The scientific and participatory perspectives ought to be confined to their proper spheres of human life. More specifically, one can think of opacity respect as based on the recognition of an agent’s integrity, where ‘integrity’ is understood in the sense of wholeness and completeness of an agent.⁵ Agents exercise their rational agential capacities to form plans and commitments and act on them. They do so in ways that are more or less coherent, both synchronically and diachronically (Carter 2018). Thus, personal ‘integration’ is a matter of degree, as is the capacity to achieve it. As Christine Korsgaard says, through the coherence of such exercises of their rational agential capacities, persons literally ‘pull themselves together’, constituting themselves as distinct and more-orless integrated wholes (Korsgaard 2009).⁶ The respect involved in adopting the participatory perspective requires us to take this integration or ‘pulling together’ of the agent as given, as long as we have reasonable grounds for thinking that it surpasses a minimal threshold. Adopting this perspective involves treating the agent simply as a single whole. To quote Bennett again, it requires us ‘to take at face value the deliverances of the person’s thought and agency, and not to look at the “mess inside” from which it stemmed’ ⁵ On the assumption of a person’s wholeness or completeness, see Cupit (2000). On respecting a ‘whole’ person, though conceived as a set of desires rather than of agential capacities, see Enoch (2022: 159–161). ⁶ I prefer to speak of ‘integration’, which is certainly a scalar property, rather than of the more binarysounding ‘unity’, as Korsgaard does (Carter 2018: 833).
134
Basic Equality and the Contexts of Opacity Respect
(Bennett 2018: 265). The scientific perspective is in tension with this external perspective because it instead involves considering the agent as a bundle of measurable capacities, where the measurement of a capacity depends on counterfactuals and probabilities regarding specific kinds of reaction to specific kinds of stimuli. Thus, an agent is seen ‘as a causal system in which various inputs cause a potentially predictable range of outputs, and where the inputs can in principle be manipulated to lead to beneficial outputs’ (Bennett 2018: 267). This is not to say that it is never appropriate to adopt the scientific perspective in interactions with persons, but it does suggest that the participatory stance can be plausibly interpreted as involving a kind of evaluative abstinence regarding the very capacities that empirically constitute those persons. An objection to this argument for opacity respect is that it exaggerates what is required of us in order to recognize others as agents. Are there not many contexts in which we are required to assess people’s agential capacities and where our doing so is compatible with our viewing them as agents? Richard Arneson cites several kinds of situation in which someone else might act on an assessment of certain of my agential deficiencies in order to promote my life chances—for example, by providing me with tailor-made advice— without failing to treat me as an agent. ‘Robust dismissal of opacity respect need not be disrespectful in the sense of treating me as though I were a nonagent or as though my agency achievements were not crucial for my life success’ (Arneson 2015: 48; cf. Beillard 2013: 54; Charvet 2013: 70–71).⁷ In addressing this concern, it is important to bear in mind the distinction between treating an object as an x, and treating that object in ways that are appropriate for an x. The first kind of treatment does not entail the second. Arneson correctly states that the assessment of a person’s agential capacities does not necessarily involve treating her as a ‘nonagent’. My point is that it might still not involve treating her in a way that is respectful towards her in her capacity as an agent in the sense highlighted above, and so might not involve treating her in a way that is appropriate in light of those of her characteristics that are morally salient in the given context. Consider a more obvious example: if Smith is a slave-owner, then I ought, at least in most contexts, to ⁷ Some of Arneson’s counter-examples (again in Arneson 2015: 48) are of judgements about the degrees to which we should hold people responsible for their own disadvantages, judgements that take into account the levels of those people’s basic agential capacities. I discussed the implications of opacity respect for this kind of judgement, and therefore for luck egalitarianism, in Carter (2011: sec. 7). The ruling out of certain aspects of luck egalitarianism is indeed an implication of my account of basic equality. Mine is a modus ponens argument: to the extent that Arneson’s objection points to a counterintuitive outcome, my argument involves biting the bullet. Arneson’s objection is a modus tollens argument: opacity respect cannot be correct because it rules out certain kinds of responsibility-sensitivity, and the latter must be correct! Such an objection needs to be followed up with an alternative account of basic equality, for otherwise the favoured kind of egalitarianism will in any case be unsound.
Ian Carter
135
treat Smith as a slave-owner. But there are various ways of doing so, some more appropriate than others. I could congratulate Smith on being a slaveowner, or I could condemn Smith for being a slave-owner and perhaps also act, as far as my powers allow me, to deprive Smith of this legal status. In all the contexts that I can think of, the latter would be the more appropriate kind of treatment of Smith considered as a slave-owner. But this is not to say that if I congratulate Smith on being a slave-owner, I fail to treat him as a slave-owner. The question of which kind of treatment, in a more abstract sense, is appropriate for persons considered more abstractly—that is, simply as persons—is much more elusive and controversial than the question of how to treat someone occupying the specific social and legal role of slave-owner. Some might say that in the case of persons considered simply as such, the appropriate treatment consists in promoting their wellbeing; others might say that it consists in according them freedom and in holding them responsible for their choices; others still that it consists in a combination of these things, or something else again. And the distinctions between these different views can be less than clear cut. For example, those favouring the promotion of persons’ wellbeing might incorporate into the very idea of wellbeing the individual’s interest in making choices and in ‘agency achievements’ (to use Arneson’s expression). Nevertheless, just as in the case of the treatment of slave-owners, the question is: what amounts to appropriate treatment of a being in virtue of certain of its characteristics? The characteristic that I have been focusing on is that of agency, and the attitude I have been discussing is that of recognition respect for a person’s agency, as opposed to the treatment of a person’s agency as something to be improved or a good to be promoted.
4. Contexts of Practical Reason Opacity respect can be appropriate to some contexts but not to others. This fact is not damaging for my account of basic equality but might in fact constitute a strength. Context-dependence is an inevitable feature of nearly all range properties, and should not therefore surprise us: we focus on a given range property in a given context, but also on the scalar properties underlying it in other contexts, depending on our particular purposes (Carter 2021: sec. 3). My account of basic equality will be effective to the extent that the contexts in which opacity respect is appropriate happen also to be those in which we tend to favour treating people as equals, so that opacity respect can indeed be seen to explain our belief in basic equality at various levels.
136
Basic Equality and the Contexts of Opacity Respect
To clarify what is meant by ‘contexts’ here, it will be helpful to distinguish between properties, relations, and relationships. Some properties of persons are internal, such as having a brain and having two arms; others are relational and denote certain social roles, such as those of father, pupil, and employee. Some properties of persons are more abstract, such as that of being a moral agent; others are more concrete, such as that of being the referee in a particular football match. The properties that interest us here, whether internal or relational, abstract or concrete, are those that appropriately elicit a kind of recognition respect. I shall use the term ‘relationship’, on the other hand, to refer to the complete set of interactions between two or more concrete individuals occurring in light of their various relations (Carter 2013: 30–31). Different relations present us with different contexts of practical reasoning. In other words, the different relations in which person A stands to person B within a particular relationship provide A with different kinds of reason for treating B in certain ways. Thus, a single relationship can simultaneously incorporate different contexts of practical reasoning: when A reasons on the basis of one kind of relation with B and then on the basis of another kind of relation with B, A is reasoning in different contexts. When I suggest that opacity respect is appropriate in some contexts and not in others, then, I am not using the term ‘context’ to mean ‘state of affairs’; rather, I am referring to a context constituted by one or more particular interpersonal relations. Different contexts of practical reason can exist within the same spatio-temporally specified physical situation and can determine different kinds of treatment of the same person, by the same person, within that same situation. I will consider some examples in Sections 6 and 7. I have suggested that opacity respect is an appropriate attitude to adopt towards individuals when relating to them simply in their capacity as agents (Carter 2011: 556–557). In such a context, the sheer fact of their agency demands respect of the kind discussed in the previous section. Now, this might sound like a rather unnatural way to relate to other people. After all, our everyday interactions as individuals are more obviously based on ‘thicker’ relations that presuppose social roles and the various expectations associated with them—relations such as mother to daughter, employer to employee, teacher to pupil, wife to husband, or friend to friend. The idea of relating to others simply as agents is a philosophical abstraction. In everyday life we do not think of ourselves so readily as relating to others simply as agents. Nevertheless, philosophical abstraction is exactly what we need in order to explain why so many of us feel we ought to relate as equals despite our different levels of capacities. By engaging in philosophical reflection, we can come to see that, as well as relating as mother to daughter, as employer to employee, and so on,
Ian Carter
137
we are also always relating simply as agents—that is, as synchronically and diachronically integrated bundles of agential capacities of various kinds.
5. Evaluation versus Evaluative Abstinence, Compatibilism versus Incompatibilism I have argued that only opacity respect can justify focusing on a range property supervening on certain agential capacities, because only opacity respect rules out assessing the levels of the subvening capacities (above a minimum threshold). I have yet to see a convincing alternative account of the moral relevance of a range property supervening on agential capacities. Still, there are those who would prefer to continue the search for such an alternative account, because they consider opacity respect to be too strong a requirement. According to Jeremy Waldron, while the range property matters, the underlying variable scalar properties matter too, as the latter are what make people ‘sparkle’ and in this way stand out as valuable individuals (Waldron 2017: 171–172). We should not, then, embrace evaluative abstinence, as this would leave individuality out of the picture. Instead, we should allow people to ‘scintillate’, shifting our focus back and forth between the range property and the underlying scalar properties (Waldron 2017: 157–160). Unfortunately, this stance continues to leave our focus on the range property without any solid grounds. I believe that further analysis of the notion of ‘contexts of opacity respect’ can allay Waldron’s worry, allowing us to confirm opacity respect in its role as the reason for our focus on the range property, and in turn for our basic equality. There are two alternative ways of making sense of the contrasting reasons for evaluation and evaluative abstinence. I will call these a ‘compatibilist’ account and an ‘incompatibilist’ account of such reasons. The compatibilist account. According to this account, the rationale for ruling out focusing on the scalar properties is appropriate to some reasoning contexts while the rationale for focusing on the scalar properties is appropriate to other reasoning contexts, and the two kinds of rationale are compatible because their adoption in those different reasoning contexts does not produce prescriptions that are physically incompossible. Stated more commonsensically, each kind of reason grounds an independent sphere of conduct: in one such sphere we view and treat others as equals (in one respect), and in another we view and treat them as unequals (in another respect), and neither kind of treatment excludes the other.
138
Basic Equality and the Contexts of Opacity Respect
The incompatibilist account. According to this account, there are two kinds of reason that pull in different directions and cannot both be acted on. This might be (but is not necessarily) because the two kinds of reason coexist and conflict in the same reasoning context.
Both of the accounts involve accepting that there are senses in which we ought to treat people as equals, by focusing on the range property as a result of showing opacity respect, and senses in which we ought not to treat them as equals, because we ought to focus on the relevant underlying scalar properties. However, only the second of them—the incompatibilist account—requires us to address a practical conflict of values. Let us assume that in their pure forms the two accounts are contradictory: the compatibilist account implies that our contrasting commitments to evaluation and to evaluative abstinence never give rise, as a matter of principle, to practical conflicts, whereas the incompatibilist account implies that they always do. Whether either of the accounts is true in its pure form will depend on the substantive theory of justice we assume. However, on most substantive theories, neither account is likely to prove true in its pure form. Waldron’s worry, that opacity respect would prevent us from acting on the basis of certain important evaluations, presupposes the truth of the incompatibilist account in at least some of the contexts in which I have been hypothesizing that basic equality depends on the appropriateness of opacity respect. As we shall see, however, the compatibilist account is the most plausible one both in the context of relations between the state and citizens within a liberal political conception of justice and in many other contexts in civil society. In what follows, I will first focus on the plausibility of the compatibilist account (Section 6) and will then ask what space is left for the incompatibilist account (Section 7).
6. The Compatibilist Account In attempting to imagine compatible spheres of conduct in which opacity respect is and is not appropriate, the simplest starting point is a kind of relationship that involves only the ‘thin’ kind of relation calling for opacity respect. Elsewhere I have suggested that, according to a liberal interpretation of the proper functions of the state, the relationship between agents of the state and individual citizens is of this ‘thin only’ kind (Carter 2011: 557–558; Carter 2013, sec. 4). On this liberal interpretation, a person acting in the specific role of agent of the state should feel no tension between treating citizens
Ian Carter
139
as moral or political equals and treating them as moral or political unequals: she should always treat them as moral and political equals because she should always show opacity respect for them.⁸ Citizens are equals in the eyes of the state, and the opacity respect shown in this context of reasoning is compatible with various private assessments carried out ‘horizontally’. However, opacity respect owed to agents considered simply as such, and our resultant equality as bearers of human rights, need not be state-based. Cosmopolitan egalitarian relations can be made sense of in terms of individuals’ relating to each other simply as agents. For the cosmopolitan, this exercise in philosophical abstraction is one that any individual can and ought to engage in when reasoning about other individuals’ most basic rights as moral agents regardless of what other ‘thicker’ relations they stand in towards one another. In considering what those rights are, we reason in the context of our relation simply as agents, because (or at least to the extent that) it is the mere fact of our agential capacities that grounds those rights. In reasoning in that context, we can (and we ought to) adopt the attitude of opacity respect, and this explains why we accord them (or ought to accord them) an equal basic moral status, regardless of the other ways in which we relate to them, as equals or unequals, at various other levels. A similar exercise in abstraction occurs unreflectively, or less reflectively, in situations in which it comes to us spontaneously to think of another person simply as a human—for example, in emergency rescue situations or in certain situations in war.⁹ Consider now complex relationships containing both thinner relations calling for opacity respect, like the state-citizen or cosmopolitan ones just mentioned, and thicker ones calling for assessments of agential capacities, but where the two kinds of relation coexist and do not produce any conflicts of duties. The clearest such cases are those involving professional relations that require a voluntary, but circumscribed, waiving of the duty of opacity respect. For example, if I enter a relation of employment, my employer might assess various of my capacities within that context, but still adopt the attitude of opacity respect, even if applied to those very same capacities, when relating to me outside of our employer-employee relation. Elizabeth Anderson refers to something like this art of separation in discussing acceptable versus unacceptable social hierarchies: ‘When a particular person … acts outside ⁸ There will be exceptions, such as where the state acts as an employer (an exception that will occur less often under liberalism than under statist socialism). ⁹ See Michael Walzer’s discussion of the phenomenon of ‘naked soldiers’, where ‘naked’ means literally lacking in clothing but also, metaphorically, shorn of certain social roles, or perhaps of all social roles (Walzer 1977: 138–143).
140
Basic Equality and the Contexts of Opacity Respect
the colour of his office, he also has no authority over subordinates … Offduty in civil society, supervisors and employees meet as formal equals, even if rarely as friends’ (Anderson 2008: 154). My additional suggestion is that this coexistence of an inequality in one context and an equality in another can be explained by a more fundamental separation between contexts of opacity respect and contexts of evaluation. This art of separation might apply even to the most basic of agential capacities. Consider a situation in which I, a private individual, seek help from a psychotherapist, where the latter is able to diagnose a personal problem and propose a therapy only by assessing, and acting on, certain deficiencies of mine in terms of basic agential capacities. (Assume that, at least in an overall sense, my basic agential capacities are nevertheless above the threshold. If they were not, there would be no case for opacity respect in any context and therefore no prima facie tension to address.) Once I start seeing the psychotherapist in her professional capacity, I enter a specific kind of relation within which it is not appropriate for her to show me opacity respect. From this fact, together with my account of basic equality, it follows that the psychotherapist lacks a basis for considering me an equal within our professional relation. Instead, she has reasons for acting on the basis of various of my scalar agential capacities (some of which might be lesser than her own, while others might be greater). However, our professional relation, which involves transparency rather than opacity, does not compromise my status as a citizen or person equal to other citizens or persons, including to the therapist outside her professional capacity. Nor does it imply that I lack any rights against the therapist, within our professional relation, in my capacity as an agent (the therapist is constrained, for example, to seek my consent to certain forms of treatment). When I bump into the therapist in the supermarket or at a party-political meeting, where the relation as therapist to patient is no longer relevant, we show opacity respect to one another, participating in the collective focus on the relevant range properties, and we therefore relate as equals. If the therapist fails to do so, I will feel an uncomfortable sense that boundaries are being inappropriately crossed and my status as an equal in the public sphere is being undermined. The therapist should not act on her professional assessments of my individual basic capacities in such a situation, and should treat such assessments as confidential: doctors are indeed expected not to render public their assessments of their patients’ personal abilities, and it is no business of other private individuals to go about collecting such information. Even during our therapy sessions, the psychotherapist can and ought to practice an art of separation between different contexts of reasoning, treating me
Ian Carter
141
simultaneously with opacity respect in my capacity as a citizen and without opacity respect in my capacity as a patient, with the two contexts occupying interchangeable positions at the back of her mind or at the fore: she can be aware at any moment that she relates to me both in her professional capacity as a therapist and as a fellow citizen and indeed as a fellow human being with a life to lead outside our professional relation. All of these considerations support the compatibilist account of the therapist’s contrasting reasons for evaluation and for evaluative abstinence. They also track plausible views about the appropriateness of treating people as equals only in certain contexts of practical reason. The art of separation looks much easier in the case of the professional relation between psychotherapist and patient than in everyday intimate relations. Nevertheless, if we take that professional relation as our starting point, we can see how the model can be extended to cover many aspects of intimate relationships. Not only can couples and friends in intimate relationships treat each other as equals when they enter the voting booth or discuss politics in a wider group, viewing each other through public eyes; they can also consider and treat each other as equals in many other aspects of their relationships, where the basis for their equality is a mutual respect for the choices the other makes and for their choice-making abilities, a kind of respect that involves avoiding viewing the other as a being whose powers of discrimination and judgement stand in need of analysis and assessment.¹⁰ Mark Schroeder has provided some interesting examples to illustrate how we might treat another person, momentarily and in a particular context, as a ‘thing’, in the sense of treating their voluntary behaviour as a mere part of the causal world rather than as a product of their agency (Schroeder 2019). One such example is where Schroeder’s wife criticizes him in an apparently irrational way and he puts this behaviour down to her having had a bad day. Such a causal explanation seems ‘diminishing’ in the sense of implying that she is less of an agent than she might be. And we might also add, in the present theoretical context, that it violates the requirement of opacity respect to the extent that it implies an agential incapacity on her part to resist certain external influences. According to Schroeder, however, not only are such isolated instances of causal explanation unavoidable in certain contexts, but they can even be necessary in order to treat the other individual as a person, identifying the heteronomous episode exactly in order to ‘see through’ it and focus exclusively on the contributions of the person that are genuinely ¹⁰ Friends and couples often divide labour on the basis of who is better at what, but the abilities in question need not be those upon which the relevant range property supervenes, in which case no duty of opacity respect is violated.
142
Basic Equality and the Contexts of Opacity Respect
agential. There is a causal background to every person’s agential contributions which includes any number of physical causes acting on our bodies. We assign the heteronomous episode to that same causal background and then adjust our perception of the person in the same way as our eyes adjust to different lighting conditions. The heteronomous episode ‘is there, but it is something to be seen through, rather than visually inspected, and once you have properly incorporated it into the background—once your eyes have adjusted to the new conditions—it is effectively invisible, and you see objects as the same colors as before you took off your sunglasses’ (Schroeder 2019: 104). This way of according the Strawsonian objective perspective a place in interpersonal relations seems again to support the compatibilist account of the two commitments to evaluation and evaluative abstinence, since it points to a way in which the treatment of the person as a thing can be contained and neutralized, so excluding the moral conflict that might otherwise arise. Indeed, the prescription that we ‘see through’ the causal explanation, rather than ‘visually inspecting’ it, can be interpreted as introducing a kind of exclusionary reason similar to evaluative abstinence, only in a second moment that follows our explicit recognition of the causal explanation and of any implied agential shortcomings.
7. The Incompatibilist Account Wherever the compatibilist account proves to be true, the contrasting commitments to evaluation and evaluative abstinence create no obvious problems for my account of basic equality. If the implied courses of action do not conflict, then basic equality is preserved in the relevant spheres of action. It is not clear that any stronger kind of conclusion is required of a theory of basic equality. But what if there are situations that instead confirm the truth of the incompatibilist account? To recall: these are situations in which the contrasting commitments to evaluation and evaluative abstinence provide reasons for courses of action that cannot all be carried out. Such conflicts might be resolved by weighing the contrasting commitments against one another or applying some kind of priority rule. Alternatively, they might be experienced as dilemmas to which there is no rational solution. It is not implausible to apply the incompatibilist account at least to certain aspects of intimate relationships, where respect for the other’s agency, on the one hand, and the desire to improve their choices or choice-making abilities, on the other, can create moral conflicts in a way that Schroeder’s examples
Ian Carter
143
of causal explanation do not appear to. We might feel, for example, both that we ought, out of a duty of care, to engage in paternalistic behaviour towards a friend or partner that we judge to be acting against their best interests, and that we ought not to engage in such behaviour. The first commitment derives from a negative evaluation of the friend’s or partner’s agential capacities, whereas the second, if I am right about the appropriateness of opacity respect in such a context, can be understood as deriving from a refusal to engage in exactly that evaluation, a refusal that confers a kind of ‘outward dignity’ on the friend or partner. The context of practical reason in which these conflicting commitments occur appears to be the same: it is the context in which we are relating to another simply as an agent. We perceive that their rational capacities, or their strength of will, are not at pathologically low levels, and therefore that we ought to allow them to choose for themselves or at most that we ought to ‘engage’ with their reasoning capacities without considering those capacities themselves to be in need of assessment. But we might also be tempted to subject their abilities to critical scrutiny with a view to improving them, or perhaps to adopt a persuasive strategy the nature of which is influenced by our perception of the level of their cognitive or volitional capacities, as in some of the examples of Arneson mentioned earlier. My purpose, here, is not to pronounce on the correct course of action in such situations. Rather, I wish to make two conceptual points. The first is that the independent case for opacity respect continues to exist as a pro tanto moral requirement even if it is permissibly rejected in specific contexts within a radically value-pluralist perspective. Adopting an attitude of care or compassion or pity based on negative evaluations in specific contexts does not involve treating the other individual as a ‘nonagent’ or make their agency disappear, but, if that other individual satisfies the requirements of moral personality, and if the incompatibilist account is true, then neither does the adoption of such an attitude occur without moral loss. The second point is that, on my account, such renunciations of opacity respect also amount to renunciations of basic equality, because they amount to denying the moral salience of the range property, which is the property that is equally possessed. And this implication, of a falling away of basic equality within a context in which it ought, for reasons of opacity respect, to be present, is not an implausible inference. For if we hold the incompatibilist account to correctly apply in a particular private relationship, and we undergo the experience of feeling torn between opacity respect and its rejection, then it does seem to be true, phenomenologically speaking, that we also undergo the experience of feeling torn between treating the other as an equal, on the one hand, and not doing so, on the other.
144
Basic Equality and the Contexts of Opacity Respect
If this second point is correct, then examples like that of the paternalist friend do not amount to a challenge to the idea that opacity respect is an essential part of what grounds basic equality. Rather, such examples suggest that in some private relationships, some individuals, perhaps even without wrongdoing (if we adopt radical value pluralism), are compromising a kind of basic equality that would otherwise characterize such relationships. Neither does this conclusion rule out the claim that opacity respect should be understood as the default from which any departure is in need of justification. Notice, moreover, that the compromising of basic equality in such cases is local in the sense of being specific to certain relations within certain personal relationships. Even if you can treat your friend paternalistically in some personal matter (whether rightly or wrongly), at least when you accompany that friend to the voting booth in a political election, the partial truth of the compatibilist account becomes self-evident: you should show opacity respect for them as a political agent and on this basis you should treat them as someone with a right to no less (or more) voting power than anyone else. (Suppose that what actually happens is that you instead treat them as a citizen whose powers of judgement stand in need of analysis and assessment, or that you try, on the basis of such an assessment, to manipulate them into voting in a certain way. On a non-instrumental, respect-based account of democracy, such attitudes and actions would be clearly inappropriate. Even worse, on such an account, would be for public officials to show such attitudes or to engage in such behaviour.) Might the incompatibilist account sometimes also be applicable in the public sphere? Only, I submit, if one assumes a normative political theory that is less liberal than the one I assumed earlier in setting out the case for compatibilism. In other words, as far as the public sphere is concerned, my thesis is that the basic equality of citizens presupposes, in at least one important sense, a liberal conception of the state, so that to compromise that liberal conception, in that important sense of ‘liberal’, is to compromise citizens’ basic equality. If I am right about this, we have the interesting conclusion that statebased egalitarianism depends on liberalism, in at least one important sense of ‘liberalism’.¹¹ To clarify this relation between liberalism and basic equality, consider again the case of the therapist or social worker. From within the liberal perspective, we can imagine individuals voluntarily seeking help, at most drawing on universally available subsidies which the state can provide ¹¹ Christian Schemmel similarly sees the requirement of non-intrusiveness as necessarily constraining public measures promoting an ideal of relational equality grounded in opacity respect (Schemmel 2021: 108–109, 159).
Ian Carter
145
without prying into the levels of any particular individuals’ basic agential capacities. Now imagine, instead, that the state instructs its officials actively to identify individuals who, despite being in the normal range of basic agential capacities, would benefit from improvements in those capacities (as in Floris 2019a). In this case, the reason those officials have for focusing on the range property of moral personality will no longer apply to any of the citizens whose capacities they assess with a view to helping them improve their powers of choice. And without the relevance of the range property, such citizens are no longer equals in the eyes of the state. This kind of illiberal intrusive state cannot be an egalitarian one.¹²
8. Conclusion In this chapter I have defended my claim that basic equality is grounded in opacity respect. I have not defended that claim against all objections, but have focused exclusively on the objection, raised by a number of critics, to the effect that opacity respect is often not a fitting moral attitude and therefore cannot provide a valid justification for basic equality. I have tried to show that, while opacity respect is indeed appropriate only to certain kinds of social relation, the same is true of basic equality, and that the liberal commitment to basic equality, both in the public sphere and in many individual interactions, tracks a liberal commitment to opacity respect.¹³
References Anderson, E. 2008. ‘Expanding the Egalitarian Toolbox: Equality and Bureaucracy’. Proceedings of the Aristotelian Society, Suppl., 82: 139–160. Andersson, E. 2022. ‘Distributive Justice, Social Cooperation, and the Basis of Equality’. Theoria, 88: 1180–1195.
¹² It might be open to such officials to carry out their investigations secretly, perhaps observing individuals’ behaviour by means of hidden cameras and microphones in order not to appear to compromise their public stance in favour of opacity respect. In this case, the agents of the state would be publicly supporting basic equality while secretly rejecting it. ¹³ I am grateful to the editors, Giacomo Floris and Nikolas Kirby, as well as to Teresa Bejan, Julie Tannenbaum, and an anonymous reviewer for Oxford University Press, for helpful comments on a draft of this chapter. I am also grateful to audiences in Manchester, Dublin, and Milan for discussions of previous versions. I am especially indebted to Giacomo Floris for any number of searching discussions on the topic of this chapter. The research for this chapter was completed with the support of the Italian Ministry of Education, Universities and Research (PRIN 2017S4PPM4_002).
146
Basic Equality and the Contexts of Opacity Respect
Arneson, R. J. 2015. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth? On Basic Equality and Equal Respect and Concern, edited by U. Steinhoff, 30–52. Oxford: Oxford University Press. Beillard, J. 2013. ‘Equality and Transparency’. American Philosophical Quarterly, 50: 51–62. Bennett, C. 2018. ‘Intrusive Intervention and Opacity Respect’. In Treatment for Crime. Philosophical Essays on Neurointerventions in Criminal Justice, edited by D. Birks and T. Douglas, 255–73. Oxford: Oxford University Press. Carter, I. 2011. ‘Respect and the Basis of Equality’. Ethics, 121: 538–571. Carter, I. 2013. ‘Basic Equality and the Site of Egalitarian Justice’. Economics and Philosophy, 29: 21–41. Carter, I. 2018. ‘Equality of Opportunity, Responsibility, and Personal Identity’. Ethical Theory and Moral Practice, 21: 825–839. Carter, I. 2019. ‘Self-Ownership and the Importance of the Human Body’. Social Philosophy and Policy, 39: 94–115. Carter, I. 2021. ‘Basic Equality Revisited: Range Properties, Status, and Opacity Respect’. Quaderni di Scienza Politica, 28/2: 179–198. Carter, I., and Page, O. 2023. ‘When is Equality Basic?’ Australasian Journal of Philosophy, 101: 983–97. Charvet, J. 2013. The Nature and Limits of Human Equality. Basingstoke: Palgrave Macmillan. Christiano, T. 2015. ‘Rationality, Equal Status, and Egalitarianism’. In Do All Persons Have Equal Moral Worth?: On Basic Equality and Equal Respect and Concern, edited by U. Steinhoff, 53–75. Oxford: Oxford University Press. Cupit, G. 2000. ‘The Basis of Equality’. Philosophy, 75: 105–125. Darwall, S. 1977. ‘Two Kinds of Respect’. Ethics, 88: 36–49. Darwall, S. 2006. The Second Person Standpoint: Morality, Respect, Accountability. Cambridge, MA: Harvard University Press. Enoch, D. 2022. ‘Autonomy as Non-alienation, Autonomy as Sovereignty, and Politics’. Journal of Political Philosophy, 30: 143–165. Floris, G. 2019a. ‘Moral Status, Equality, and Distributive Justice’. PhD Thesis, University of Manchester. Floris, G. 2019b. ‘On the Basis of Moral Equality. A Rejection of the Relation-First Approach’. Ethical Theory and Moral Practice, 22: 237–250. Floris, G. 2023. ‘Contingency, Arbitrariness, and the Basis of Moral Equality’. Ratio, 36: 224–34. Korsgaard, C. M. 2009. Self-Constitution. Agency, Identity, and Integrity. Oxford: Oxford University Press. Lippert-Rasmussen, K. 2016a. Luck Egalitarianism. London: Bloomsbury.
Ian Carter
147
Lippert-Rasmussen, K. 2016b. Opacity, Respect, and Equality. Paper presented at the workshop on equality, LUISS, Rome, November 2016. Miklosi, Z. 2022. ‘The Problem of Equal Moral Status’. Politics, Philosophy and Economics, 21: 372–392. Phillips, A. 2021. Unconditional Equals. Princeton, NJ.: Princeton University Press. Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Sangiovanni, A. 2013. ‘On the Relation between Moral and Distributive Equality’. In Cosmopolitanism versus Non-Cosmopolitanism: Critiques, Defenses, Reconceptualizations, edited by G. Brock, 55–75. Oxford: Oxford University Press. Sangiovanni, A. 2017. Humanity Without Dignity. Cambridge, MA.: Harvard University Press. Scanlon, T. M. 1998. What We Owe to Each Other. Cambridge, MA: Harvard University Press. Schemmel, C. 2021. Justice and Egalitarian Relations. Oxford: Oxford University Press. Schroeder, M. 2019. ‘Persons as Things’. In Oxford Studies in Normative Ethics, vol. 9, edited by M. Timmons, 95–115. Oxford: Oxford University Press. Sher, G. 2015. ‘Why We Are Moral Equals’. In Do All Persons Have Equal Moral Worth?: On Basic Equality and Equal Respect and Concern, edited by U. Steinhoff, 17–29. Oxford: Oxford University Press. Steinhoff, U. 2015. ‘Against Equal Respect and Concern, Equal Rights, and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On Basic Equality and Equal Respect and Concern, edited by U. Steinhoff, 142–72. Oxford: Oxford University Press. Strawson, P. F. 1974. ‘Freedom and Resentment’. In P. F. Strawson, Freedom and Resentment and Other Essays, 1–25. London: Methuen. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: Harvard University Press. Walzer, M. 1977. Just and Unjust Wars. New York: Basic Books. Williams, A. 2020. ‘The Basis of Equality and its Implications’. Paper presented at workshop on ‘Basic Equality’, MANCEPT Workshops in Political Theory, 9 September 2020. Wollner, G. 2015. ‘Basic Equality and the Currency of Egalitarian Justice’. In Distributive Justice and Access to Advantage. G. A. Cohen’s Egalitarianism, edited by A. Kaufman, 180–203. Cambridge: Cambridge University Press.
6 Equal Moral Status and the Collective Nature of Rationality Thomas Christiano
1. The Puzzle In this chapter, I explore the establishment of a conception of fundamental moral status for human persons and whether it can be equal moral status. The problem can be expressed in terms of a set of propositions that seem not to fit well together: there is a hierarchy of statuses; the moral status of persons is grounded in rational capacity; persons have rational capacities to different extents; persons have equal moral status. It seems to some that these propositions cannot all be true, and yet they all seem to be true. How can this be? I want to propose an argument that reconciles these propositions and I want to propose a positive argument for why they should be reconciled. First let us get some basic ideas on the table. What is moral status? I take fundamental moral status to have an explanatory role. The grounds of the fundamental moral status explain why it is fitting that the object that has this status is due a certain kind of treatment. And the kinds of treatment in the case of persons are respect for certain basic rights and the equal advancement of their wellbeing or opportunities for wellbeing. The explanatory role of status is that the grounds of status identify a kind of fundamental and impersonal worth that the being has. And it is in virtue of that impersonal worth that the being has the moral status that it has and is owed what it is owed. There are different moral statuses, and there is hierarchy among statuses. Living things have a kind of moral status, sentient beings have a kind of moral status. Beings that move themselves have a kind of moral status. Rational beings have distinctive moral status grounded in the impersonal worth of rational nature, which is a reflective and progressive notion. I will say more
Thomas Christiano, Equal Moral Status and the Collective Nature of Rationality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0007
150
Equal Moral Status and the Collective Nature of Rationality
about this kind of rationality below. Some kinds of beings have a higher status than others. This is because they have a greater impersonal worth.¹ The consequence of hierarchy is that beings with higher status have greater moral advantages than beings with the other moral statuses. Greater moral advantages means that more demanding rights are associated with the status. This may involve the same duties but with greater weight in the case of higher status beings. For example, two beings with rights to life may have different conditions for defeating that right; the being with lesser status may have the right to life defeated by lesser competing conditions than those that defeat the rights of the being with a higher status. In some cases, beings with different status will be associated with different duties. A right to life is normally not associated with merely living beings. Some think that animals have rights to life and most think that human persons have rights to life. A right to freedom of expression is normally associated only with rational beings such as human persons, but not other animals or living beings.
2. Autonomy and Moral Status One way to think of status and the hierarchy of statuses is to think of status as being associated with a level of autonomy.² I do not use the word ‘autonomy’ here in the sense of rule over oneself or rational self-governance, but in a broader sense. I have in mind that any being that in some sense is the source of its own activity is autonomous. We can attribute an aim to it and some means for achieving that aim. But there are increasing levels of autonomy. Living things have some autonomy, which non-living things do not, in the sense that they have an internal principle of growth and survival. Conscious beings have greater autonomy because they represent the world to themselves as a part of their reaction to that world. They are in greater possession of themselves by doing this. They do not merely move towards an aim; they represent the aim and the world to themselves and pursue the aim on the basis of their understanding of the world. Beings that are minimally rational are moved by desires and beliefs about how to achieve those desires. And reflectively rational beings reflect on their understandings of value and the world and try to improve them. These are ascending levels of autonomy. ¹ In this chapter, I do not defend the central role of worth in the account of status or the idea that rational nature is the main element in the status of persons. Here my concern is with equality of status. I defend the worth of rational nature at more length in Christiano (2014). See Sher (2014) for an alternative grounding. See also Kittay (2005) for an argument against properties as the basis of status. ² See Kagan (2019) for a similar claim about the relation of status and autonomy.
Thomas Christiano
151
These beings are not merely increasingly sophisticated. They are more autonomous in the sense that they are more the source of their own activity. It is one thing to be orientated to achieve some aim; it is quite another to represent that aim to oneself and represent the world to oneself and then choose how to pursue the aim in the light of these representations. It is a much greater level of autonomy to be able to evaluate one’s aims and to be able to revise them rationally and reflectively. The increase in autonomy is not always to be understood in terms of increase in the degree of the possession of some other property. The increase can have a qualitative character in that the possession of certain new capacities (such as the powers of reflection and self-evaluation) imply greater autonomy. The greater level of autonomy is connected with the greater impersonal worth of the being, though it is not the only feature that accounts for that impersonal worth. There is something extraordinary in the possession of life but even more wonderful in the possession of beliefs, desires, and consciousness and their determination of action. The most wonderful of all is the capacity to reflect and rationally transform one’s beliefs, desires, and even consciousness. This impersonal worth is deeply connected with moral status. The intuition behind this is that autonomy in the broad sense in which I intend it here implies that there is a being to which one attributes its own aim and its own activity to achieving that aim. It is this quality of having its own that gives the being status and that explains why it has wellbeing and, ultimately, rights.
3. Moral Status and Public Status Furthermore, fundamental moral status is distinct from equal public status elaborated by democratic theorists, which serves to ground the inclusion of persons in civil and political society. Young children, for example, do not have full civil and political status in democratic societies. They do not have rights to make contracts or vote or run for office. Yet, young children do have equal moral status which they share with other human persons. Equal public status is meant to ground equal civil, economic, and political rights in the context of the normal disagreement and conflict of interest we find in any moderately complex society. It is a kind of principled resolution to disagreements about morality and justice and even about the capacities people have for arriving at correct conceptions of these matters. We recognize this disagreement and resolve it by taking a public view of citizens, awarding them
152
Equal Moral Status and the Collective Nature of Rationality
equal public status. Institutions that realize equal public status such as democracy and liberal rights then form the background in which persons can argue with each other and try to realize their competing conceptions of value. Equal moral status is distinct from equal public status because it is fundamental to all of morality and justice and is not a response to disagreement or other problems of rationality or morality in a complex society. It attaches to persons qua persons, while different civil and political statuses attach to different parts of a person’s life. Moral status grounds all the ways in which we deal with the problems of disagreement and with the different stages of a person’s life.
4. The Puzzle and Range Properties Now we can see that there is a potential worry in the collection of propositions that have been made regarding moral status. The superior moral status of persons is grounded in their possession of rational capacity, but individuals are generally thought to have greater or lesser rational capacities and yet we still think that they have equal moral status. How are these propositions to be reconciled?³ One might simply say that the propositions are reconciled by saying that the ground of status is a range property, as Rawls asserts.⁴ This would mean that the rationality of persons is a range property such that every being that satisfies some minimum threshold of rationality is equally rational in the range sense, even if some have greater rational abilities or capacities. There may be an upper bound to the range as well. This shows that the problem of the propositions about status is not a problem of logical inconsistency. I believe that the idea that the status ground is a range property is correct and it shows that there is no problem of logical consistency. But it does not show us all that we want to know. The problem of the propositions about status has to do with the grounds of status and equality. It raises the question why we should think that rationality is a ground of status but not think that greater rationality is a ground of greater status. To see the distinctive nature of the problem, we can formulate it in the following way: why should we think of rationality as a range property in this context? Of course, it is logically possible to view it in this way, but we have not been given an argument for
³ See Arneson (1999) for a seminal and powerful discussion, and see also McMahon (2008). ⁴ See Rawls (1999). Waldron (2016) also endorses this approach.
Thomas Christiano
153
thinking of it this way. Certainly, the idea that the status ground is not a range property is not an implausible one on its face. So, though the possibility of status being grounded in a range property does show that there is no logical inconsistency in the propositions about status, we are not out of the woods yet. This is because we need to have in addition some positive reason for thinking that the grounds of the status of persons really are range properties. To try to answer this question, I will articulate my conception of imperfect rational nature and the collective character of impersonal rational nature. I will articulate my conception of equal moral status. I will then articulate some intuitive arguments for the idea of equal moral status, despite the above sources of concern. I will show that we are not forced to adopt a thesis of unequal moral status on the basis of unequal rational capacities. But, going beyond this merely negative argument, I will then give some positive argument for the thesis that reflective rational natures are equal in moral status based on the collective nature of rationality. But I will argue that there are limits to this strategy, and I show what they are. I will close with a discussion of some possible further objections to the conception and argument I lay out.
5. Imperfect Rational Nature I think of rational nature as a reflective capacity that is constantly in a position to re-evaluate the ideas it has about the different kinds of values that it uses to guide its behaviour and attitudes to the world. It is in this position because it understands that its ability to apprehend what is valuable is highly fallible and potentially subject to improvement and because it is aware of the deficiency of its grasp on what is valuable. Because of its rational nature, it desires always to have a correct grasp of value. When it reflects on its imperfection, it must desire to improve its grasp. What it means to be a limited and embodied rational being on this view is that it is a being that is always in a position to struggle to improve its understanding of value and the capacity of value to guide its activity. To modify a passage of Hobbes’s, human rational nature is engaged in a restless pursuit of understanding that ceases only in death.⁵ In this sense, human rationality is radically different from what we might conceive divine rationality to be. Divine rationality is perfect and complete. It does not struggle, but contains within itself the complete realization of understanding and its realization in action. It does contain a self-conscious conception of value, but that conception is never in doubt. ⁵ See Hobbes (2000).
154
Equal Moral Status and the Collective Nature of Rationality
The elements of this picture of imperfect rational nature are that a being that has this has the capacity to have: (1) a conception of what is valuable and what is not and some ability to put this conception into individual and collective action; (2) an awareness that whatever conception one has is a possibly mistaken conception of value; (3) a capacity to rethink what is valuable in the sense of re-evaluating the conception of value and modifying it; and (4) a desire to improve its understanding of what is valuable and the ability to conform its actions to it. The notion of value I have in mind is a capacious one including goods, rights, justice, and worth. And the struggles that people undergo are not necessarily or even usually about the most abstract issues, but often about very concrete issues. Sometimes it is not about the values on their own, but about the trade-offs among the values. I certainly do not mean to suggest that people are always trying to improve their understandings of value in the sense that it is a daily or even hourly activity. Perhaps the best way to put it is that most people do this some of the time and many people do it a lot of the time. Also, in order to grasp this restless striving, it is important to note that the struggle is a collective one. While individuals are often content with the understandings they have, the society overall is struggling with problems, contradictions, absurdities, and inadequacies in its conceptions of value and in the implementations of value. The society struggles in the sense that many members are engaged in ongoing dialogues on these matters, ones in which any and all can join, and which influence the beliefs of nearly everyone. This restless ongoing dialogue and debate is part of what is responsible for transforming the society over time. And the society struggles with these because it is constantly facing new problems and difficulties and people disagree in the society on these matters and argue with each other over them. As societies go through changes in values, the individuals in them experience the struggle between an older set of understandings of value and the new. They often experience these new ones as revelations, and not mere changes of attitude; they wonder what they were thinking before. And of course, sometimes people experience the new valuations as serious mistakes and long to retain the older valuations or perhaps to develop different ones. This vision of rational human beings is a kind of Millian modification of the scholastic Aristotelian picture. It introduces a kind of dynamism in the conception of rationality that is not really fully present until we get to Mill.⁶ And it introduces a new evaluative component that is not present earlier. The traditional conception of rationality affirms that it is irrational not to conform ⁶ See Mill (1989).
Thomas Christiano
155
one’s actions and emotions to reason and it is irrational not to develop the right conception of value. The Millian addition partly denies the second thesis and affirms a third thesis, which is that it is often irrational to remain content with the conceptions that one has. There is always room for improvement. A rational being necessarily desires an increase in its understanding of value. This is because a rational being desires things because they are conceived to be genuinely valuable. To be content with an inferior understanding of value would be to undermine that aim. This may not entail that there are always reasons to change one’s valuations, since one may not have anything better in hand. But it does mean that there are times in any person’s life when they can be held accountable for not updating their understandings, and each person understands this. Hence each person participates in this restless striving for better understanding. None of this is meant to imply that there is unending and perpetual progress in understanding of value. Nor does it imply that individuals are never complacent or stuck in muddled ideas. A human being is, in Montaigne’s memorable words, a frail bark at sea in a terrible storm.⁷ There is a lot of backsliding and regression; there are a lot of blind alleys and horrible ideas. But the restless striving does not stop. There is the possibility of progress here, but it is by no means a guarantee. And none of this is meant to suggest that ideas rule the world. But they do play an important part. It is worth pausing briefly on the notion of ‘capacity’. I can only develop this idea intuitively here. This notion is meant to be distinct from ‘potential’, which is a wider notion. To have a capacity to x implies that one is in possession of a mechanism that under normal circumstances of development, such as nutrition and non-violence, will enable one to develop the ability to x. To have a rational capacity does not entail that one can actually think rationally. It entails that one has all the complete mechanism that will become able to think rationally under normal circumstances of life. Hence, infants have this capacity, though they are not yet able to exercise it. Mice do not have this capacity because they are not in possession of a mechanism that will develop into the ability to reason. A mouse has a potentiality for rationality in that if one adds a new mechanism to the mouse’s brain, it might be able to do this. One further idea is that one may be able to enhance a capacity with drugs or training without fundamentally changing the capacity. It is essential to see the difference between capacity and potentiality to see how the argument works. The possession of a capacity of reason is essential to rational nature. ⁷ Montaigne (1980).
156
Equal Moral Status and the Collective Nature of Rationality
6. The Collective Character of Rational Nature Rationality is essentially collective in a number of different ways. The society in which we live serves as a necessary background to our thinking; it contains much of the content of our thinking, it tests the warrant for our thinking, and it provides purposes to our thinking. First, collective activity is a necessary background condition to the exercise of rational capacities as we know it. The prime example of this is language, without which the capacity for reflecting on the good, the right, and the true would be non-existent, or so fundamentally primitive as to not be the same kind of thing. More substantively, we all start with a very rich background of priors, which we inherit from family, school, and in many other ways from the larger society and which we continue to receive throughout our lives. Without this rich background of priors, we would have nothing to reason from or to. Our own thinking usually just consists in repetitions and local applications of, or dialogues with, and sometimes marginal changes to, this massive body of inherited thought. Without this massive background we would be nothing more than stupid animals. To be sure, some people contribute a bit more to this collective activity than others, but we are mostly just rehearsing the background ideas which define our thinking. Second, collective activity is necessary because of the central importance of the division of labour in cognition. One way in which this occurs is through distributed cognition. Our thinking simply cannot have any kind of serious sophistication without being linked up with other people’s thinking. Any society with a division of labour has a cognitive division of labour, which enables people to act and think on good reasons without actually possessing those good reasons in their heads.⁸ Doctors, mechanics, lawyers, policy economists, building inspectors, civil engineers, and many other professionals confer warrants on our thinking that we are not able to confer ourselves. We depend on their expertise in a myriad number of ways. Indeed, they may confer warrant on our thinking even when we don’t know that they do. A second way involves a different kind of division of labour that is essential to the growth of human knowledge, which is the one Mill points us to.⁹ That is the whole system of argument and counter-argument on which scientific reasoning, artistic development, and political deliberation depend. We depend on others and communities of others developing their own ideas and arguments, and on their attempting to criticize our ideas and arguments. This process of ⁸ See Hutchins (1995). ⁹ See Mill (1989: Ch. 2).
Thomas Christiano
157
give and take of reasons amounts to a kind of system of trial and error in the evaluation of all the ideas available to society.¹⁰ Many ideas and arguments are presumed to have survived this process and thus acquire some warrant just from that. Third, a large part of practical reasoning itself has an essentially distributive character to the extent that for collective activities, on which the welfare of societies depends, we each think in terms of contributing our share on the assumption that others contribute theirs. The reasons we have for action depend on the reasons others have. This is essential given the previous two observations because our contributions to the world around us and to the thought of that world are fairly small at most. So, the reasons we have to make those contributions, aside from the individual rewards of status and resources or our role in the division of labour, are primarily that we are doing our small part. For example, we have reason to pay our taxes in many societies even though if one person does not pay their taxes, no significant bad would occur. We also have reasons not to contribute to outcomes to which we make insignificant contributions, such as carbon-producing activities that contribute to climate change. Our own action may be insignificant, but many such actions in combination can be significant. Here our reasons for action depend on a kind of collective possession of reasons for action of many people.¹¹ These collectively derived reasons for action are given concrete articulation in social norms and law which are essential to coordinating those reasons. In essence we have to think of our rational capacity in large part as a capacity to participate in the collective activity that shapes the background, content, warrant, and purposes of the exercise of reason. The quality of the warrant each person derives for her thoughts from these processes depends essentially on the quality of the institutions and networks that people find themselves in. Institutions that encourage the development of a multiplicity of points of view and reasoned debate among them are likely to encourage much higher quality ideas. They can be undermined to some extent by corruption in the professions, acrimonious polarization, mutual distrust, or institutions that discourage or actively suppress discussion and debate. Moreover, the social system may work better for some than for others, given their positions in the division of labour. The institutions that are productive of reasoned thought in one part of the society may be absent in other parts. It can also be undermined by a weak or very uneven system of ¹⁰ See Muldoon (2013). ¹¹ See Nefsky (2017) for an analysis of this kind of reason for action.
158
Equal Moral Status and the Collective Nature of Rationality
education or institutions that do not help people become informed. All of this is a consequence of the fact that rationality is essentially a collective product. Its quality depends on the quality of the collective itself. There are two big questions that arise here. One is, how is individual moral standing based on rational capacity possible here when it is really the collectivity that is the main driver of this capacity, and no individual is necessary though there have to be some individuals? After all, the collective rational activity does require that there be some persons to sustain the activity, but it does not require that any particular person be there to sustain it. This is a potentially unsettling question because it would seem to permit the sacrifice of individuals to the collective. Why not just say that the collective has standing, and individuals are subordinate to it? Or perhaps the standing of individuals depends on their relations to the larger collective? The question is, how can persons have standing as inviolable persons when they and the rational activity that is the basis of their impersonal value are so deeply embedded in the collective activity? In fact, it is not clear that it is a genuine worry. Recall that a basic marker of status is autonomy or self-governance. The collective is not self-governing, at least not to the degree that even plants are self-governing; only individuals have this feature. Even though their self-governance is deeply dependent on the collective, it does not subordinate them to its status because the collective itself is not a self-governing entity and thus does not have independent moral status. No doubt the deep dependence of individual rationality on the collective does diminish their autonomy, but it does not imply that the collective has autonomy or status. To be sure, the collective can acquire a certain level of self-governance, but that is the consequence of individuals organizing themselves in a certain way in terms of a certain conception of justice and of collective decision-making. The self-governance of the community even in these cases is deeply dependent on that of persons. Furthermore, the deep dependence of individual rationality is not inconsistent with the idea that the autonomy of individual persons is at a higher level than the autonomy of sentient or self-moving beings.
7. An Intuitive Argument for Equality Despite Different Rational Capacities The worry suggested by the set of propositions above is that equal status is not compatible with variation in possession of the natural rational capacities. By variation in possession of the natural rational capacities, I mean variation
Thomas Christiano
159
in the capacity to achieve insight into value or to make progress in achieving deeper or clearer insight into value. The idea is that if possession of a rational capacity grounds status, then why doesn’t possession of greater rational capacity ground greater moral status? The grounds of status are continuous. I call this the continuity argument. Different statuses might entail a greater claim to opportunity to achieve wellbeing, and though the basic rights may be the same, there may be greater weight attached to the rights of the more rational. A first response to this argument helps us see how it is possible for the underlying ground of status to be a range property. I argue that possession of natural rational capacity is the basis of equal moral status while differences of natural rational capacity are differences in natural talent. First, it is a natural question to ask concerning a principle that assigns greater claims to wellbeing to those who are more rational whether it is unfair that some have more opportunity for wellbeing than others. The approach that asserts that inequality of status arises from different rational capacities suggests that this is not unfair or unjust because the differences in question are status-conferring differences. But the question does not simply go away as a result of this theoretical manoeuvre. There is still a lingering sense that greater rights or opportunity for wellbeing for the more rational is ‘giving more to those who have more’. This suggests that when people are thinking about differences of rational capacity, they are thinking about differences in native talent, which seem generally to be unjust grounds for differences in access to wellbeing. Let us call this the fairness consideration. The fairness consideration is not that it is unfair that someone might have a lower status. That would be a problematic assertion, since status is supposed to be the basis of how we are to treat people. The fairness consideration is an intuitive consideration that suggests that there is unfairness in someone having lesser rational capacities that then leads to lesser opportunities for wellbeing. The intuition that there is unfairness suggests that we are not dealing with beings of different status, or at least that differences in rational capacity are not grounds for different status, because there is an intuition of unfairness here. Second, the fairness consideration gives another reason for status being equal for all despite differences in rational capacity when we think in terms of the following intuitive example. Suppose that we have two people, who are otherwise the same, having very different appreciations when listening to a piece of music, say Beethoven’s first Razumovsky Quartet. One person really plumbs the depths of the piece while listening to it, the other does not, though he does thoroughly enjoy it. And let us suppose that this difference is
160
Equal Moral Status and the Collective Nature of Rationality
entirely the result of greater rational capacity. They have both had the same education and other opportunities and they both love the music and make efforts to listen to it and understand it. Now, on the view that differences in rational capacity call for differences in claims to opportunity for wellbeing, there is nothing to be regretted here. The greater wellbeing of one is simply the appropriate effect of their greater capacity. There is nothing more to be said. But this is not satisfying. Suppose we had some drug or other therapy that could enhance the capacity of the lesser listener; we might think that we ought to give it to that person and thus enhance his listening experience. We might even think that there was something unfair in not giving the capacity-enhancing drug to the lesser listener. But the idea that this makes sense should give us pause about attributing different statuses to persons with different rational capacities. To see how strange this would be, the view that differences in rational capacity ground differences in status would suggest in this case that we want to increase the status of this person. But on what grounds could we do this? If the ultimate ground of treatment is status and the traits that ground it, then there is no possible ground for this change. If we hold to the fundamental conception of status, then we should not think of status as something that we have reason to promote or enhance. It is simply something we respond to appropriately. At the very least, it is not a matter of fairness that we are required to increase the status of a being. That is the role of status that I have been working with in the discussion. So here the argument is if there is unfairness in the above example, that unfairness cannot be explained in terms of it being unfair that a being has a lower status than it could have. Status is a ground of appropriate treatment, not a form of treatment. The most natural way of thinking about this intuitive example is to think that persons with lesser rational capacities are unfairly burdened. The proper response to the situation in which it is possible to improve the rational capacities is that we should improve those capacities out of a sense of fairness. It should be noted that we do this all the time. The advance of medicine has enabled us to produce medicine that increases concentration, that gives a person an ability to read text, to see better and to hear better, and to sort out what they see and hear. But this suggests that difference of rational capacity need not imply difference of status at all. They are differences of natural endowment that ought to be rectified out of fairness to each person. Of course, someone might object that I am begging the question and presupposing that difference of rational capacity does not generate difference of
Thomas Christiano
161
status. In response, my claim here is that the above reasoning and practices concerning the desirability and fairness of increasing capacity are commonplace. I am suggesting that the commonplace reasoning presupposes the thesis that difference in rational capacity does not generate difference in status. That strikes me as an intuitive argument for that last claim. I want to say something here to vindicate the intuitive argument. An imperfectly rational being rationally desires to improve its understanding of value. And one major means by which to improve its understanding is to improve its capacity for understanding. What this discussion is meant to show is that every imperfect rational being has this same aspiration. Hence, once a being has a rational capacity, it desires the enhancement of its rational capacity. Such an aspiration is central to its being an imperfectly rational being. Because the being is reflectively rational, it strives to achieve a better understanding of value and consequently it strives to have a greater capacity to achieve a better understanding. This is meant to argue for the ideas that (1) the possession of a rational capacity of the sort I have defined is sufficient for the distinctive moral status of personhood, and (2) once a person has a rational capacity then every increase in natural rational capacity is an increase in a kind of natural resource that every rational being has reason to desire. These two theses are meant to support a third thesis: (3) differences in rational capacity do not imply differences in moral status among those who possess some rational capacity. All other things being equal, they amount to a kind of unfairness because some have greater capacity than others and so are able to advance their wellbeing more fully. One might object that the above argument for enhancement might also lead to an argument for enhancing the capacities of non-rational beings so that they become reflectively rational. But, in response, we can say that their status does not call for this kind of enhancement. I give two reasons for this. The first reason has to do with the fundamental nature of status that I have postulated in this chapter. Status is the basis for certain kinds of treatment and certain judgements of justice and fairness. It is not itself a matter of fairness that someone has one status or another. So, there cannot be a moral reason for increasing the status of a being that is grounded in that being’s status. The second reason is that even if we could enhance the capacity of a dog so that it could become a rational being, we do not owe that to the dog. To enhance the capacities of the dog in this way would be to create a new kind of being. It would so transform the nature of the thing as not to preserve identity, much like a change from non-living to living or from plant to animal would seem to undermine identity. A third related reason is that to make a rational being out
162
Equal Moral Status and the Collective Nature of Rationality
of a dog, even if that is possible, is not to advance any aim that the dog has. So, in that sense, the creation of a rational faculty would not be the creation of a new resource for the dog, since it responds to no aim of the dog.¹² In contrast, to enhance the capacities of the lesser listener would not be to create a new kind of being. It would in effect merely extend the capacities of the listener and presumably in a way that the listener does in fact rationally desire. To enhance the rational capacities of an ordinary human being so that they are greater than they were is not to turn the ordinary human being into a new type of being. The nature of the being is preserved and thus the identity of the being is preserved. Indeed, we do this more and more. We have drugs that enable people to concentrate better; we have various ways in which we make it possible for people to overcome disabilities. And thus, it makes sense to think that this kind of being comes under the same kind of principle whether it has more or less rational capacity, as long as it has the requisite kind of rational capacity. This strikes me as intuitive and plausible, but it does invoke a difficult metaphysics of essences that may prove intractable. The fundamental principle of status on this view would be that all beings that have the reflective rational capacities have an equal status. That equal status is grounded in the possession of a capacity that can be enhanced and ought to be enhanced if possible. Differences in degree above the threshold do not imply differences in status; they imply that some kind of remedy ought to be applied if possible. This solution in effect denies the relevance of differences of rational capacity to the puzzle about status. It asserts that the differences in the possession of the rational capacities do not have relevance to determining moral status beyond the achievement of the threshold.
8. A More Positive Theoretical Argument for Equality I have so far attempted to construct a kind of negative argument for equal status and an intuitive argument. The first argument tries to demonstrate that there is some reason to think that unequal rational capacities need not rule ¹² There is a puzzle that is generated internally by the view I have sketched and defended. The worry is the following. Were a perfectly rational god to exist, it seems obvious that it would have a higher moral status. But imperfectly rational beings have some reason to want to be perfectly rational. The argument I gave above suggests that the perfectly rational god must have the same status. In response to this, one might claim that imperfectly rational beings cannot aspire to be perfectly rational. This is because it would require complete transformations of their natures. Whether physicalism is true or not, the rational capacities we have are in some sense essentially tied to our physical nature. The traditional view of divine nature is that it is not material. That is what makes it possible for it to be perfect. It is outside of time and has unlimited resources. Human rational capacity is exercised in time and with very limited material resources.
Thomas Christiano
163
out equal status. The second, intuitive unfairness consideration suggests that differences in rational capacities can be unfair for reflectively rational beings so that such differences do not ground difference of status. But do we have anything stronger in support of equal status? We might think that we do not need anything stronger than this because the intuition that people have equal status is a commonplace and is very strong. As long as we dispel the air of paradox that the common view seems to suggest, we have done all we need to do to support equal status. But I think we need more. There is disagreement on the issue that needs to be engaged with.¹³ And there are important puzzle cases connected with severe disability that are not fully understood.¹⁴ I will proceed by constructing a series of arguments in defence of the moral equality of all finitely rational beings. These arguments are based on the collective nature of rationality. I am not sure if the arguments bring us to a completely egalitarian position, though they do seem to get us closer to it. My intention here is to show the promise of the appeal to the collective nature of rationality and to display where I think there is still some uncertainty about the appeal. One possible argument from the collective nature of rationality to equal status is the following. Let us call this the participation argument. A. Each person’s status is grounded in their capacity to participate in the collectively rational process. B. If their status is grounded in the capacity to participate in the collectively rational process, then their status is determined by the impersonal worth of the collectively rational process. But C. If each person’s status is grounded in the same thing, then it is grounded in something of the same impersonal worth. D. If each person’s status is grounded in a thing with the same impersonal worth, then each person must have equal status. Therefore, each person has equal moral status. This argument appears to ground equal status among persons. But there are a number of difficulties in the argument. First, as I have argued above, and will continue to argue below, it is not clear how the collective rational process can have the kind of impersonal worth that grounds moral status. Second, one might wonder here about whether each person’s worth should not be based on their contribution to the collectively rational process as well as the quality of the participation in the collectively rational process. The ¹³ See the contribution of Steinhoff (2014). ¹⁴ See Kittay (2005) and Silvers and Francis (2009). See also McMahan (2008).
164
Equal Moral Status and the Collective Nature of Rationality
worth of each person would be grounded in their contribution to and participation in the collectively rational process. And the contribution must involve the exercise of the rational capacity such as responses to others’ contributions or innovative ideas. This would seem to be a reasonable modification. After all, it is the ability of the collective process to enhance the understanding of value that is important. One might think that the individual contributors have impersonal worth in proportion to their contribution to that ability.
9. The Contribution Argument for Inequality The contribution argument for unequal status would be best put in terms of capacity to contribute to the collective process so that differential capacity to contribute to the collective process could be the basis of differential impersonal worth and so differential status. But here we arrive at the same point we started at: differential rational capacity (which now involves differential capacity to contribute to the collective process) implies differential worth. The trilemma seems to arise again from this approach. There are two important related considerations pointing to problems with the contribution argument. First, the contribution argument suffers from the concern that what I am capable of contributing is a function of many different apparently morally arbitrary facts. Two persons with the same capacities will be capable of very different contributions depending on the society they are members of and the needs of that society. This is the traditional argument for not allowing distribution to be skewed by natural talent alone. One’s moral status must not be determined by extraneous factors. Rawls hints at this argument himself in his discussion of moral status.¹⁵ The value of a person’s capacity for rationality does depend essentially on the background social order of which they are a part. A set of capacities for rationality may be extremely useful and worthwhile for one kind of society and not as useful or worthwhile for a different kind of society. This follows simply from the fact that societies have needs for different capacities and the same society may have needs for different capacities in different periods in its history. Moreover, the value of a capacity’s contribution will vary from historical moment to historical moment. So, the value of a particular capacity to contribute to the development of rationality will vary from society to society and from historical moment to historical moment. If moral status is based on capacity for contribution, then the moral status of a person will depend on ¹⁵ See John Rawls (1999: 443). For an allied discussion, see also Buchanan (1990).
Thomas Christiano
165
the society they are in or the particular historical moment they are in. And this seems to be an unacceptable result.
10. The Problem with the Contribution Argument This leads us to the arbitrary societal variation argument against the contribution argument for unequal status. And this enables us to see the egalitarian potential of the deep dependence of the development and realization of the rational capacity on society. But it is important for us to get a clear picture of the argument and why it exerts pressure on the argument for inequality. The basic observation is that the value of rational capacity depends on the social circumstances in which it is exercised. The very same rational capacities that are very valuable in one set of circumstances may not be so valuable in others. The idea is that this is an arbitrary basis for the determination of status. The one underlying rationale for this condition on our thinking about moral status is a commitment to moral individualism with regard to status (status individualism) at the base of our moral thinking. The moral status of persons and the consequent basic grounds for moral advantages and disadvantages must be found in the individual entity whose status is in question. This does not require that we ignore the deep dependence of individual rationality on society. But it does require that the qualities that ground moral status and that explain the basic moral advantages and disadvantages be ones that do not vary on the basis of external social factors. A person’s moral status cannot vary merely depending on what society they are born into. And it cannot vary depending on what basic relationship they have with the particular society they are members of. The only way that we might try to defend the contribution argument against the arbitrary societal variation problem is if we conceive of persons as parts of a larger entity that has impersonal worth and thus status. But this consideration is undercut by the conception of moral status as associated with autonomy that we discussed earlier in the chapter. The capacity for autonomy understood as a kind of self-development is one main basis for the distinctive kind of worth that is associated with the status of an entity. Living things have this capacity, animals have this capacity; rational human beings have this capacity to a greater extent than the previous two categories of entities. What do not clearly have this capacity to any high degree are collectives. They are largely the accumulated results of many people acting together in ways that do not confer autonomy on the larger group. They lack the kind
166
Equal Moral Status and the Collective Nature of Rationality
of integrity of action that living things, animals, and persons possess. To the extent that they do act together, their actions are the results of artifices that the persons who are members create. The dependence of individual rationality on the larger group does not result from the larger group having much significant autonomy. So, the collective itself does not possess the kind of impersonal worth that could give it a moral status. And the status of persons cannot be merely dependent on the impersonal worth of the collective. That said, human beings are in general political animals, as Aristotle would have it.¹⁶ That is, they seek social and political life, and their full development cannot take place without the background context of social and political life. Once we hold to the status individualistic picture, we can see why it is problematic to confer on persons different moral status depending on the contribution they can make to the community they are in. It looks like the moral status would have to depend on a lot of extraneous factors outside the person. This dependence is not merely a causal dependence. I am not thinking here only that the capacity to contribute is causally dependent on the larger society, though it is. The idea is that the value of the capacity to contribute is dependent on how the capacity fits into the division of labour of a larger society. The very same capacity has very different value if it is inserted into one society or another. To make moral status depend on that seems to make the moral status depend on a problematic extraneous factor. This does not mean that one must neglect differences in capacity among persons when organizing the society. It still makes sense for the person with the appropriate abilities for contribution (which will to some degree depend on underlying capacity) to go to the position in the division of labour for which they are best suited. It makes sense for positions of authority to go to those who have the most ability and willingness for filling them well. But these distributions of power and authority must be based on principles that assign equal basic advantages and disadvantages to all. They must be based on an appropriate realization of equality of opportunity and a proper distribution of wealth and income that is grounded in equality. Equal moral status is what makes it appropriate for persons to have equal basic moral advantages and disadvantages. One objection to my reasoning so far might be that I am importing a notion of fairness into the question of who has what status.¹⁷ The argument is that if ¹⁶ See Miller (1995) for an illuminating analysis of this idea that is consistent with the ideas developed in this paper. ¹⁷ I thank Rainer Forst for suggesting this in discussion.
Thomas Christiano
167
status is based on capacity to contribute to collective rationality, then status will depend on morally arbitrary differences among persons such as what society or historical moment they live in. It seems that the argument implies that this would be unfair so we must not make status grounded in contribution. But, the objection continues, this makes the attribution of status depend on fairness, and this would seem to get the order of grounding wrong. The problem is that principles of fairness apply to persons because they have the status of persons, while the above argument implies that we ought to select a conception of status because the alternative is unfair. This would seem to be a problem. As we pointed out above, fairness does not apply to the attribution of status; it only applies to beings with status. But I do not think that this objection works. The use of non-arbitrariness as a standard for thinking about the attribution of status does not entail that we are using a notion of fairness to adjudicate among different conceptions of status. Non-arbitrariness is a general standard that can be used in many different contexts. It is meant to exclude factors that are irrelevant to the assessment of an object. In this case, non-arbitrariness directs us to look at features of persons that can be described in an individualistic way. Status cannot vary as a consequence of features that are outside the individual. Hence, status must be based only on the capacity to participate in collectively rational activities and not on the particular uses such a capacity has in particular contexts.
11. The Argument for Equality from the Collective Character of Rationality and Status Individualism So, we have here two superficially opposed thoughts. One is that human beings’ rational natures are deeply dependent in a multitude of different ways on the social context in which they find themselves. And this dependence affects the value of the actual contributions persons with particular capacities can make in different social contexts; the very same capacities can have highly beneficial effects in some social circumstances and relatively less beneficial effects in other social contexts. Two is status individualism, which asserts that the moral status of persons must be grounded in features that they have independent of factors outside the person. This status individualism holds because, though rationality is deeply dependent on society, only individual human beings have the autonomy that can qualify them as autonomous rational beings. Society does not have that kind of autonomy.
168
Equal Moral Status and the Collective Nature of Rationality
To these two assertions we add the premise that capacity for contribution to collective rationality varies by society. The very same capacity that contributes highly to the rational project in one societal context may not contribute so much to another societal context. The consequence of these two assertions coupled with the societal variation premise is that the bare possession of rational capacity must be sufficient for moral status and that that moral status must be equal among all finitely rational beings. The basic argument here is that for all differences of rational capacity that we detect, they are based on differences in social context and so, because of status individualism, the differences cannot be bases of difference in moral status.
12. Objection from Extreme Disability to the Argument for Equal Status But there is a more serious objection, which is that this societal variation argument for equality of status may not take us all the way to equality of status for persons. The reason is that there may be some persons whose capacities are such that they are less valuable in any society. At the very least, this is a conceptual possibility. Some severely disabled but still rational persons may be less suited for any role in any society than other rational persons. Hence, the lesser value of their capacity is not merely the result of external extraneous circumstances. One might wonder whether such an extreme difference in capacity to contribute among persons might not suggest an argument for lower moral status. Let us call this the extreme disability argument. Just to be clear, I am here discussing severely disabled rational beings.¹⁸ These are beings that have some rational capacity though it is limited in important ways. I am not speaking here of anencephalic children or other beings who have no rational capacity. The status of these beings will be like that of animals or plants. They may have a derived status as a result of the fact that they are in relationships with persons, but they do not have the fundamental status of persons that is grounded in rational nature. It is important here to remind ourselves of the distinction between civil and political status and moral status. Civil and political status is the basis of rights ¹⁸ See Kittay (2005). I think of her examples of congenitally severely mentally retarded humans as persons on this standard since they have aesthetic appreciation and moral responses. Admittedly there is uncertainty here since we do not have a complete account of the inner worlds of these people, but careful loving attention seems to detect significant elements of rational nature here.
Thomas Christiano
169
to participate in the economic, civil society, and political institutions of the society. Those rights may be lower if there is a clear sense in which a person is not able to exercise those rights in a way that advances their interests or moral concerns. Those rights are also not rights that attach to persons qua persons, but attach rather to persons at various stages in their lives. Children have lesser such rights for example. But children do not have inferior moral status to adults. Their civil and political status is determined in a way that reflects underlying equal concern for their legitimate interests and their equal dignity. There are two ways to respond to the objection that invokes more extreme disability. One, we can argue that even the most severely disabled rational persons could make a superior contribution in some social contexts. Here we could extend the society variation argument by including many different possible societies in the space of societies to which one can make a contribution. Even if a particular person is not able to make a contribution to the societies of which we are aware or any similar possible societies, there may be more distant possible worlds in which these persons could make a similar or greater contribution than other currently not so disabled persons. The extreme disability presumably does not hold in all remote possible worlds. If we then reaffirm the premise that status cannot depend on extraneous factors such as which possible society a person inhabits, we can invoke the individualistic standard to support equality of status for those who are severely disabled rational persons. The second response to the objection could reject the contribution conception of the value altogether. After all, we have good reason to think that contribution cannot be the only dimension of importance in the value of a person’s participation in the collective rational activity. How well one is able to reflect that collective activity should matter as well. For example, if one were to contribute somehow to the collective rational activity but were unable to appreciate its products, that might also be a serious mark against the quality of one’s participation in the collective rational activity. I don’t think it makes sense to abandon contribution altogether, though it does make sense to suppose that a richer conception of the value of one’s participation is better than merely holding to the idea that contribution alone is of value. Hence variations in the extent to which persons make contributions cannot be altogether neglected as a possible source of difference in status. The first response to the extreme disability objection suggests that the societal variation argument be extended from the societies that we have seen over
170
Equal Moral Status and the Collective Nature of Rationality
the last 100,000 years or so to those that plausibly could have existed in this period (given psychological and social possibilities) to much more remotely possible societies that are not plausible ones given basic facts of psychology, sociology, economics, and politics. The societal variation argument initially permits us to take the whole set of plausible societies given these facts and tells us that if persons are more capable in one kind of society than other people are but less capable in another society than those other people are, then we should not invoke value of contribution to assign different status. If we expand the set of societies from the socially possible to the psychologically possible or even the physically possible, then we will have a much more egalitarian argument because even the most severely disabled will be able to make greater contributions in some possible societies. It is important to be clear here about the possible world argument. The idea is not that this being in another possible world could have capacities that make a greater contribution. That would be to assert that the being might have the potential to make a greater contribution. The societal variation argument in its possible world variant rigidly designates the capacity of the being and then asks whether a being with that capacity could make a high contribution, perhaps with some enhancements but not such as to create a new mechanism or capacity. I do not have a clear sense of whether this response to the extreme disability argument works. In favour of the argument is the idea that moral status is a deep fact about morality and the possession of moral status is a deep fact about the being that possesses it. Thus, the propositions that characterize it must have a deep kind of necessity attached to them. That suggests that attributions of moral status ought to be invariant across many possible worlds, maybe even ones that are not psychologically possible. But it is not evident that the necessity of attributions of status must be that deep. They are certainly not logical truths. But how we should further limit the set of possible societies in which the variability of contribution is relevant to assessing moral status is not clear to me at this time. So here I think we do meet a limit to the positive argument I have been proposing for equal status. This is not to say that we are forced to accept unequal status even if we do accept that society variation must be evaluated relative to psychological and social facts. We are not forced to accept equal status even then, because we still have the idea that increases in rational capacity can be looked at as increases of resources. But we do fall short of a complete positive argument for equality here. We may have to look elsewhere for such a positive argument, but the jury is still out on the completeness
Thomas Christiano
171
of the societal variation argument for equality. But I cannot pursue that argument further here.
13. Conclusion In conclusion, why does difference in capacity suddenly become less normatively powerful above the threshold identified by the onset of rational capacity? First, a change from below the threshold to above the threshold involves some kind of substantial transformation of the nature of the being involved while changes above the threshold do not involve such substantial transformations. To give a chimpanzee the kinds of rational capacities that nearly all humans have is fundamentally to transform the being into a new kind of being. It is not to improve the chimpanzee; it involves replacing the chimpanzee with a different kind of being. The change is not identitypreserving. Thus, one cannot owe it to the chimpanzee to enhance the capacities in this way. Second, we can say it makes sense to enhance the rational capacities of rational beings even though it is not owed to non-rational beings that they be turned into rational beings. The basic argument is that it does seem fitting to enhance the rational capacities of a rational being. This must be the rational desire of any rational being, and it seems to advance the fundamental concerns of rational beings. On the other hand, it does not seem a fitting response to a non-rational sentient being to transform it into a rational being. It is not within the realm of concerns of the sentient being to become rational. That is simply beyond its nature. This might be a good thing to do in some circumstances if it were possible, but this is not because it is a fitting response to the nature of the sentient being. Third, the equality of status for the vast majority of persons can be positively argued for on the grounds that moral status ought not to be a function of factors extraneous to the person. Value of contribution is, however, dependent on the society and the historical moment in which the person exists. To make status depend on value of contribution would be to violate the individualistic constraint on assignment of moral status. I have attempted to respond to the internal difficulty of trying to base equal moral status on human rationality. I have articulated a conception of equal basic moral status and a conception of human rationality. I have argued that despite appearances, there is no reason to think that the internal tension requires us to reject equal moral status. And I have tried to show that there
172
Equal Moral Status and the Collective Nature of Rationality
is positive reason for believing that there is equal moral status. That positive reason, however, may not take us all the way to equality for all rational beings. There is, unsurprisingly, more work to do.¹⁹
References Arneson, Richard. 1999. ‘What, If Anything, Renders All Humans Morally Equal?’. In Peter Singer and His Critics, edited by Dale Jamieson, 103–128. Oxford: Blackwell. Buchanan, Allen. 1990. ‘Justice as Reciprocity versus Subject-Centered Justice’. Philosophy and Public Affairs, 9/3: 227–252. Christiano, Thomas. 2014. ‘Rationality, Equal Status, and Egalitarianism’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 53–75. Oxford: Oxford University Press. Hobbes, Thomas. 2000. Leviathan, edited by Edwin Curley. Indianapolis: Hackett Publishers. Hutchins, E. 1995. Cognition in the Wild. Cambridge, MA; MIT Press. Kagan, Shelly. 2019. How to Count Animals, More or Less. Oxford: Oxford University Press. Kittay, Eva Feder. 2005. ‘At the Margins of Moral Personhood’. Ethics, 116/1: 100. McMahan, Jeff. 2008. ‘Challenges to Human Equality’. Journal of Ethics, 12/1: 81–104. Mill, John Stuart. 1989. On Liberty, edited by Stephan Collini. Cambridge: Cambridge University Press. Miller Jr., Fred D. 1995. Nature, Justice, and Rights in Aristotle’s Politics. Oxford: Oxford University Press. Montaigne, Michel de. 1980. An Apology for Raymond Sebond. Harmondsworth: Penguin Books. Muldoon, Ryan. 2013. ‘Diversity and the Cognitive Division of Labor’. Philosophy Compass, 8/2: 117. Nefsky, Julia. 2017. ‘How You Can Help, Without Making a Difference’. Philosophical Studies, 174: 2743. Rawls, John. 1999. A Theory of Justice, rev. ed. Cambridge: Harvard University Press. Sher, George. 2014. ‘Why We Are Moral Equals’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 17–29. Oxford: Oxford University Press. ¹⁹ I thank Andrew Williams, Pablo Gilabert, Ryan Pevnick, Matthew Palynchuk, Richard Arneson, Giacomo Floris, Nikolas Kirby, Arash Abizadeh, Nandi Theunissen, and participants in a conference on equal status organized by Giacomo Floris and Ian Carter in the summer of 2020, participants in sessions at the APSA meetings in Montreal in September 2022, the APA meetings in Montreal in 2023, and a conference on Human Dignity at Concordia University in April 2023 for valuable comments and discussion.
Thomas Christiano
173
Silvers, Anita, and Francis, Leslie. 2009. ‘Thinking about the Good: Reconfiguring Liberal Metaphysics (or not) for People with Cognitive Disabilities’. Metaphilosophy, 40: 3–4. Steinhoff, Uwe. 2014. ‘Against Equal Respect and Concern, Equal Rights and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 142–172. Oxford: Oxford University Press. Waldron, Jeremy. 2016. One Another’s Equals. Princeton: Princeton University Press.
7 Basic Equality Worth, Luck, and Weight Nikolas Kirby
1. Introduction Can we explain why human beings, who are unequals in almost every nonnormative property, at least by matter of degree, are yet still one another’s ‘equals’ in some basic moral sense? It is indeed a widely held assumption that (at least, almost) all human beings are one another’s equals in some basic moral sense. Furthermore, this assumption is fundamental to much of contemporary moral, political, and legal theory.¹ However, presumably, our basic moral equality must be grounded on each human’s possessing some non-normative property, that is, some property that explains why we (and, perhaps, not other beings) are one another’s moral equals. Yet, as a matter of empirical fact, the most plausible candidate non-normative properties for this role come in different degrees across the human population. So the following problem arises: if having some of a non-normative property (like ‘rationality’) is so important that it grounds having some of a basic moral property (like ‘intrinsic value’ or being due ‘respect’), then why does having more of the non-normative property not also ground having more of the basic moral property? Here lies not simply a gap in our explanation as to why we are equals. Here lies a ‘pressure of reason’ to conclude the opposite: we are, if anything, one another’s unequals (Arneson 2014: 36). A widely accepted response to this problem has eluded philosophers: leading some to declare the problem insoluble (Arneson 2014; Husi 2017), others to reject basic equality (Steinhoff 2014), and others to back to theology (Waldron 2017: 185). In this chapter, I offer a novel solution by leveraging three key concepts: moral worth, moral luck, and moral weight.
¹ On the ‘egalitarian plateau’ see Dworkin (2000: 11); Kymlicka (2002: 4). Nikolas Kirby, Basic Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0008
176
Basic Equality
First, after clarifying the problem of basic equality (Section 2), I start my argument (Section 3) with the assumption that, in virtue of possessing some degree of rational agency as a non-normative property, (almost) all human beings, unlike other animals, are able to be morally responsible for their actions to some degree of moral credit. However, it is a mistake to equate such degrees of credit with the moral worth of one’s actions. Instead, such moral worth is equivalent to the actual value of such credit attained relative to the possible values of credit available to that individual. In short, our moral worth is always our ‘choice’ assessed relative to ‘circumstance’, and these circumstances include any unchosen differences in our degree of rational agency. Once such differences in circumstance, therefore, are accounted for then, it turns out that each individual has the equal ability to attain moral worth. We are equals in at least this one moral property. This putative solution to the problem of basic equality then raises two questions. First, why should we accept such claims about moral worth to be true? My answer is that they are the most plausible way to avoid the problem of circumstantial moral luck (Section 4). Second, assuming that these claims are indeed true, what follows? My primary answer is that, ceteris paribus, it is only fitting that each individual’s response to the challenge of attaining moral worth has ‘equal weight’ in the world; only in this way will how they lead their life matter as much as any other (Section 5). I end by addressing four objections regarding discontinuous options, determinism, lack of alternative options, and the exclusion of humans without agency (Section 6).
2. The Problem of Basic Equality Philosophers articulate our ‘basic equality’ using a diversity of formulations. They speak, inter alia, of equal ‘worth’, ‘status’, ‘intrinsic value’, ‘concern’, ‘basic rights’, and/or ‘respect’. I shall, however, simply address the following proposition: Basic Equality: All (or nearly all) humans are one anotherʼs equals in their possession of some basic moral property.2
What is a ‘basic moral property’? The term is a functional placeholder for any property that will do the kind of work we expect of a concept of ‘basic equality’. Thus, a property is ‘moral’ if, depending upon one’s most primitive ² For reasons why, see Kirby, 2024.
Nikolas Kirby
177
moral concepts, the property is: having some (degree of ) moral value; and/or being the subject or object of a moral ought proposition (of some degree of binding force or lexical rank); and/or being the subject or object of a moral reason for action (of some degree of weight or lexical rank). Such a moral property is ‘basic’ if it, roughly speaking, satisfies the following set of conditions. The property is generally natural, that is to say, its possession arises as a consequence of the normal course of environmental processes, such as birth, growth, and/or maturity. The property is generally inalienable, that is, it is not liable to be easily extinguished or transferred by its possessor to another, without the possessor themselves ceasing to exist. The property and its implications, generally, have great weight or lexical priority: we rarely (if ever) have reason, all things considered, to act in any way inconsistent with the value, ought proposition, and/or reasons for action constituting the basic moral property. Thus, collectively, these three conditions mean that, when articulating a moral theory, the distribution of the property (equal or unequal) across a set of individuals will be theoretically foundational, or very close to foundational. This is to say that other major claims within a theory will be implied, or at least conditioned, by the fact of that distribution. The advantage of Basic Equality—qua proposition—is that it is neutral at the outset between various concept(ion)s of basic equality. However, it also allows us to highlight that the problem of basic equality is similarly neutral. This problem starts with the assumption that the fact of any human’s possession of any basic moral property must be grounded in her possession of a non-normative property, that is, its ‘basis’ (Rawls 1999: 441). However, on the one hand, the most plausible candidate non-normative properties for such grounding are scalar: rationality; sentience; intelligence; empathy; agency; a sense of justice; a good will; even ‘conceptions of the good’ can be more or less complete (Schaar 1964: 867). And, further, as a matter of empirical fact, human beings do possess such properties to varying degrees (Spiegelberg 1944: 106; Williams 1973). On the other hand, there is what Richard Arneson calls a ‘pressure of reason’ to presume that, if a scalar nonnormative property grounds a scalar morally basic property, then ceteris paribus a greater (or lesser) degree of that non-normative property must surely proportionally ground a greater (or lesser) degree of the morally basic property (Arneson 2014: 36). As Louis Pojman put it, ‘If P constitutes human worth, then it would seem that the more of P that a person has, the better he or she is. … If reason is really all that makes us valuable, then the more of it the better … If our ability to will the good is what gives us value, then it would seem that some people are more valuable than others because they
178
Basic Equality
have greater ability to will the good than others’ (Pojman 1991: 484–485; see also Carter 2011: 541; Christiano 2014: 56). This presumption is defeasible. Hence, the ‘ceteris paribus’. But it does mean that the onus of proof is shifted onto any proponent of Basic Equality. The presumption is indeed in the contrary direction: No Basic Equality: All (or nearly all) humans are not one anotherʼs equals in their possession of any morally basic property.
Of course, a number of replies to this argument against Basic Equality have been proposed. One might reject the starting assumption that the possession of a morally basic property needs to be grounded in any non-normative fact about humans: it is just an ungrounded fact (Gosepath 2014: 125; but see Husi 2017: 388). Or, insofar as it is grounded, it is grounded in our ‘humanity’, and whatever then grounds the fact that we are human (Frankena 1962; Vlastos 1984; but see Schaar 1964: 875–882; Wilson 1966: 93). Or, one might argue that our equal possession of a particular morally basic property is not even a fact about humans, but rather it is a proposition that we each (should?) choose to assume about humans when engaging with them (Arendt 2006; Macdonald 1946; but see Waldron 2017: 55–61). Or, one might accept the starting assumption of the argument above, but argue that the non-normative may be theological (Waldron 2017: 175–214; but see Thomas 1979: 539–540) or transcendental (Kant 1996; but see Williams 1973: 230–249), thus voiding any presumption that it is likely to be spread unevenly amongst us. Or, one might argue that it follows as a formal principle of rule application (Frankfurt 2015; Lucas 1965; Westen 2014; but see Waldron 2017, 66–83). Or, one might aim to identify a non-scalar property to ground our basic equality, defined in terms of a scalar property (Carter 2011; Christiano 2014: 62; Parr and Slavny 2019: 6–8; Sher 2017: 36–37; Rawls 1999; but see Arneson 2014: 40–48; Husi 2017: 390–400; Kirby, 2024). Or, one might look to a universal prohibition against treating others as inferiors (Sangiovanni 2017; but see Floris 2020). Or, one might even concede that Basic Equality may be false but hold that some (many?) moral arguments need not actually rely upon it to justify many forms of so-called egalitarian rights (Husi 2017: 381ff; Steinhoff 2014; but see Waldron 2017: Ch. 1). I shall not rehearse these extant arguments here. I take them to already meet convincing counters supplied in the literature, cited in the respective references above. I simply proceed to my own novel argument in the next section. However, one final preliminary note, before doing so: the ‘pressure of reason’ that creates the problem of Basic Equality is sometimes thought
Nikolas Kirby
179
to create a cognate problem with respect to grounding moral inequality, or more accurately, the moral distinction between human beings and other animals. The claim is that not merely will any candidate non-normative property (such as ‘rationality’) vary across the human population, it will also include at the lower levels various other animals (such as higher primates). Many, therefore, take the fundamental problem of basic equality to be one of justifying a non-arbitrary ‘threshold’ level on such a non-normative property (for example, a certain level of rationality) that grounds equality in a basic moral property for any being possessing that property above the threshold level (or within a range), but not below (see Arneson 2014: 34ff; Waldron 2017: 119ff; Christiano 2014: 56ff ). I suggest, however, that this is a distraction from the real ‘pressure of reason’ problem that I have detailed above. This is because we can easily justify a discontinuity between (at least almost all) human beings and other animals,³ by simply narrowing our candidate bases for solving Basic Equality to only non-normative properties that (at least almost all) human beings possess to some (any) degree, and other animals have to no degree. We might call this a ‘bare property’—that is, possessing another property to some (any) degree at all (see Kirby, 2024). Indeed, that is my assumption about my own putative basis: the bare property of having some (any) rational agency at all. But it is the move also made, I suggest, by many other theorists, whether they quite understand themselves to be doing so or not. They speak of our ‘capacity to reflect’ (Christiano 2014: 62); our ‘subjectivity’ (Sher 2017: 36–37); ‘having a conception of the good’ (Parr and Slavny 2019: 6–8)—all properties that plausibly, under a relevant conception, (at least, almost all) human beings have to some (any) degree, and other animals have to no degree at all. With such a non-normative property in hand—uniquely possessed in the human range alone—one does not need to identify any particular threshold level on such a property to ground our equal possession of a basic moral property. Indeed, one should simply look for an argument to justify why any level will do. Yet, that is precisely where the pressure of reason kicks in: if having a particular non-normative property grounds having some basic moral property, then why does having more of the former not ground more of the latter? How can the most marginal amount of that non-normative property
³ Under almost all plausible candidate properties, empirically some human beings will lack the property. For my own response, see Section 6.4. Further, depending upon the candidate property, some other animals may instantiate the property. This may not be problem, but instead a productive result. For example, if it turns out as a matter of empirical fact that some other animals (say, higher primates) do indeed hold some degree of rational agency (as I define it), then I would happily embrace their equality too in the consequently grounded basic moral property.
180
Basic Equality
in one human being ground the same moral result as having an incredibly high amount? That is the real problem of basic equality. It is true that the possession of any such non-normative properties that start ‘in the human range’ are likely, themselves, to be grounded upon us surpassing a certain threshold level of underlying non-normative grounding propert(ies), and these non-normative properties are possessed to varying degrees across both humans and animals. For example, I fully accept that my proposed non-normative property, rational agency, as defined below, will be grounded upon our reaching some threshold level of a complex of, inter alia, sensory, perceptual, and rational faculties, that other animals also have to a lower degree. However, the claim that reaching a threshold on such an underlying non-normative scalar property triggers having some (any) degree of another non-normative scalar property is not at all an uncommon grounding relation, nor presumptively arbitrary. Take, for example, the rising water level of a river and the property of being flooded. The water rises to a threshold level—the riverbank—and then flooding begins, and indeed continues to increase as a function of the immediate explanatory variable—the water level. Of course, whatever reasons for action are then triggered by some (any) flooding, we should expect will increase as a function of higher flooding levels. It is precisely this latter type of intuitive inference that we are seeking to counteract in defending Basic Equality.
3. Moral Worth The relevant non-normative property that I aim to ground our basic equality upon is, as foreshadowed, having some (any) degree of rational agency (simply ‘agency’ hereafter). For the purposes of illustration, I shall assume that agency is something like: the capacity to conceive of reasons to prefer certain states of affairs over other states of affairs; to form a judgement about what there is most reason to do given such reasons; and to be successfully motivated to act (or not) by such reasons.⁴ However, ultimately, the particular conception of agency preferred does not matter, so long as it is consistent with the next three claims. The first claim is: Human Agency: All (or nearly all) human beings have degrees of agency, a, across their lifetime, where a > 0. ⁴ There is a long tradition that aims to draw a line of moral distinction between humans and other animals in cognate fashion, even if this tradition has not always interrogated so rigorously our basic equality above this line: see (Korsgaard 2006).
Nikolas Kirby
181
The important concession is that agency comes as a scalar property. It comes in different degrees as a result of the different properties upon which it, itself, is grounded: some human beings can conceive reasons more easily than others; some may be able to form more consistent, more rational, more accurate judgements than others; and some may be more motivated by such reasons than others. Furthermore, the expectation is that such capacities may change over a lifetime. An individual may hold different degrees of agency at different points in their lifetime. They may even, at various points, lack such agency altogether. However, so long as a human being has some degree of agency, a, where a > 0, at some point in their lifetime, then they can be taken to fall within the scope of Human Agency (for those who might not, see Section 6.4). The second claim is that having a degree of agency, for any individual, in any scenario, will upon exercise (partially) ground another property, that is, a degree of moral responsibility for the resulting state of affairs. We might label this property, for stylistic reasons that will become apparent below, the individual’s ‘choice’: Choice: Xʼs degree of agency, a, in any scenario, S, grounds upon exercise Xʼs ʻchoiceʼ, that is, having moral responsibility for a resulting state of affairs, R of scope m with degree of credit c.
This claim assumes, following Michael Zimmerman, a definition of ‘moral responsibility’ that distinguishes between its scope and its degree (Zimmerman 2002: 560). The ‘scope’ of X’s moral responsibility for a resulting state of affairs R is the extent of the resulting state of affairs that is attributable to X. This is to say that it is appropriate to take these particular aspects of the resulting state of affairs as a basis for the moral appraisal of X’s choice. The ‘degree’ of X’s moral responsibility is the particular moral appraisal of that choice. We often speak of such appraisal in non-scalar terms—paradigmatically, praise or blame—but I shall assume that it can also be spoken of in scalar terms so that we can put a cardinal value on the term. I shall speak of degrees of ‘credit’ (which includes a negative range of degrees of discredit).⁵ The third claim is that having a degree of agency in any scenario will also (partially) ground a ‘domain of moral responsibility’, that is, a set of options that the individual could be morally responsible for, depending upon how they exercise their agency. We might label this property the individual’s ‘circumstance’: ⁵ Of course, this is effectively what others have termed the ‘moral worth’ of our actions (Arpaly 2002; Beardsley 1957; Markovits 2010; Smith 1991). But, as will become clear below, I reserve the latter term for another purpose.
182
Basic Equality
Circumstance: Xʼs degree of agency, a, in any scenario, S, grounds Xʼs ʻcircumstanceʼ, that is, having an option set, O, of possible resulting states of affairs, (R1 … RN ), each with scope (m1 … m N ), with degree of credit (c1 … c N ), for which X could be morally responsible depending upon how they exercise their degree of agency, a.6
This claim assumes that, in some sense, agents generally do have multiple options available to them when acting. Many think this necessary for being morally responsible for any resulting state of affairs (see Van Inwagen 1983), and I shall continue to articulate this model of calculating moral worth under that assumption. However, of course, following Harry Frankfurt, others do not share this assumption (Frankfurt 1969). This is not fatal to my argument (see Section 6.3). It just requires an appropriate variation, mutatis mutandis. It is important to note that whilst, on the one hand, for each individual some degree of agency will ground having some ‘set of options’ for which an individual may be morally responsible, on the other hand their particular degree of agency is also liable to determine, in part, what particular options they have within that set. For example, Sue’s rational faculties may be such that, in a particular scenario, her choice to hit John will not be very blameworthy, if she hits him. The option ‘hit John in a very blameworthy manner’ does not lie in Sue’s option set because of her particular degree of agency. By contrast, Ann’s rational faculties may be such that, in otherwise exactly the same scenario, hitting John would be very blameworthy. The option to ‘hit John in a very blameworthy manner’ does lie in her option set because of her particular degree of agency. In this way, one’s particular degree of agency will, in part, ground one’s ‘circumstances’, even if one’s exercise of that particular degree of agency will then ground one’s ‘choice’, given those circumstances. My first three claims establish a ‘dual grounding’ structure. The first claim grounds both the second and third: Human Agency: All (or nearly all) human beings have degrees of agency, a, across their lifetimes, where a > 0, grounding in any scenario, S, upon exercise both: Choice: having moral responsibility for a resulting state of affairs, R of scope m with degree of credit c; and ⁶ The terms ‘choice’ and ‘circumstance’ deliberately recall Ronald Dworkin’s distinction in the domain of distributive justice: Dworkin 2000. However, here ‘choice’ is defined in terms of moral rather consequential responsibility. And, ‘circumstance’ is defined by what one might possibly be morally responsible for in any scenario, rather than what one is not consequentially responsible for.
Nikolas Kirby
183
Circumstance: having an option set, O, of possible resulting states of affairs, (R1 … RN ), each with scope (m1 … M N ), with degree of credit (c1 … c N ), for which X could be morally responsible depending upon how they exercise their degree of agency, a.
This grounding structure explicitly contemplates that different degrees of each human’s agency, a, will both (partially) ground choices of different degrees of actually attained credit, c, and also circumstances offering options with different degrees of credit, (c1 … c N ). In short, we expect that scalar differences in agency amongst human beings will ground scalar differences in both the credit of their actual choices and credit available to them as they make those choices. However, the rest of the argument aims to explain how it is possible that such scalar differences amongst these latter two properties—choice and circumstance—can interact to generate a persisting equality in another property among such individuals. First, let us assume that in any particular scenario, within the option set available to an individual, degrees of credit c1 … cN are ordered from lowest to highest: with c1 being the minimum and cN the maximum, although the absolute value of each may change between scenarios. Given this ordering, any such scenario might be understood to set a kind of ‘challenge’, that is, to attain the highest degree of moral credit, c, available given the options (that is, by stipulation cN ), and avoid the lowest degree of moral credit, c, (that is, by stipulation c1 ). With this assumption in place, we can represent what I shall call X’s degree of attained ‘moral worth’ as a function of the degree of credit, cX , for the choice that X actually takes, given the highest degree of credit available c N and controlling for the lowest degree of credit available c1 in her circumstances: Worth: In any scenario, S, given Xʼs degree of agency a, option set O, and resulting state of affairs R such that X has: Worth, w =
cx – c1 cN – c1
Thus, in each and every scenario where X exercises agency, worth, w, can be represented by a value between 0 and 1. If X chooses the worst available option, then the numerator will be c1 – c1 = 0, and thus w = 0. If X chooses the best available option, then both the numerator and denominator will be cN – c1 , and thus w = 1. Across their life a person will face a constant stream of different scenarios where they exercise agency. Thus, we might assume them to offer up, to adapt Ronald Dworkin’s phrasing, a ‘life performance’, which in some sense is the
184
Basic Equality
‘sum’ of the performances in each individual scenario.⁷ One interpretation might simply be to sum the values of worth across each scenario. However, this would give the same weight to each scenario in a person’s life regardless of the stakes, and also assign greater worth simply to anyone for living through more scenarios. A better interpretation, instead, is to see one’s overall worth, across any number of scenarios, L in life as a ratio of the sum of each value cX – c1 for each scenario over the sum of each value cN – c1 for each scenario. Life Performance: In any life, X will face a number of agential scenarios L, given Xʼs degree of agency a, option sets OL , and resulting state of affairs RX , such that X accumulates: Worth, w =
ԕcXa – c1a ԡ + ԕcXb – c1b ԡ + ... + ԕcXL – c1L ԡ ԕcNa – c1a ԡ + ԕcNb – c1b ԡ + ... + ԕcNL – c1L ԡ
In this way, one’s degree of worth at any point in one’s life will always be representable by a value between 0 and 1, no matter how many scenarios, but scenarios where the stakes are higher will matter more than other scenarios in one’s life. By terming this concept, so described, an individual’s ‘moral worth’, I mean to imply that it is the proper basis for appraising an individual’s overall moral performance in their life, that is, to say how well they have performed morally. It forms the basis upon which we might say that an individual agent, rather than just any particular choice of theirs, is worthy of praise and esteem; or disdain and contempt. It is the currency of what Zimmerman calls that individual’s ultimate ‘moral ledger’ (Zimmerman 1988: 38).⁸ Overall, this model of calculating such worth is meant to capture the idea that each individual’s whole life can be considered as a kind of moral ‘challenge’. That challenge is to attain the highest degree of overall credit for one’s choices, given their circumstances—their ‘options sets’. In other words, each individual is set the challenge of attaining the highest degree of worth. Now, at the end of my argument I shall also add one final axiological claim, which is the least plausible of all my claims. It is not necessary for my ⁷ See Dworkin 2000: Ch 7. However, unlike Dworkin I hold that one’s performance value, qua ‘worth’, controls for circumstance (as I have defined it). ⁸ To clarify: credit grounds the praiseworthiness (or blameworthiness) of particular choices on my view, and worth grounds the praiseworthiness (or blameworthiness) of the individual. Given the analytic relationship I have set out between the two concepts, changes in degrees in both will be directly correlated, but not in any one-to-one manner. In particular, an individual whose life includes no options to perform very praiseworthy actions (high credit choices) can still be a very praiseworthy individual if they have chosen the best options available to them (high worth life performance). The distinction is akin to that between two kinds of praiseworthiness (and blameworthiness) drawn by Elizabeth Lane Beardsley, although I am inverting her terms (Beardsley 1957). I do this because it seems to me most intuitive to reserve the term ‘moral worth’ for what constitutes, at least to my mind, the more important assessment of the moral performance of the individual.
Nikolas Kirby
185
argument, as I establish below (see Section 6.1), but it will aid our initial conceptualization. This claim is that between c1 and cN in any scenario there lies an infinite set of options reflecting continual gradation of degrees between the higher and lower bounds: Continuous Options: For any option set O, of possible resulting states of affairs (R1 … RN ), each with scope (m1 … mN ), with degree of discredit (c1 … cN ), there is a state of affairs with the lowest value c1 , another with the highest value cN , and other states of affairs for all values of c in between.
The importance of this assumption is that for any individual, across any life, including any option sets, they not merely can all achieve the highest and lowest levels of worth, but also any level in between. Now, bringing all these claims together we get a kind of equality in a moral property for all individuals that have some (any) degree of agency. Given that an individual’s aggregate level of worth, for any level of aggregate credit, is discounted relative to that individual’s circumstances, then it holds that differences in one’s circumstances will not affect the level of worth attainable by any agent. Any individual is able to achieve any level of worth (between 0 and 1), independent of their circumstances, that is, entirely independent of the range of degrees of credit available to them. And most importantly, for the sake of basic equality, this ability will not be different for any set of individuals because they have different degrees of agency. Any aspect of one’s agency that affects one’s circumstances (that is, the degrees of credit available to the agent) is controlled for just like any other factor that might affect one’s circumstances. In conclusion, each individual with a degree of agency, regardless of the particular degree of agency, will have the same (equal) ability to lead a worthy, or indeed unworthy, life so calculated. This is to say that they have the same (equal) range of worth values available to them. In this way, no matter the proportions of one’s life—the greater or lesser one’s circumstances—what we might think truly matters is equally available: that is, to live a worthy life relative to circumstance.
4. Moral Luck If the claims constituting the ‘model’ of moral worth above are true, then we would be equals in at least one moral property. My argument for its functionally defined ‘basicness’ qua ‘basic moral property’ (as per Section 2)
186
Basic Equality
follows in the next section. But here, I simply ask, why should one accept such claims about moral worth to be true? In particular, what argument can we offer that does not itself depend upon intuitions about our basic equality? The independent reason I offer is that this model of moral worth is the best explanation for the impossibility of circumstantial moral luck. What is, putatively, circumstantial moral luck? ‘Moral luck’, in general, would arise if the action of an agent were properly attributed a degree of moral worth, although the relevant aspect of that action that grounded such moral worth was, in some way, beyond that agent’s control (Williams 1981; Nagel 1979). ‘Circumstantial moral luck’, as a sub-species, would arise if the moral worth of one’s action were to depend in some way upon one’s circumstances, yet such circumstances were not fully within their control (Nagel 1979: 33–34). The standard test case is the life of a person who immigrates from Germany to Argentina in the early 1930s for business reasons before the Nazi regime comes to power. They end up living a reasonably praiseworthy life in Argentina, yet counterfactually if they had stayed in Germany they would have taken up the opportunity to be a part of the regime (Nagel 1979: 34). It is, ex hypothesi, only by sheer luck of circumstance that the individual ends up leading a reasonably praiseworthy life, rather than a truly despicable one. Such a case is a ‘test case’ because it is meant to test our intuition that, despite appearances, circumstantial moral luck—like all moral luck—is impossible. As Bernard Williams says, ‘when I first introduced the expression moral luck, I expected to suggest an oxymoron’ (Williams 1993: 251). Why this expectation? The reason is internal to the very concept of ‘moral worth’ itself. By definition, moral worth is a degree of moral appraisal we are due by virtue of what we are responsible for across our lives, that is, what is attributable to us. Yet, if one’s putative moral worth depends upon factors that one is not responsible for, then it cannot to that extent be one’s ‘moral worth’ qua only a moral appraisal of what one is responsible for. Moral luck seems to create a contradiction in the very concept of moral worth. My argument, however, is that the problem of circumstantial moral luck only arises if one fails to make the distinction between moral credit and moral worth as I have done above. If, by contrast, we make this distinction and conceive of moral worth, so defined, then the problem is solved. This is a strong
Nikolas Kirby
187
independent reason to accept my account of moral worth, and thus, a fortiori, my account of our equal ability to attain moral worth.⁹ Let me illustrate. To begin: as a simplified formalization, the (apparent) problem of moral luck might be construed as follows. On the one hand, an individual has their actual circumstances with a range of possible options available, all with credit values ranging (5c … 10c), and within that range, let us say they choose a moderately creditable option c8 . Let us call these ‘Argentinian circumstances’. On the other hand, in an alternative scenario, the individual has a different range of possible options available, all with credit values ranging (1c … 10c). In this latter alternative scenario, the individual would have chosen the most discreditable option, 1c. Let us call these ‘Nazi circumstances’. If we equate the individual’s resulting level of credit, 8c, with their moral worth, w, then we have circumstantial moral luck. This is because in part, they only end up with this level of moral worth due to factors beyond their control, that is, because 1c was not available to them under the alternative circumstances. However, if our alternative conception of moral worth, detailed above, is presumed, then regardless of circumstance, that is, regardless of the particular range of credit options available to them, the individual will always have the same range of worth options available to them, that is, in Argentinian circumstances (0 … 1), and in Nazi Circumstances (0 … 1). It is true that in Argentinian circumstances the individual will not be able to choose anything so discreditable as the option 1c—available only in Nazi circumstances. But, in terms of worth, the option associated with 1c will have the same value as the most discreditable option in the former circumstances, 5c, such that w = 0 for both. If this is the case, then there is no problem of circumstantial moral luck, because regardless of the factors beyond the control of the agent, each agent’s level of moral worth across their life will always lie within their control. They
⁹ Of course, I am assuming here to some extent that one finds the other possible responses to circumstantial moral luck unpersuasive, or at least far less attractive. In particular, I don’t think this is a hard argument to make against what seems to be the dominant ‘modal solution’ proposed by Zimmerman himself (Zimmerman 2002; see also Peels 2015; Richards 1986). In short, the solution holds that our moral worth will be a function not merely of what we do in this actual life, but of what we would have done in all possible other lives. But this solution is so unattractive because, in effect, we must revise our concept of ‘moral worth’ such that we are morally responsible for acts we never actually perform (Hanna 2014). Indeed, Zimmerman acknowledges upfront that such a proposition is prima facie mystifying, or, indeed, laughable (Zimmerman 2002: 553). It is only, I suggest, the fact that he seems to be offering the sole plausible alternative to circumstantial moral luck that means we should take this proposition seriously. If my argument is correct, however, then he loses this dialectical advantage.
188
Basic Equality
have the same range of worth values available to them, even if not credit values. Hence, their level of moral worth will always be entirely dependent on their ‘choices’ and not their ‘circumstances’. Of course, at this point, one might still see some form of luck at play: but for being in Argentinian instead of Nazi circumstances, the individual would have chosen the least worthy option. This may be true, but this is not inconsistent with saying, in both cases, that it lay within the individuals’ control to choose between all worth options 0–1. It is simply to assert that, given different circumstances, one will exercise one’s control in different ways. Of course, one might press that this illustrates just why we are never in control of our choices, because they are always causally determined by circumstances in this manner. But this is to press the ‘problem of causal luck’ (qua the ‘problem of free will’) rather than the problem of circumstantial luck—the import of which I address below (Section 6.2).
5. Moral Weight Let us accept that we are equals in this one peculiar way: in virtue of having any degree of agency, we each have equal ability to attain moral worth because we have some (any) degree of agency but also regardless of having any particular degree of agency. This is, indeed, equality in a moral property. But what of it? Does this moral property satisfy the functional conditions requisite to be ‘basic’, and can we expect something recognizably ‘egalitarian’ to be implied? Let us assume it is sufficiently ‘natural’, and ‘inalienable’, as defined above. Does the equal ability to live a worthy life itself ground implications that have great moral weight or lexical priority? I will consider two arguments. The first is that given each individual has the equal ability to achieve any degree of worth regardless of circumstance, then it is a justified basis for desert.¹⁰ By ‘desert’, here, we do not merely mean moral appraisal. We mean rewards and/or punishment, whether substantive (prizes, fines) or merely expressive (public recognition, condemnation). Some might object that there would be nothing egalitarian about such a system of desert. Instead, it would be just another token of a type of inegalitarian theory, as ancient as Aristotle (Aristotle 2020: 1131a20–1131a30) and as modern as Uwe Steinhoff (Steinhoff 2014), taking human beings to vary in a particular basic moral property ¹⁰ In the ‘pre-institutional’ sense: see (Scanlon 1986).
Nikolas Kirby
189
(worth, merit, or virtue), and then rewarding and punishing appropriately. Such a type of theory only relies upon formal, proportional equality: Proportional Equality for Desert: Equal desert is grounded on equal worth; unequal desert is grounded on unequal worth.
However, this objection fails because unlike such inegalitarian theories, the basis for equal or unequal desert—that is, equal or unequal worth—itself is based upon a deeper moral equality—that is, the equal ability to have any level of worth. This gives the resulting theory of desert an ‘equal opportunity’ -like structure. An (un)equal distribution of desert is justified precisely because each agent had equal ability (‘opportunity’) to have any level of desert (depending upon how they exercise that ability to achieve any level of worth). Hence, by contrast, we actually have: Equal Opportunity for Desert: Equal ability to have any particular level of desert grounded on the equal ability to have any level of worth.11
A greater weakness, or, more accurately, limitation, to such a theory is it being relatively politically inert. It may justify expressive responses, and even some responses impacting upon the distribution of property and political rights qua prizes and punishments. However, prima facie, it seems quite consistent with almost any set of social circumstances, at least ex ante. Since individuals are equally able to be worthy, and thus gain desert, regardless of initial circumstances, this seems to imply an indifference to those initial circumstances. And, whilst forms of desert might impact circumstances, including the distribution of political rights and resources, in media res, degrees of virtue are unlikely to be the primary fact grounding the just distribution of resources and/or political rights (see Rawls 1999: sec. 17, 48; Scheffler 2001: Ch. 10; Scanlon 2013: 111ff ). In short, the risk is that our putative concept of basic equality has some moral implications, but they are unlikely to be important enough to be ‘basic’ enough. My second argument aims to meet this concern, although drawing inspiration from an unlikely source: Friedrich Nietzsche. Following Nietzsche, an action or a life of actions has weight insofar as something of value turns upon it (Nietzsche 1974: 341; 1992: sec. 8). Within the Christian metaphysical system, Nietzsche asserts, the only thing of (dis)value is eternal punishment or ¹¹ Something akin to this is proposed in one of the earliest pieces on basic equality: see (Spiegelberg 1944: 108).
190
Basic Equality
reward. On the one hand, therefore, our actions in life have ‘weight’ because of the process of divine judgement after life: their moral worth (as determined by God) will ground infinitely (dis)valuable eternal punishment or reward. On the other hand, so Nietzsche argues, everything in life itself remains devalued because, ex hypothesi, nothing of value turns upon one’s actions in life. With the ‘death of God’, we rid ourselves of the first part of the theory of divine value—the infinite value of divine of reward and punishment. Yet, we risk being left with the second part—the devaluation of anything in life on its own terms. We risk being simply left with ‘ghastly levity’ (Nietzsche 1992: sec. 8). Hence, with questionable success (see Ridley 1997), he turns to the idea of eternal recurrence, in an attempt to give weight to actions within life (Nietzsche 1974: 341). Now, like Nietzsche, we may refuse to rely upon theological metaphysics to ground value. But unlike Nietzsche, we may be more comfortable with a secular metaphysics that ascribes intrinsic value to various actions, and outcomes, in this world. In fact, I will simply assume that it is of real importance how any agent performs in our model of worth, regardless of any consequences. How one responds to the challenge of being worthy matters—and matters deeply— in itself. However, and this is central to my argument, there are other things that also matter deeply in themselves too, such as truth, justice, welfare, and beauty. Now, on the one hand, together, the existence of all these values means that we need not risk ‘ghastly levity’ from the death of a Christian God. Each individual’s life performance will have some weight because the realization or not of objects of real value turn upon their actions—even if it is just the achievement (or not) of their own level of worth. On the other hand, however, something is still lost by the death of such a God: the assurance that each individual’s life will have equal weight. Within the logic of divine judgement, according to Nietzsche at least, only one thing mattered—the outcome of judgement after life—and each individual’s actions have equal power to determine that outcome: they determine the fate of one’s soul, that is, their own. However, in secular mode, everything of value is only within life, but if that includes values the promotion of which by any agent depends upon contingencies outside of their control, such as truth, justice, welfare, and beauty, then we have no assurance that the occurrence of those things will turn equally upon each individual’s actions regardless of circumstance. Each life will not necessarily have the same weight, the same potential to author what else matters. For example, a person with great circumstances will have great ability to shape the world around them, for good or ill. We may presume, if they
Nikolas Kirby
191
act with much worth, then given their circumstances they are also likely to bring much truth, justice, welfare, and/or beauty into the world. If they act with little worth, then they will not. It lies within their control. Much, therefore, will likely turn upon their life performance. It will really matter, for the sake of realizing value in general, whether this individual performs worthily or not. Contrast this with a person of poor circumstances who has little ability to shape the world around them, for good or ill. This ‘poor’ person’s life performance is, I assume, just as important as the ‘greater’ person’s in itself. It matters, in itself, just as much whether they attain great moral worth or not. But, most likely, given their circumstances nothing much else of value will turn upon that performance. It will not really matter much for the sake of realizing other values. In this way, life performances of the same importance in themselves will have entirely different ‘weight’ in this world given different contingent circumstances. How some lives are led will just not matter as much as others, through no fault of their own. So, my final substantive claim is that this is a misfit. Ceteris paribus, it is unjust that one person’s life performance has any greater weight than another’s. Each person’s life performance relative to circumstance is equally valuable in itself, and thus ceteris paribus those life performances should matter equally in terms of their consequences. If we cannot assume that there is a final divine judgement after life that attributes equal weight to each person’s performance relative to circumstance—regardless of circumstance— then we owe it to each individual within this earthly life to do so instead. Indeed, each person’s ‘circumstances’ are largely arbitrary. It lies within our power to redistribute resources, to change the options available to others, and thus the denominator in anyone’s worth ratio. Given the imperative to attribute equal weight to each person’s performance, we owe it to each individual to equalize that denominator. Each individual’s performance, therefore, will matter equally because the quality of each person’s performance will have equal potential to determine the overall nature of the world that they all inhabit. This basic intuition might provide the grounding for a kind of (retrofitted) responsibility-sensitive, or ‘luck’, egalitarianism. Of course, at this point, one should recognize the potential costs of such redistribution: it may be inefficient; it may involve the transfer of great resources from those who can be reliably expected to be worthy to those who might be expected to be unworthy; and, as a result of the first two, far less of value may be created in the world. However, these are the usual ripostes within the general cut and thrust of egalitarian theorizing. They are disputes
192
Basic Equality
about the ‘ceteris paribus’, above. My aim here has merely been to sketch how we might get that initial egalitarian claim going in the first place.
6. Objections 6.1 Difference in moral options At the end of the model proposed, I introduced one assumption that many might find deeply implausible: in any scenario, an individual will face an infinite set of options between the two options with the highest and lowest degree values, creating a continuous function of credit and thus worth. This assumption has permitted me to claim that any agent has equal ability to have any level of worth, both within any decision and across their life as a whole, entirely regardless of circumstance. Now one might indeed accept that agential scenarios in life rarely have options with the bipolar character of many philosophical thought experiments, but instead generally have many options. But one might baulk at there being an infinity of options in between, creating a continuous range of credit.¹² Of course, as we collect scenarios across a life, we might get closer to a continuous function of possible values for aggregate worth across that life. But even then it will not be completely continuous, and, most importantly, it is unlikely to be the same across different individuals. This introduces a subtle kind of moral luck into our model: individuals will not have access to the same spread of values between absolute worth and unworthiness. Thus, they will not, strictly speaking, have the equal ability to live a life of any degree of worth. My response to this is twofold. First, assuming this is correct, the most important kind of equality does remain in having the equal ability to attain either extreme of (un)worthiness. In this way, each agent is still set the same challenge in life: to achieve the highest degree of worth (and avoid the lowest). Secondly, the problem will not arise because agents have different degrees of agency. It will arise because, given their degree of agency, the rest of their circumstances provide a limited set of options. Change those other circumstances, and one may increase the number and continuity of credit values for any agent. If this is true, then all individuals have the potentiality to have the equal ability to attain any degree of worth regardless of their degree of agency. We retain our conception of basic equality with the introduction of ¹² This is not entirely implausible. For example, if credit turns upon the strength of one’s will to do X, and strength is a continuous function, and one could have any strength in any scenario, then credit will be continuous in any scenario too.
Nikolas Kirby
193
a limited modal component. Further, if this is the case, then plausibly it is a kind of injustice that individuals with the same potentiality for any degree of worth have it contingently limited in different ways. Rather than necessarily being a difference in ‘weight’ in the world, we might think of it as a kind of difference in moral ‘dexterity’ in the world.
6.2 Determinism The second objection to the model is that it is inconsistent with causal determinism, and thus must assume an incompatibilist account of freewill. The objection holds that if causal determinism is true (and it probably is), then we are all causally determined to do whatever we do, and we cannot do otherwise (see, the ‘problem of causal luck’ referred to above, Section 4). Thus, no one can have the ability to be worthy (or unworthy or anything else in between), let alone each have equal ability to do so. Instead, each of us can only have the ability to be worthy or the ability to be unworthy or the ability to be something else in between. Thus, given that we are all causally determined to be worthy to different degrees, we will each have different such abilities. We will not have an equal ability to live a worthy life, relative to circumstances. Contrary to this objection, the model is perfectly consistent with determinism and a compatibilist account of freewill. This is because it is consistent with a compatibilist account of ‘ability’. For example, to adapt the classic analogy of a ‘weathervane’: in one sense a well-functioning weathervane has the ‘ability’ to turn in any direction—north, south, east, west, or anywhere in between; in another sense, of course, it only has the ‘ability’ to turn in the direction the wind blows. We might say, therefore, that it has the ability to turn in any direction, depending upon the wind. Further, if we have two weathervanes, then it is perfectly possible that they might have such an ability equally, but be facing in different directions because their winds blow in different directions. We can say that they both have ‘equal ability’ to turn in any direction, depending upon the wind. So, there is a perfectly common usage of the terms ‘ability’ and ‘equal ability’, which are consistent with the ability to do multiple different things, depending upon a particular fact, even if that fact is entirely causally determined. In the case of our equal ability to lead a worthy life, that fact is whatever determines what we are morally responsible for, even if that is also entirely causally determined.¹³ ¹³ Of course, if one is a determinist, and therefore rejects freewill, and thus also moral responsibility for action, then my account of the basis of basic equality will not impress.
194
Basic Equality
6.3 Lack of alternative options A third line of attack is based upon the version of compatibilism that denies that the ‘ability to do otherwise’ is a necessary condition for moral responsibility, even in the ‘weathervane’ -like sense described above. This position has been convincingly developed using a range of cases where an individual believes that they have alternative options, they choose to undertake one of those options, and do so act, but in fact, contrary to belief, this was the only option available to them because another agent is invigilating their actions, ready to stop them doing anything else (Frankfurt 1969). This creates a problem for the model, because, if moral responsibility does not require the ability to do otherwise, then in such scenarios (perhaps all scenarios) the highest level of moral credit available would be the same as the lowest level would be the same as actual realized level. Thus, both the numerator and denominator would be zero. The value of ‘worth’ would be undefined.¹⁴ However, this result can be avoided, by simply adapting the model mutatis mutandis. If the upshot of the argument against the need for actual alternative possibilities is that an agent only needs to believe that there are such alternative possibilities, then the domain of control within the challenge model need only be understood as relative to those beliefs. This is also consistent with the intuitive idea that our beliefs do not merely constrain our moral responsibility (I am not blameworthy for not saving your life by administering a drug that I reasonably believed I did not have access to, although in fact I did), but also expand it (I am blameworthy for not trying to save your life by trying to access a drug to administer that I reasonably believed I had access to, although in fact I did not) (Frankfurt 2018: 290–292).
6.4 Those without agency Finally, some might think, even accepting the validity of our argument, that it still fails. This is because, possibly, not all human beings have some (any) degree of agency. Instead, the severely mentally incapacitated are left out, as unequals. For some, part of the very imperative of any discussion of basic equality is to explain the inclusion of such individuals. To this objection, I can only press two points. On the one hand, if the critique is that these human beings lack the underlying rational capacities ¹⁴ Substituting CX for CN and C1 : Worth, w =
CX –CX CX –CX
=
0 0
= undefined
Nikolas Kirby
195
to live with any (even the most marginal) degree of agency, and thus moral responsibility, credit and worth for their actions so defined, then to note that they are not equals in being able to attain worth for their actions is no great offence. It simply follows from the opponent’s own premise. If, contrary to that premise, such human beings do have any degree of agency, then they are equals. On the other hand, the critique may be that in according this normative status deep moral significance qua a ‘basic equality’, I am therefore excluding those below the threshold from our moral community. However, this does not follow. Just because some, perhaps many, and important implications follow from our equal ability to attain worth, it does not mean other moral claims might not follow from other morally salient features of our lives. In fact one, very plausible, implication might be that in being morally responsible in this world, we are responsible for the welfare of those who are not. Further, it does not foreclose the discovery of other bases to other basic equalities, perhaps of the more typical claims to ‘equal intrinsic worth’ or ‘equal value’. However, it seems to be a step forwards for the time being to have grounded at least one basic equality amongst almost all human beings.
7. Conclusion All (or nearly all) human beings are one another’s equals in at least one basic moral property, that is, the ability to attain (or disdain) moral worth throughout their life. They each have this ability in virtue of the fact that they have some (any) degree of agency, and they have this ability equally regardless of having different degrees of agency. This equality has fundamental moral implications because it is fitting that the life performance of any such individual has weight, that is, the realization or not of value in the world turns upon such an individual’s performance. Since each individual has the same ability to offer such a life performance, then ceteris paribus each individual’s life performance should have the same (equal) weight. What are the implications of this imperative to accord each individual’s life equal weight? To what extent, if any, might it be justifiably outweighed by other aims? If the argument of this chapter is correct, then these questions should shape a novel research agenda in philosophy grounded in a clear account of why and how human beings are one another’s equals.¹⁵ ¹⁵ Thanks to the anonymous reviewer who read this manuscript, as well as Ian Carter, Alexander Horne, Edmund Tweedy Flanagan, Giacomo Floris, Jonathan Wolff, and Richard Arneson, who have all given invaluable feedback on various iterations of its content.
196
Basic Equality
References Arendt, Hannah. 2006. On Revolution. New York: Penguin Books. Aristotle. 2020. Nicomachean Ethics. Oxford scholarly editions online, edited by Christopher Rowe and Sarah Broadie. Oxford: Oxford University Press. Arneson, Richard. 2014. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Arpaly, Nomy. 2002. ‘Moral Worth’. Journal of Philosophy, 99/5): 223–245. Beardsley, Elizabeth Lane. 1957. ‘Moral Worth and Moral Credit’. The Philosophical Review, 66/3: 304–328. Carter, Ian. 2011. ‘Respect and the Basis of Equality’. Ethics, 121/3: 538–571. Christiano, Thomas. 2014. ‘Rationality, Equal Status, and Egalitarianism’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 53–75. Oxford: Oxford University Press. Dworkin, Ronald. 2000. Sovereign Virtue. Cambridge: Harvard University Press. Floris, Giacomo. 2020. ‘Two Concerns about the Rejection of Social Cruelty as the Basis of Moral Equality’. European Journal of Political Theory, 19/3: 408–416. Frankena, William K. 1962. ‘The Concept of Social Justice’. In Social Justice, edited by Kenneth E. Boulding and Richard B. Brandt, 1–32. Englewood Cliffs: PrenticeHall. Frankfurt, Harry G. 1969. ‘A lternate Possibilities and Moral Responsibility’. Journal of Philosophy, 66/23: 829–839. Frankfurt, Harry G. 2015. On Inequality. Princeton, N.J.: Princeton University Press. Frankfurt, Harry G. 2018. ‘What We Are Morally Responsible For’. In Perspectives on Moral Responsibility, edited by John Martin Fischer and Mark Ravizza, 286–295. Ithaca, NY: Cornell University Press. Gosepath, Stefan. 2014. ‘On the (Re)Construction and Basic Concepts of the Morality of Equal Respect’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 124–141. Oxford: Oxford University Press. Hanna, Nathan. 2014. ‘Moral Luck Defended’. Noûs, 48/4: 683–698. Husi, Stan. 2017. ‘Why We (Almost Certainly) are Not Moral Equals’. Journal of Ethics, 21/4: 375–401. Van Inwagen, Peter. 1983. An Essay on Free Will. Oxford: Clarendon Press. Kant, Immanuel. 1996. Practical Philosophy. edited by Mary J. Gregor and Allen W. Wood. Cambridge: Cambridge University Press. Kirby, Nikolas. 2024. ‘The Problem of Basic Equality: A Constructive Critique’. Journal of Ethics and Social Philosophy.
Nikolas Kirby
197
Korsgaard, Christine M. 2006. ‘Morality and the Distinctiveness of Human Action’. In Primates and Philosophers: How Morality Evolved, edited by Frans de Waal, Stephen Macedo, and Josiah Ober, 98–119. Princeton: Princeton University Press. Kymlicka, Will. 2002. Contemporary Political Philosophy. Oxford: Oxford University Press. Lucas, J. R. 1965. ‘Against Equality’. Philosophy, 40/154: 296–307. Macdonald, Margaret. 1946. ‘Natural Rights’. Proceedings of the Aristotelian Society, 47: 225–250. Markovits, Julia. 2010. ‘Acting for the Right Reasons’. Philosophical Review, 119/2: 201. Nagel, Thomas. 1979. Mortal Questions. Cambridge: Cambridge University Press. Nietzsche, Friedrich. 1974. The Gay Science, translated by Walter Kaufmann. New York: Random House. Nietzsche, Friedrich. 1992. Ecce homo, translated by R. J. Hollingdale. London: Penguin. Parr, Tom, and Slavny, Adam. 2019. ‘Rescuing Basic Equality’. Pacific Philosophical Quarterly, 100/3: 837–857. Peels, Rik. 2015. ‘A Modal Solution to the Problem of Moral Luck’. American Philosophical Quarterly, 52/1: 73–87. Pojman, Louis. 1991. ‘A Critique of Contemporary Egalitarianism: A Christian Perspective’. Faith and Philosophy, 8/4: 481. Rawls, John. 1999. A Theory of Justice, rev. ed. Cambridge, MA: Belknap Press of Harvard University Press. Richards, Norvin. 1986. ‘Luck and Desert’. Mind, 95/378: 198–209. Ridley, Aaron. 1997. ‘Nietzsche’s Greatest Weight’. Journal of Nietzsche Studies, 14: 19–25. Sangiovanni, Andrea. 2017. Humanity without Dignity: Moral Equality, Respect, and Human Rights. Cambridge, MA: Harvard University Press. Scanlon, Thomas M. 1986. The Significance of Choice. The Tanner Lectures on Human Values, 7: 149–216. Scanlon, Thomas M. 2013. ‘Giving Desert its Due’. Philosophical Explorations, 16/2: 101–116. Schaar, John H. 1964. ‘Some Ways of Thinking About Equality’. Journal of Politics, 26/4: 867–895. Scheffler, Samuel. 2001. Boundaries and Allegiances: Problems of Justice and Responsibility in Liberal Thought. Oxford: Oxford University Press. Sher, George. 2017. Me, You, Us. Oxford: Oxford University Press. Smith, Holly M. 1991. ‘Varieties of Moral Worth and Moral Credit’. Ethics, 101/2: 279–303.
198
Basic Equality
Spiegelberg, Herbert. 1944. ‘A Defense of Human Equality’. The Philosophical Review, 53: 101. Steinhoff, Uwe. 2014. ‘Against Equal Respect and Concern, Equal Rights, and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 142–172. Oxford: Oxford University Press. Thomas, D. A. Lloyd. 1979. ‘Equality Within the Limits of Reason Alone’. Mind, 88/352: 538–553. Vlastos, Gregory. 1984. ‘Justice and Equality’. In Theories of Rights, edited by Jeremy Waldron, 41–76. Oxford: Oxford University Press. Waldron, Jeremy. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: The Belknap Press of Harvard University Press. Westen, Peter. 2014. Speaking of Equality: An Analysis of the Rhetorical Force of ‘Equality’ in Moral and Legal Discourse. Princeton Legacy Library. Princeton, N.J.: Princeton University Press. Williams, Bernard. 1973. Problems of the Self. Cambridge: Cambridge University Press. Williams, Bernard. 1981. Moral Luck: Philosophical Papers, 1973–1980. Cambridge Books Online. Cambridge: Cambridge University Press. Williams, Bernard. 1993. ‘Postcript’. In Moral Luck, edited by Daniel Statman. Albany: State University of New York Press. Wilson, John. 1966. Equality. Philosophy at Work. London: Hutchinson. Zimmerman, Michael J. 1988. An Essay on Moral Responsibility. Totowa: Rowman & Littlefield. Zimmerman, Michael J. 2002. ‘Taking Luck Seriously’. Journal of Philosophy, 99/11: 553–576.
8 Equality and Moral Status Challenges to Their Grounding Agnieszka Jaworska and Julie Tannenbaum
There are three key questions about moral status as it relates to moral equality and inequality: (Q1) Is there a sufficient basis of moral status, call it B, that could justify the intuition that the moral status of both human babies and humans with severe, permanent cognitive impairments is equal to that of cognitively unimpaired adult humans? (Q2) Is there some necessary basis B, possessed by most humans but not most animals, that justifies the moral inequality between individuals from these two groups? Lastly, (Q3) Do different degrees to which beings exhibit B engender different degrees of moral status? In this chapter, we first distinguish between different types of moral status (Section 1), clarifying which type of moral equality is at issue in (Q1)–(Q3). After introducing the notion of degrees of moral status (Section 2), we explain (in Section 3) the importance of determining the grounds of moral status and thus of moral equality, and then focus on the challenges of finding affirmative answers to (Q1) and (Q2). We briefly touch on our own previously presented solution, noting its chief shortcoming. Thereafter, we turn to (Q3), and describe the Discrete Interval View, which answers this question in the negative, and the Continuity View, whose answer is affirmative; we show that attempts to defend the Discrete Interval View have thus far been question-begging against the Continuity View (Section 4). Finally, we make explicit several general principles that might seem to motivate the Continuity View and then present various counter-examples to these general principles, thereby showing that an affirmative answer to (Q3) has not been established (Section 5).
Agnieszka Jaworska and Julie Tannenbaum, Equality and Moral Status. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0009
200
Equality and Moral Status
1. Types of Moral Equality For individuals to be moral equals is for them to have equal moral status. To have moral status is to matter from the moral point of view. More specifically, one’s moral status consists in there being certain moral requirements,¹ for one’s own sake, for how one is to be treated by all morally responsible agents. Although the expression ‘having moral status’ might suggest that moral status is some independent moral property one has (perhaps in virtue of some non-moral property) that explains the moral requirements on one’s treatment, this is not the case. ‘Moral status’ is simply a shorthand way of referring to these moral requirements. For utilitarians (who prefer the term ‘moral considerability’ rather than ‘moral status’), the requirement is to treat one’s interests as inputs into the utilitarian calculus, while for non-utilitarians the treatment is typically specified by a set of positive and negative moral claims or rights. At the most general level of description, we can speak of an entity’s having moral status if and only if there is at least one requirement on how that entity is to be treated for its own sake. In this sense we may say that fallen twigs don’t have moral status, but the birds that carry them do; we can do whatever we please to the twigs, but we owe some consideration to the birds—although we might be uncertain or disagree about what type of consideration (for example, including them in the utilitarian calculus? not harming them for fun?). At this most general level, saying that an entity has moral status leaves open what the moral requirements are. Correspondingly, the claim that two entities have equal moral status entails that whatever the requirements are, they are the same for both entities. Below this ‘genus’ level description of moral status and moral equality, there are several different ‘species’ or types of moral status distinguished by different areas of moral concern (for example, a being’s welfare versus a being’s autonomy). As we will show, these types of moral status are not competing notions; an entity can have more than one type of moral status. By contrast, within each type of moral status there are competing, more fine-grained interpretations.
¹ We use a broad sense of ‘requirement’ here that includes moral reasons of various strengths that pertain to how one is treated. Note that not all moral requirements are constituents of moral status. For example, the requirement to keep one’s promises, while it might be partially derived from the promisee’s moral status, is nevertheless not a requirement associated with moral status.
Agnieszka Jaworska and Julie Tannenbaum
201
Moral status #1 concerns regard for an entityʼs existence and welfare.2
This type of status is often interpreted to involve the protection of the life of the individual in the form of moral reasons not to kill them.³ Some writers proceed as if this is all that moral status #1 consists in (Feinberg 1980). But reasons not to destroy the individual altogether—namely, that it negatively impacts their welfare—go hand in hand with other reasons not to interfere with the individual, such as reasons not to torture or otherwise cause severe bodily harm.⁴ More controversially, some interpret this type of status to also encompass positive reasons to aid the individual, especially in order to prevent death, severe pain, or loss of important capacities.⁵ But such interpreters would agree that the reason to aid is less weighty, and hence more easily overridden, than the reason not to destroy the being. Equality of this type of status entails that each and every relevant negative and positive reason governing one’s treatment has equal strength. Moreover, to say that individuals A and B are moral equals is not merely to say that there are equally strong reasons, for example, not to kill either of them, but also to say that if that reason is overridden for individual A on certain grounds, it should also be overridden for individual B when the same grounds apply (we’ll call this the consistency condition).⁶ Moral status #2 concerns regard for an entityʼs autonomy.
Regard for an entity’s autonomy is also called ‘respect’. Respect is typically interpreted to entail protections in the form of reasons not to undermine autonomy (for example, reasons to avoid paternalistic treatment, deception, manipulation, etc., as well as torture and killing), and entitlements in the form of reasons to maintain and enhance autonomy (for example, ² There are several closely related notions here: welfare, interests, well-being, flourishing, what is good for the entity, etc. ³ In certain circumstances, not being killed might not be in the being’s interest (for example, severe incurable pain). The requirements of moral status #1 in such circumstances are open to debate. ⁴ On some views (for example, Tooley 1972), many animals lack any serious protection of life while retaining protection of their other interests. This still amounts to having moral status of type #1. ⁵ Jeff McMahan (2022), for example, does not accept these additional claims, while Warren Quinn (1984) and Agnieszka Jaworska (2007) do. ⁶ The source of the consistency condition is the rational principle to treat like cases alike. Accordingly, the requirement that when the reasons are of equal strength the beings must be treated consistently is not a further moral requirement of equal treatment to be added to the requirements of moral status. It may appear that beings of equal status don’t have to be treated equally when their status is low: while feeding pigeons in the park, I don’t violate any moral constraint if I give all the crumbs to the one pigeon I particularly like. But the best explanation of why this is permissible is that my liking of the one pigeon can override the weak reasons constitutive of the moral status of the other pigeons. In fact, the other pigeons are treated the same way in the sense that if I liked any of them, I would be permitted to give all the crumbs to the one I liked.
202
Equality and Moral Status
reasons to provide certain forms of education, information, etc., as well as life-sustaining aid and pain relief ). The protections and entitlements for individuals with this type of status include but extend beyond the protections and entitlements associated with moral status #1. The more fine-grained details of how autonomy is appropriately protected and enhanced are open to debate—for example, should an autonomous agent be permitted to undermine her own autonomy, for instance, by suicide, to avoid a degrading form of continued existence (see, for example, Hill Jr. 1991; Velleman 1999)? Two individuals have equal status of this type if they merit the same and equally strong protections and entitlements of the sorts indicated above (along with the consistency condition described previously). Some writers discuss the concept of ‘basic equality’. Do they mean equality of moral status #2 or of moral status #1? Basic equality is associated with Ronald Dworkin’s (1977: xii, 180, etc.) expression, ‘equal concern and respect’.⁷ But this expression involves an ambiguity, depending on which term in the phrase one focuses on. When ‘basic equality’ is thought of as ‘equality of respect’ (usually interpreted as respect for autonomy), it amounts to equality of status #2 at a rather general level of description. On the other hand, if ‘basic equality’ is thought of as ‘equality of concern’ (usually interpreted to track a being’s interests), this suggests equality of moral status #1 at its most general and unspecified level of description. Following Dworkin, Jeremy Waldron presents basic equality as involving both aspects (for example, 2017: 52–53). We describe problems with combining these two types of moral status in one concept in Section 3. Moral status #3 concerns opacity toward an entityʼs agential capacities.
The moral requirement that constitutes this status is to deliberatively ignore (treat as opaque) differences in how robust an individual’s agential capacity is, or how often it manifests, when determining how we treat that individual.⁸ For example, suppose a person has moral status #2 in virtue of their autonomy. Instead of focusing on reasons to protect, maintain, and enhance autonomy, as moral status #2 does, moral status #3 entails that, when considering whether, for example, one can paternalistically override the requirement to obtain consent, we not deliberatively investigate or take into consideration how robust the person’s autonomous capacity is. Equality ⁷ Other authors use the term ‘basic equality’ in different senses (for example, Carter, this volume). We will not address whether these other conceptions also give rise to the difficulties we describe. ⁸ Of course, one does not always ignore these differences. For more details, see Carter (2011 and this volume). Note that Carter does not use the term ‘moral status’ when describing opacity respect.
Agnieszka Jaworska and Julie Tannenbaum
203
of moral status #3 should thus be understood as equality of the strength of the reasons to treat the robustness of one’s agential capacity as deliberatively opaque (along with the consistency condition). One could imagine a moral status that combines moral statuses #2 and #3, that is, a status concerned with autonomy that involves not only reasons to not undermine and to enhance autonomy, but also reasons to deliberatively ignore the degree of autonomy a person has. Moral status #4 concerns treatment of individuals by social systems (political, legal, economic, linguistic, etc.).
Some understand this status to centrally involve requirements on how the society’s material resources (for example, income and wealth) are allocated. Equal status is then interpreted as requiring fair distribution of resources (presumably along with the consistency condition on when a distribution scheme is overridden). The advocates of the Rawlsian Difference Principle and luck egalitarians are in this camp, though they disagree about what kind of distribution is fair. Others think that this status primarily entails requirements on how citizens relate to each other in the public sphere (for example, their access to economic cooperation and education, and their power to influence societal processes, institutions, and decisions).⁹ Equality of status is then interpreted as requiring a fair allocation of access and power in the societal realm, where possession of material resources might play only a derivative role. Yet other interpretations of this status include libertarianism (society chiefly needs to provide certain liberties to its members) and the capabilities approach (the requirements on society are to equalize the capabilities necessary for a flourishing life). Status #4 is clearly a moral notion since it involves moral requirements on how social systems—and so the individuals composing those systems— should (morally speaking) relate to one another. Nonetheless, there is a serious concern whether these moral requirements do in fact amount to a notion of moral status. For C to have moral status is not merely for there to be moral requirements on how some moral agents treat C, but on how any moral agent (or institution) treats C (Jaworska and Tannenbaum 2023: sec. 2.4). Consider, for example, the right to vote. Citizen C does not have a right to vote in every society; not every society has a duty to give C a voting say in its affairs. Only C’s society has this duty. Similarly, one does not have a ⁹ Access sometimes depends on linguistic practices, such as the use of honorific titles (‘Sir’, etc.). See Elizabeth Anderson (1999) and Samuel Scheffler (2003) for details.
204
Equality and Moral Status
right to other societies’ resources; one has a right only to one’s own society’s resources. Of course, there might be a strong reason for any morally responsible agent in the world to aid those in need or to help one gain representation in one’s society, but these reasons are a matter of moral status #1 or #2. So, strictly speaking, the requirements of #4 do not constitute moral status. However, they are, in our opinion, sufficiently analogous to moral status—in the sense that every morally responsible agent within a given society must treat those in that society in accord with these requirements—to be included here. Moral statuses #1–4¹⁰ need not be competing notions (though in some contexts they might come into conflict). An individual can have more than one type of moral status. Moreover, the same act can violate more than one of an individual’s types of moral status. For example, if a being has all forms of moral status above, then torture would violate the requirements of moral statuses #1 and #2, and also moral status #4 when done by the state to its citizens. Relatedly, the same acts can treat individuals unequally in more than one sense. For example, colonialism and segregation within a state, which involve treating some individuals as inferior to others, violate more than one type of equal moral status. Both actions violate the subordinates’ equal moral status #1 insofar as their wellbeing, but not that of the ruling class, is put at high risk; both violate their equal moral status #2 insofar as the freedoms and opportunities to exercise autonomy are severely restricted for the subordinates but not for the ruling class; and both violate the subordinates’ equal moral status #4 insofar as these violations occur in the context of claiming political authority over both groups.
2. Degrees of Moral Status For each type of moral status, it is possible to distinguish varying degrees of that status.¹¹ While types of status identify different areas of moral concern (for example, a being’s welfare versus the being’s autonomy), degrees of status mark variations in the extent to which we should have concern in that area. Just as each type of moral status is open to competing interpretations, so too are degrees of moral status. For example, with respect to moral status #1, for those who hold that there is an upper bound,¹² the highest degree of ¹⁰ Are there other notions of moral status and equality that we have not yet captured? At first, it might seem that the classical social contract theory offers a different notion of moral equality in terms of the power to bind oneself to another and no one’s having authority over another in the state of nature. But this form of moral equality derives from the equality of moral status #2 and hence is not really a separate kind of moral equality. ¹¹ Many theorists, such as Kant and the utilitarians, deny or do not distinguish degrees of moral status. ¹² Richard Arneson (1999: 120) suggests that there might not be an upper bound on the degrees of moral status.
Agnieszka Jaworska and Julie Tannenbaum
205
this status would involve the fullest set of the protections and entitlements associated with this status.¹³ Moreover, the reasons not to kill would be particularly stringent (amounting to a right not to be killed), whereas the reasons to aid would be strong but not stringent (on the interpretations that include them). Lower degrees of moral status #1 might involve all the same requirements, but correspondingly weaker (that is, more easily overridden) ones. For example, benefits to beings of a higher moral status would more readily serve as reasons to override protections and entitlements of beings of a lower status than vice versa. Alternatively, a lower moral status might lack particular protections or entitlements (for example, one might lack the right not to be killed but still retain the right not to be tortured). A third alternative might combine these approaches: at a lower degree of moral status not only are some protections omitted, but also the remaining protections are weakened.¹⁴ Lastly, degrees of moral status may also differ in how the concern for the interests of the individual is implemented. For example, while the highest degree of moral status #1 involves a stringent reason not to be killed which overrides reasons to maximize the overall good, a lesser moral status might simply entail factoring the individual’s good into the utility calculation for maximizing the overall good (McMahan 2002: 245–247).¹⁵ Note that two individuals have equal moral status of a given type if they have the same degree of that status. Equality is most often discussed with respect to the highest degree of moral status.
3. The Grounds of Moral Status and Why They Matter For each type of moral status, a key question arises: in virtue of what does an individual merit this status (what are its grounds)? If degrees of a type ¹³ Some writers might add that being treated fairly is an aspect of the highest degree of moral status #1. As John Broome (1990–1991) sees it, fairness is not to be equated with merely giving equal weight to equally weighty reasons (that is, it is not equivalent in content to the rational requirement of treating like cases alike—see n6 above). Rather, beings are not treated fairly if their equal claims are simply settled by tiebreakers. To illustrate this, suppose that a general must select one soldier from the battalion for a deadly mission and that one outstanding soldier is most likely to accomplish the mission. It would be unfair of the general to let the importance of the mission simply override the outstanding soldier’s claim to escape death, which is equal to that of every other soldier. That is, in addition to the reasons associated with each soldier’s claim to escaping death and the reasons to accomplish the mission, there are competing reasons generated by considerations of fairness, to give each soldier’s claim an equal chance of satisfaction by choosing one solider randomly. By contrast, presumably (Broome does not address this), at lower degrees of moral status no such reasons generated by fairness arise and a tiebreaker can directly determine whose claim to survival is overridden. At the highest degree of moral status, fairness is relevant even in distributing the satisfaction of weak (trivial) equal claims, while at lower degrees of status, fairness is inapplicable even in adjudicating the claims not to be killed. ¹⁴ Degrees of moral status can in principle be constructed, in analogous ways, for other types of moral status; for example, for status #2, we do seem to respect an 8-year-old, a teenager, and an adult differently. Nonetheless, degrees are most commonly explicitly or implicitly invoked for status #1. ¹⁵ See Jaworska and Tannenbaum (2023: sec. 3) for a fuller discussion. Note that McMahan’s proposal implies that Broome’s reasons of fairness do not apply at lower degrees of moral status.
206
Equality and Moral Status
of status are distinguished, the question of grounds arises for each degree of that type of status. The answers to these questions of grounds enable one to determine which beings have what type and degree of status. In this section we focus on the grounds of the highest degree of each type of status. On the prevalent Kantian interpretation of status #2, the relevant ground is the capacity to reason about, and act on, judgements of what it is good to do. Some think these judgements must include moral judgements, others do not.¹⁶ Regarding status #4, many philosophers take it to derive partly from status #1 or status #2, independently or in combination, even though, as we saw in Section 1, the requirements of #4 do not strictly amount to moral status.¹⁷ If so, we are then led to ask about the grounding for status #1 and/or status #2. For status #3, we read Ian Carter as proposing these agential capacities as the grounds: ‘to reflect on one’s desires and circumstances, to set ends for oneself, to form coherent plans’ (2011: 552).¹⁸ When it comes to grounding the highest degree of moral status #1, most of the accounts that have been proposed—including those in this volume—are plagued by several problems, just two of which we will note here.¹⁹ Under-inclusivity
Consider grounding the highest degree of moral status #1 on having a sophisticated capacity, such as a form of rational agency,²⁰ the capacity to care (Jaworska 2007), or the capacity to have ‘moral worth’ (Kirby, this volume). One might even expand the grounds to include the potential for such capacities (Arneson, this volume). However, these accounts exclude some human beings who we intuitively think do have the highest degree of moral ¹⁶ It may be tempting to assume that the grounds of the highest degree of moral status (of whichever type) must themselves be moral (for example, Anderson [1999: 312] discusses views making this assumption). After all, what else could ground moral status but a moral property? In fact, the term ‘moral property’ is ambiguous; it might mean a property that is morally relevant (for example, capable of grounding moral requirements) but not itself moral, or it might mean a property that is itself moral (for example, a feature of moral agency). So, for example, the highest degree of moral status #2 might be grounded on the capacity to reason morally (itself a moral property) or grounded merely on the general capacity to reason (not itself a moral property, but a morally relevant property). On Christine Korsgaard’s interpretation of Kant’s Formula of Humanity, moral status #2 is grounded on the capacity to act for reasons of any kind, moral and non-moral alike (2007: 110–114, esp. 114). ¹⁷ Waldron (2017) clearly thinks that equal degrees of moral status #4 partly derive from ‘basic equality’ (equal degrees of moral status #1 and moral status #2). Anderson (1999) and Scheffler (2003) suggest a similar view, but with respect to only moral status #1. However, others think that even if individuals’ moral status #1 is not equal, there would nevertheless be other grounds for equal degrees of moral status #4 (see Lippert-Rasmussen, this volume; Sangiovanni, this volume: 115). ¹⁸ Carter borrows the cited wording from Thomas Hill Jr. (2000: 87). ¹⁹ For a detailed discussion of the deficiencies of most attempts to ground moral status #1, see Jaworska and Tannenbaum (2023). ²⁰ There are many who hold such a view, from those inspired by Kant to Thomas Christiano, this volume.
Agnieszka Jaworska and Julie Tannenbaum
207
status #1: those whose relevant capacities will not develop beyond the earliest stages due to genetic or other deficiencies. One can now see how ‘basic equality’, if is it understood as combining the highest degree of status #1 and status #2, faces an even more extensive threat of under-inclusivity. In general, combining the two moral statuses prejudges that these two types of status will apply to the same individuals on the same grounds. If the grounds of these statuses is autonomy (which is a natural assumption for status #2), those with permanent and severe cognitive impairments, and perhaps also 3-year-olds and even some young teens, will be excluded,²¹ and thus various human beings will not be moral equals to autonomous beings, even with respect to status #1. We cannot assume that these two types of status will apply to the same individuals on the same grounds ahead of inquiring into these grounds (for example, the grounds of moral status #1 might be the capacity to care, not autonomy). In addition, the assumption that the two types of status apply to the same individuals leads to a problem of intelligibility. Insofar as status #2 is interpreted in the standard Kantian way, it is inapplicable to beings (currently or permanently) incapable of autonomy, and hence its requirements make no sense with respect to such beings, and yet, at least some of these beings (for example, humans with permanent severe cognitive impairment) should be considered to have the highest degree of moral status #1. Over-inclusivity
Consider the proposal that being a human animal is a sufficient ground for moral status #1 (see Benn 1967). On this view, anencephalic babies would have the same moral status as you and I, but many think that such babies do not have the same right not to be killed and that we do not have the same reason to aid them as we do babies with other disabilities, even severe cognitive disabilities. So this implication is intuitively very implausible.²² In light of problems such as the under- and over-inclusivity of the various accounts above, one might wonder whether we can sidestep the question of the grounds of the highest degree of moral status #1 (or of any other type of ²¹ Interpretations of moral status #2 do not converge on whether beings that exist but are not yet autonomous have an entitlement to the proper development of autonomy (see Wood 1998). Most Kantians agree that there is no duty to bring autonomous beings into existence. Elizabeth Harman thinks there is a reason to create, though not for reasons of respecting autonomy (2004). ²² According to Eva Kittay, ‘being born a human being in a condition that is compatible with life’ (this volume: 284, note 35) is sufficient for moral status #1. This sufficient condition, according to Kittay, does not cover anencephalic babies. However, anencephaly is compatible with a short life, so it is unclear why this sufficient condition would not imply that anencephalic babies have the same right to available lifeprolonging treatment as the rest of us.
208
Equality and Moral Status
moral status) and hence the grounds of the corresponding equality. Would we do better instead to simply point to, for example, the harms associated with treating one as inferior (Sangiovanni 2017)? On Sangiovanni’s view, the key harm of treating one as inferior is an injury to one’s sense of self. But human infants and severely cognitively impaired humans are not susceptible to this kind of injury since they lack a sense of self. So this approach cannot account for the applicability of the requirements constitutive of moral status #1 to these beings. Also, various treatments (actions or deliberations) are required or prohibited as part of any of the types of moral status distinguished above, regardless of whether the individual is aware of the treatment, and thus regardless of the treatment’s effects on the individual’s sense of self. So, Sangiovanni’s approach also cannot account for this aspect of moral status.²³ Thus, we would sidestep the question of grounds only at these high costs.²⁴ In previous work, we proposed an account that attempts to avoid the problems of under- and over-inclusivity in grounding a high degree of moral status #1 (Jaworska and Tannenbaum 2014; 2015). Briefly, our account rests on the idea that sophisticated cognitive and emotional capacities can be incompletely realized. If cognitively sophisticated capacities ground a higher degree of moral status #1, so too can the incomplete realizations of those very same sophisticated capacities. To understand the notion of incomplete realization, consider first learning a skill ‘by doing’: for example, a child is learning to read by sounding out individual letters, guided by the end of fluent reading. Reading badly is not an apt description of this activity; rather, the child is in the process of piecing together the activity of reading but has not mastered it yet. Against certain background conditions, the sounding out of the letters is an incomplete realization of the activity of reading. Similarly, in the course of their development, humans learn cognitively sophisticated activities ‘by doing’. Consider playing a game like peekaboo. It involves following a simple rule, which is a rudimentary version of practical reasoning. Peekaboo can thus be used as an early lesson in teaching a novice to master practical reasoning. When this end ²³ An additional problem is that the approach would deny moral equality to those who stand to suffer different degrees of harm from being treated as inferior. While this approach generates requirements not to treat beings as inferior regardless of whether they would suffer a lot or a little from such treatment, it is committed to there being stronger reason to avoid treating as inferior those beings who would suffer more rather than less from the same such treatment—they are thus not moral equals. ²⁴ Anne Phillips claims that equality is ‘the commitment we make to regard one another as equals, and the claim we make on others who have so far failed to see us in this light’ (this volume: 226) and that it ‘needs no such justification’ (225). Nonhuman animals are incapable of such commitments and claims (228–229). But so too are some humans, and yet moral equality extends to all human beings (228). Why not then also to nonhuman animals? She claims that ‘drawing that line [between human and nonhuman animals] says nothing about one species being superior to another, and requires nothing more substantial than the very basic recognition that all humans are indeed humans’ (229). This sounds like a species membership justification for why nonhumans are not our moral equals.
Agnieszka Jaworska and Julie Tannenbaum
209
guides a parent or a caretaker, and certain background conditions are met, a child (or a developmentally impaired adult) incompletely realizes practical reasoning by modelling practical reasoning in these rudimentary ways. Insofar as a child can incompletely realize a cognitively sophisticated activity, the child thereby has this capacity and the corresponding moral status. The role of the background conditions is to secure a sufficiently robust connection between the novice’s rudimentary activity and the caretaker’s end of the novice’s mastering the sophisticated activity such that the former is an incomplete realization of the latter. The connection is inadequate if it’s unreasonable for the caretaker to adopt this end or if the rudimentary activities are not a feasible means of achieving the caretaker’s aim. In the case of unimpaired babies, their rudimentary first steps of cognitive sophistication are a feasible means of achieving the end of cognitively sophisticated mastery, and thus the babies have the relevant capacity. But what about rudimentary activities of human beings whose path to cognitive sophistication is permanently blocked? There can still be a sufficiently robust connection between these activities and the caretaker’s end of raising a cognitively sophisticated being. All caretakers are required, in virtue of their role as caretakers, to have their charges’ flourishing as an end—regardless of whether this is attainable—and human beings can flourish only if they eventually develop sophisticated cognitive capacities. So in this case, it is still reasonable to adopt the end of attaining sophisticated capacities, not as an aim to be realized, but rather as a guide to what next-best aim to adopt. The rudimentary activities of many permanently severely cognitively impaired human beings would then be a feasible means to the caretaker’s reasonable next-best aim, which in turn would be guided by the end of attaining cognitive sophistication. The appropriate connection is thereby secured. This establishes these humans’ capacity to incompletely realize cognitively sophisticated activities and the corresponding high moral status. Our account thus overcomes the problem of under-inclusivity. It also avoids over-inclusivity. Anencephalic babies cannot perform even the most rudimentary lessons in cognitive sophistication, so they cannot incompletely realize cognitively sophisticated activities. Thus, they lack the relevant capacity and so don’t get any special moral status on our account. In addition, though many animals can perform activities (such as playing fetch) that look like early lessons in cognitive sophistication, these activities cannot meet the background conditions we outlined above. For example, it is not feasible to make a dog cognitively sophisticated, and so the caretakers who know this cannot reasonably adopt the end of developing the dog’s cognitive sophistication, since the dog can flourish perfectly well without such sophistication.
210
Equality and Moral Status
Although this account shows that babies and those with severe permanent cognitive impairment have a higher moral status #1 than, for example, a dog—and thereby makes considerable progress towards answering questions (Q1) and (Q2) with which we began—we did not attempt to show that incompletely realized capacities are sufficient for the highest degree of moral status #1. Nor did we address question (Q3), to which we turn in the remainder of this chapter.
4. Mapping Degrees of Moral Status onto Their Grounds There are two ways to think about how degrees of moral status map onto their grounds. The first approach is the Discrete Interval (aka, Threshold) View. Suppose, for example, the capacity for consciousness grounds the lowest degree of moral status #1, while sentience grounds a higher degree, and cognitive sophistication grounds the highest degree.²⁵ On the Discrete Interval View, if two beings’ only relevant cognitive capacities are consciousness and sentience, then they both have the same degree of moral status, and are thus moral equals, regardless of the degree to which they are conscious or sentient. If there is a highest degree of moral status, once individuals meet the ground for it (for example, by having a certain cognitively sophisticated capacity), they all merit the same degree of moral status, which is the highest, regardless of how well they fulfil that condition (that is, regardless of the degree of their cognitive sophistication). Discussions of the highest degree of moral status typically presuppose this approach. But once the importance of some feature (for example, sentience or the capacity to reason) is proposed as a ground of some degree of moral status, it seems to some that one’s moral status depends not only on one’s having that capacity, but also on the degree to which one has it (that is, it matters how well one can exercise it or how often).²⁶ This is the Continuity View: two beings ²⁵ Keep in mind the general point that, while the same individual can have different types of moral status, she cannot have different degrees of the same type of moral status. If an individual is both sentient and cognitively sophisticated, the individual does not have two degrees of one type of status (moral status #1), but rather the higher degree of moral status #1. ²⁶ Arneson (1999: for example, 105) can be so interpreted, or, alternatively, understood along the lines of his more recent article (2015: for example, 36), which frames the issue somewhat differently. There he assumes that lower degrees of moral status depend on lower degrees of rational capacity, rather than on lacking that capacity altogether and having an entirely different capacity. He emphasizes the tension in holding that differences of degree of rational capacity make no difference to the relative moral statuses of two humans but do explain the difference in moral status between animals of different species (for example, of a bug versus a bird). On our framing of the problem, the thresholds of moral status are marked by altogether new capacities, but the challenge remains to explain why the degrees of a given capacity do not matter for moral status. The responses we discuss, including our own in Section 5, apply to both versions of the problem.
Agnieszka Jaworska and Julie Tannenbaum
211
are moral equals if and only if they not only share the same grounds of moral status (for example, a certain capacity), but also possess those grounds to the same degree. The Continuity View poses a challenge to the Discrete Interval View, since the latter cannot simply stipulate that, for example, how effectively one can reason does not affect one’s moral status and hence moral equality with others who can reason. The problem is essentially the same for any type of moral status and regardless of whether the feature proposed as the ground is actual, potential, or incompletely realized. The Continuity View poses a serious challenge to any claim that there is moral equality among humans in spite of how much they differ in their levels of cognitive abilities (or other possible grounds for moral equality). The Continuity View allows for moral equality in only philosophically trivial cases: namely, when two beings are alike not only in whether but also in the degree to which they possess the feature(s) that ground moral status. But these philosophically trivial cases are not the ones of interest to moral and political theories of equality. Views which insist that entities that merit any moral status at all have equal moral status might seem to avoid this type of challenge by setting one, low threshold of moral status and denying that moral status comes in degrees. For example, according to Peter Singer and Tom Regan, sentience or being a subject of a life, respectively, is the condition of having moral status, and all entities with this capacity are moral equals in that their like interests are assigned the same weight in the utilitarian calculus that determines what action to take (for example, Singer 1975: 9) or they equally have rights that protect their inherent worth (Regan 2004: esp. secs. 7.5 and 8.5). Richard Arneson (1999: 126) posits that avoiding the intractability associated with the Continuity View is a strength of Singer’s position. However, an advocate of the Continuity View could still note that some entities instantiate the capacity for sentience or being a subject of a life better or more fully than others—by sensing many more types of things or by having more sophisticated lives— and then press for an explanation of why such differences have no effect on the entities’ moral status (that is, on the weight assigned to the like interests of such beings in the calculus or the strength of their rights).²⁷ Some try to avoid this objection by treating the proposed ground of moral status (for example, rational capacity, sentience, or capacity for welfare) as a range property as opposed to a scalar property.²⁸ For example, the property ²⁷ Note that these views will not necessarily treat beings with equal moral status alike, due to their interests’ being impacted differently by the same actions. This does not affect the applicability of the present objection. ²⁸ John Rawls (1971: 508) introduced this idea.
212
Equality and Moral Status
of being more than 5 feet tall is a range property and thus does not admit of degrees, even though its corresponding scalar property, one’s height, does. Likewise, having some minimum rationality is a range property corresponding to the scalar property of rationality. However, this approach begs the question against the Continuity View. The question is why the ground of moral status is the range property of minimal rationality as opposed to the corresponding scalar property. Moral status #3 (involving a requirement to ignore differences in how robust a capacity is when determining how we treat an individual) also doesn’t solve this problem, contrary to Carter (2011). Suppose one has some degree of moral status #1 or #2 in virtue of having the scalar property of agential capacity. In thereby having a minimal degree of this capacity, one would, per Carter’s view, also have moral status #3. But we cannot simply assume that moral status #3 has priority—that is, we cannot assume that the requirement (stemming from status #3) to ignore differences in capacity trumps the requirements stemming from status #1 or #2, which are (by our hypothesis) sensitive to the degree to which the being possesses the capacity. Moreover, the priority of status #3 (the strength of its requirement) seems to (at least partly) rest on its grounds, which leads us back to the original problem: if the grounding property of status #3 comes in degrees, then status #3 would, on the Continuity View, come in degrees (for example, degrees to which we should ignore how rational a person is). One cannot beg the question and assume that a range property (for example, a bare minimum of agential capacities, as Carter proposes) is the ground for status #3.²⁹ Christiano (this volume: 159 ff ) addresses the same problem by pointing to intuitions that only make sense on the Discrete Interval View. Two of these intuitions are that it is unfair for those who have merely achieved minimal rational capacities to lack the advanced rational capacities that others possess,³⁰ and relatedly, that there are moral reasons to enhance the rational capacities of the former individuals. However, the proponents of the Continuity View readily accept that their view contradicts powerful egalitarian common-sense intuitions. Insofar as Christiano’s argument simply adds more contrary intuitions into the mix, it may do little to dislodge the ²⁹ Carter seems to think that a relatively low level of agential capacities grounds a high (perhaps the highest possible) degree of status #3. This approach does not solve the continuity problem for moral status #3, but instead raises the question of why degrees of capacity don’t matter for this status. Carter’s response invokes the Strawsonian distinction between the ‘participatory perspective’ and the ‘scientific perspective’ and suggests the two ‘ought to be confined to their proper spheres of human life’ (this volume: 133). But, as Strawson (1962) points out, reactive attitudes can be mitigated, without being wholly removed, by factors that are part of the scientific perspective, which in principle includes the degree of one’s agential capacities (see especially Strawson’s discussion of children). ³⁰ See footnote 13 above for the connection between fairness and moral status.
Agnieszka Jaworska and Julie Tannenbaum
213
theoretical considerations that appear to support the Continuity View. Moreover, Christiano seems to think that the capacity to participate in some possible collective process of practical reasoning and rational activity (which he proposes as the ground of the moral status of individuals above a certain threshold of rational capacities) does not admit of degrees, barring extreme cases of disability. However, a scalar interpretation of this capacity is still possible: for example, one has greater capacity if one can participate in and contribute to more possible collective rational processes. In light of the challenge to the Discrete Interval View, why not simply accept the Continuity View? After all, all sorts of differential treatments based on capacity level are intuitively permissible. For example, we admit only those who score highly on an intelligence test into an advanced educational program. However, according to common-sense intuitions, differential treatment is only sometimes justified; there is a presumption in favour of treating people as moral equals which is outweighed only by very significant counter-considerations (just as the right not to be killed can sometimes be overridden by very significant counter-considerations). Hence, the presumption in favour of equal treatment matters, including when overridden, and this is what the Continuity View denies. Moreover, differential treatment is intuitively not justified when it comes to, for example, the right not to be killed. While it might be true that the intelligent have more to lose in death, intuitively, that does not give them a stronger right not to be killed. By contrast, if intelligence grounds a high degree of moral status #1, then, on the Continuity View, the more intelligent among us would have a stronger right not to be killed than the less intelligent. So the standoff between the Continuity View and common-sense beliefs about equality remains.
5. Reasons to Doubt the Continuity View The Continuity View must be argued for, not merely assumed. One way to do so is to derive it from a more general principle, such as this: if B is the basis of an M degree of X, then a greater degree of having or exhibiting B corresponds to a greater degree of X. However, this general principle does not hold. Take a simple example: the depth to which a loaf of bread can be cut depends on the pressure put on the knife. But there is a threshold of pressure at which the bread can be fully cut and any further increase in the pressure of the knife will not increase the depth to which the bread can be cut. Or consider the relation between the number of reasons one sees and the quality of one’s decision. A human’s seeing that there are two reasons, not just one, that
214
Equality and Moral Status
bear on deciding what to do improves the quality of her decision. But past a certain point, the more reasons she sees, the worse her ability to come to a well-reasoned decision is. One might see these cases as belonging to a special category in which some factor independent of B sets an upper limit to a possible degree of X: there is an upper limit to the depth of the cut, determined by how much bread there is in a given loaf. Similarly, the size of a balloon depends on how much air is pumped into it, but there is an upper limit to how far the balloon can stretch before it bursts. And in the case of decision-making, there is a psychological upper limit to the number of things a mentally finite being (such as a human) can keep in mind. Let’s then only consider counter-examples from this point forward in which there is no physical or psychological limit, independent of B, to the degree of X. Take life and its basis in metabolism and other self-maintaining processes and adaptations to the environment. Increasing the number and complexity of the self-maintaining processes and adaptations beyond the essentials does not make a being more alive. One might object that, in this case too, there is a limit to the degree of X, but this time the limit is not physical but conceptual: the concept of life is such that life does not admit of degrees (one either is or is not alive; one cannot be more or less alive). Similarly, a heap is based on the number of grains, but once there are enough grains for a heap, adding more grains, though it makes the heap larger, does not make it more a heap. Moral status is different because it is conceptually coherent to understand moral status as admitting of degrees.³¹ Furthermore, it is not built into the concept that moral status has an upper limit. For any degree n of moral status, we can always construct degree n+1 of moral status, which includes all the requirements of status n, plus an additional requirement that the beings with n+1 degree of moral status take priority over those with only n degree of moral status (at least in some cases of conflict). Is there then a counter-example (to the claim that, if B is the basis of an M degree of X, then a greater degree of B corresponds to a greater degree of X) that rests neither on a conceptual limit to the degree of X nor on a physical limit set by something other than B? We offer this one: the density of water depends on its temperature.³² Starting at 0ºC, the higher the ³¹ Of course, some (for example, utilitarians and Kant) deny that the moral requirement(s) constitutive of moral status in fact come(s) in degrees, but they do not deny that such moral requirements can, in principle, come in degrees. ³² The density of water is a matter of two further factors, which are directly affected by temperature: the vibration of H2 O molecules and how rapidly their bonds are being broken and remade (and hence whether extra H2 O molecules are trapped in the water lattice). The vibrations decrease density, and the trapping of H2 O molecules increases density.
Agnieszka Jaworska and Julie Tannenbaum
215
temperature, the higher the density. But at 4ºC water reaches its highest density. As temperature increases beyond 4ºC, the density of water decreases. Density of water can coherently be thought to have no upper limit, but in fact does have an upper limit dependent only on the temperature. This counter-example to the refined principle does not involve evaluative terms. Perhaps the more general principle from which the Continuity View is derived holds only when B is a non-evaluative term and V is an evaluative term: if B (a non-evaluative property) is the basis of an M degree of a valueladen feature V, then, so long as there is no upper limit to possible degrees of V (independent of B), a greater degree of B must correspond to a greater degree of V. However, even this principle remains open to counter-examples. Consider the right to drive a passenger car on public roads based on meeting criteria assessed on a driving test. Some prospective drivers will be at various levels above the threshold necessary to pass the test, but all drivers meeting the minimal criteria have a right of the same strength to drive a passenger car.³³ Note that, in contrast to the life and heap examples, variable strengths of the right to drive are conceptually coherent. A weaker right to drive is exemplified by restrictions on driving alone (for example, those who have only learner permits) or on night driving (due to vision defects). And, in principle, one could have a stronger right, such as to use highway lanes designated for drivers with higher test scores. But does this counter-example succeed? Suppose (as will likely become the case) self-driving cars become even better drivers than we are. Suppose further that they sometimes aimlessly drive around without passengers and that this, along with human drivers, leads to traffic congestion. It seems that restricting self-driving cars without passengers, while allowing human drivers, would be justified, and hence self-driving cars would not have the same right to drive as we do. Perhaps what explains this intuition is that having interests or ends that can be satisfied by driving is also a necessary condition of the right to drive, and so the right to drive is not based merely on a sufficient skill level, but also on the kind of grounds that give rise to moral statuses #1 and #2. However, this is, in fact, grist for the mill, since the example then shows how varying degrees of having interests or being an autonomous agent can (in combination with varying degrees of skill) nonetheless give rise to rights of equal strength. What if the right to drive is, however, not merely based on one’s having interests or being an autonomous agent, but instead directly derived from moral status #1 or #2? For example, the right to drive might, at least partly, derive from the right to freedom of movement, a right included in moral ³³ Thanks to Nicholas Hanson-Holtry for this example.
216
Equality and Moral Status
status #2 or #1. If so, we must find an example involving a value-laden term V that is independent from moral statuses #2 and #1. So, consider instead an alternative counter-example involving a valueladen term (V ) that in principle has no upper limit: justified indignation and other forms of blameworthiness for what one did. The degree of blameworthiness (V ) depends partly on how horrendous what one did is (B). One is more to blame for killing one versus ten people. But, other things being equal, killing a billion people does not call for more indignation than killing a million people. More indignation could, in principle, be called for, but it is not. And here is a final counter-example. There are non-instrumental reasons to promote the playing of a particular (physical) sport and these partly depend on the sport’s involving physical skills. There are other reasons as well (for example, in playing a sport one learns to cooperate and to compete), but if we hold these constant, there is not more reason to promote one physical sport over another if the former involves more physical skills. The decathlon (encompassing ten track-and-field events) involves more skills than pole vault alone, but there is no more reason to recruit people to play the former rather than the latter. These reasons could, in principle, vary in strength, but they in fact do not. Note that, in the last two examples, the value of B (the negative value of killing people, the positive value of physical skills) is the basis of the valueladen V (blameworthiness, reason to promote a sport). So, these are also counter-examples to even this narrower version of a principle from which the Continuity View might be derived: if the value of B is the basis of an M degree of a value-laden feature V, then, so long as there is no upper limit to possible degrees of V (independent of B), a greater degree of B must correspond to a greater degree of V.³⁴ ³⁴ After drafting this chapter we became aware of Zoltan Miklosi’s (2022) article, in which he too argues against the Continuity View by objecting to a principle from which he takes it to be derived, namely: ‘if the presence of a valuable property warrants a certain kind of response towards its bearers, then every variation in the degree to which the property is present necessarily constitutes a reason for a corresponding variation in the response that is warranted towards its bearers’ (372). His counter-examples are philosophical or historical inquiry and conversing: if two topics are sufficiently significant (or two people are sufficiently rational), and hence one has a reason to engage in philosophical or historical inquiry on that topic (to converse with the person), then even if one is far more significant (rational) than the other, the reasons that govern how one engages in philosophical inquiry (conversation) do not vary. We agree that in virtue of engaging in an activity (for example, philosophical inquiry) each participant has the same reason to abide by the rules constitutive of that activity. Nonetheless, the reasons to abide by those rules more or less perfectly do vary with the value (for example, significance) of the particular instance of that activity. Just as engineers have less reason to diligently error proof their designs when working on videogame gadgets than when working on vehicles for human transportation, so too a philosopher has less reason to make all the premises explicit and to defend those premises in great detail when the topic is less worthwhile. In the case of the strength of reasons to have regard for those with whom one converses (for example, reasons to take genuine interest in what they say), these might not vary with the interlocutors’ differing degrees of rational capacities (assuming they meet the relevant threshold). But this would be
Agnieszka Jaworska and Julie Tannenbaum
217
Thus far then, our best candidates for a general principle which could justify the Continuity View must be abandoned. Of course, a better principle might yet be found.
6. Conclusion In this chapter, we have distinguished different types of moral equality and discussed difficulties with justifying the equal moral status of babies, severely permanently cognitively impaired humans, and adult cognitively unimpaired humans while also defending the common-sense intuition that these human beings have higher moral status than most animals. Further, we argued, via a series of counter-examples, that the Continuity View of how degrees of moral status map onto their grounds cannot be backed by the various general principles we considered. This evens out the playing field between the proponents and the opponents of the Continuity View. It becomes at least an open question whether moral status is more akin to blameworthiness, reasons to promote a sport, and the right to drive, or whether it varies with the degree of one’s capacity in the way proposed by the Continuity View. The Continuity View is no longer the default position, and the Discrete Interval View remains a viable option.³⁵
References Anderson, E. 1999. ‘What is the Point of Equality?’. Ethics, 109: 287. Arneson, R. 1999. ‘What, If Anything, Renders All Humans Morally Equal?’. In Singer and His Critics, edited by Dale Jamieson, 103–127. Oxford: Blackwell. Arneson, R. 2015. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. due to their equal moral status #2 (which is at issue and so cannot be appealed to). Miklosi also claims that ‘the reason for each rational being to make that one life [i.e., their own life] valuable has the same strength, regardless of how valuable the most valuable life that is available for each of them actually is’ (383), which Miklosi takes to imply that the reasons other agents have to protect each being are also equal in strength. However, if A’s life were more valuable than B’s life, then it would be a graver mistake for A to fail to respond to the reasons A has to make A’s life valuable than for B to fail to respond to the reasons that B has to make B’s life valuable, and so A has stronger reason than B does to avoid the respective mistake. Also, while it might seem as if A and B face the decision to live a life that has value versus not to live such a life, in some cases they face this decision: to protect one’s own life and let the other die versus protect the other’s life and let oneself die. One cannot beg the question by assuming that the value of each of the two lives does not affect the strength of reasons they have in making the latter type of decision. ³⁵ We would like to thank Giacomo Floris, Shmuel Gomes, Joseph Bernardoni, the students in the 2022 graduate seminar ‘Moral Status of Human Beings’ at University of California, Riverside, and the members of the 25 November 2022 online workshop ‘Moral Status and Basic Equality’ for very helpful comments on earlier drafts.
218
Equality and Moral Status
Benn, S. 1967. ‘Egalitarianism and Equal Consideration of Interests’. In Nomos IX: Equality, edited by J. Roland Pennock and John W. Chapman, 61–78. New York: Atherton Press. Broome, J. 1990–1991. ‘Fairness’. Proceedings of the Aristotelian Society, 91: 87. Carter, I. 2011. ‘Respect and the Basis of Equality’. Ethics, 121: 538. Dworkin, R. 1977. Taking Rights Seriously. Cambridge, MA: Harvard University Press. Feinberg, J. 1980. ‘Abortion’. In Matters of Life and Death, edited by Tom Regan, 183–217. Philadelphia: Temple University Press. Harman, E. 2004. ‘Can We Harm and Benefit in Creating?’. Philosophical Perspectives, 18: 89. Hill Jr., T. E. 1991. ‘Self-Regarding Suicide: A Modified Kantian View’. In Autonomy and Self-Respect, 85–103. Cambridge: Cambridge University Press. Hill Jr., T. E. 2000. Respect, Pluralism, and Justice: Kantian Perspectives. Oxford: Oxford University Press. Jaworska, A. 2007. ‘Caring and Full Moral Standing’. Ethics, 117: 460. Jaworska, A., and Tannenbaum, J. 2014. ‘Person-Rearing Relationships as a Key to Higher Moral Status’. Ethics, 124: 242. Jaworska, A., and Tannenbaum, J. 2015. ‘Who Has the Capacity to Participate as a Rearee in a Person-Rearing Relationship?’. Ethics 125: 1096. Jaworska, A., and Tannenbaum, J. 2023. ‘The Grounds of Moral Status’. In Stanford Encyclopedia of Philosophy, Fall 2023 ed., edited by Edward N. Zalta and Uri Nodelman, URL = Korsgaard, C. M. 1996. ‘Kant’s Formula of Humanity’. In Creating the Kingdom of Ends, 106–132. Cambridge: Cambridge University Press. McMahan, J. 2002. The Ethics of Killing: Problems at the Margins of Life. Oxford: Oxford University Press. Miklosi, Z. 2022. ‘The Problem of Equal Moral Status’. Politics, Philosophy & Economics, 21: 372. Quinn, W. 1984. ‘Abortion: Identity and Loss’. Philosophy & Public Affairs, 13: 24. Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Regan, T. 2004. The Case for Animal Rights. Berkeley and Los Angeles: University of California Press. Sangiovanni, A. 2017. Humanity without Dignity: Moral Equality, Respect, and Human Rights. Cambridge MA: Harvard University Press. Scheffler, S. 2003. ‘What is Egalitarianism?’. Philosophy & Public Affairs, 31: 5. Singer, P. 1975. Animal Liberation: A New Ethics for Our Treatment of Animals, 2nd ed. New York: New York Review/Random House,
Agnieszka Jaworska and Julie Tannenbaum
219
Strawson, P. 1962. ‘Freedom and Resentment’. Proceedings of the British Academy, 48: 187. Tooley, M. 1972. ‘Abortion and Infanticide’. Philosophy & Public Affairs, 2: 37. Velleman, J. D. 1999. ‘A Right of Self-Termination?’. Ethics, 109: 606. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: Harvard University Press. Wood, A. W. 1998. ‘Kant on Duties Regarding Nonrational Nature’. Proceedings of the Aristotelian Society, 72: 189.
9 When Equality Needs no Justification Anne Phillips
Why should we think of others as our equals, and how did this astonishing idea get going? It is, in one sense, an idea with a long history, and people often cite the early Christian notion of equality in the sight of God as an important forerunner. But as Teresa Bejan (2022 and this volume) has argued, that early notion of equality was a kind of equality-as-indifference, indifference, that is, from the perspective of the deity. God is indifferent to the multiplicity of social, economic, physical, and intellectual distinctions on which we may pride ourselves. Outside the more intolerant versions of Christianity, it is indifferent also to a past history of sinfulness and error, for what matters to God is one’s current spiritual state. This is the essence, I take it, of the story of Mary Magdalene, depicted through centuries as the sinful woman who had previously made her living as a prostitute, but saved by Jesus’ teachings and becoming one of his most faithful followers. In doing so, she becomes far better than her previous self and far better than the many who deny him. Equality in the sight of God signifies an indifference to the unimportant things, but the very point of the indifference is precisely that God will then judge us, and rank us, according to our ‘true’ worth. It is a route to moral ranking, not an assertion of ‘moral’—or any other—equality. That conception of equality was still at work in the arguments Bartolomé de las Casas employed in the 1550s to challenge the Spanish treatment of the indigenous peoples of the Americas, and specifically, to challenge the idea that they were barbarians who could therefore be legitimately enslaved. It was, he argued, an ‘irreverence’ towards God to ‘write that countless numbers of natives across the ocean are barbarous, savage, uncivilized, and slow witted’ (de las Casas 1992: 33–34). Las Casas accepted, in principle, Aristotle’s argument that some individuals might be so barbarous by nature that one could describe them as ‘natural’ slaves, but he refused the suggestion that this could ever be applied to an entire people. In his understanding, the key distinction between humans was the degree to which they approached a state of spiritual perfection, and in this regard he saw no intrinsic difference Anne Phillips, When Equality Needs no Justification. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0010
222
When Equality Needs no Justification
between the often brutal Spanish and the not yet Christian indigenous inhabitants. His view was, however, on the wane. As Sylvia Wynter (2003: 287) has put it, ‘the medieval world’s idea of order as based upon degrees of spiritual perfection/imperfection, an idea of order centred on the Church, was now to be replaced by a new one based upon degrees of rational perfection/imperfection’. The new ranking brought intrinsic difference much more to the fore, and in the Spanish Americas that difference was increasingly framed as a matter of ancestry and skin colour. By the eighteenth and nineteenth centuries, this had spawned a classification of peoples into (at least) sixteen distinct categories; and a genre of casta paintings that depicted the different types of people according to their mixture of European, African, and Indian ‘blood’. This kind of differentiation is the important other face of the natural equality that gradually displaced equality-in-the-sight-of-God from the seventeenth century onwards. In the European context at least, the facts of nature increasingly came to figure as the justification for human equality, but they figured at the same time as the justification for gradation. Men were depicted (and it was men) as rough equals in a pre-political state of nature; as not sufficiently differentiated in their physical or intellectual capacities to deliver any clear reason why some of them should rule and others be ruled; and by the eighteenth century were being depicted as not only born free and equal but remaining so. As deployed in these arguments, ‘nature’ suggests a potentially progressive contrast with culture. It grounds the claim to equality in features of our natural lives that may coexist with immense social difference, but that all can nonetheless be said to share. Part of the trickiness, however, of appeals to nature is their tendency to fudge the supposed clarity of that distinction with culture, often projecting onto nature differences that are social in origin, or simply conjuring up differences and describing them as differences of nature. Think of the way women became identified as, by nature, more sentimental and less rational than men, as if the constraining of women’s activities or their confinement to a domestic sphere were just a sensible response to ‘natural’ difference, an effect rather than in any way a cause. Think also of the division of humans into hierarchically ordered and supposedly discrete ‘races’, and the way this process of racialization layered presumed capacities of a psychological, intellectual, emotional, and moral character onto visible (‘natural’) features of physiognomy and skin colour.¹
¹ For further elaboration of the notion of ‘race’, see Appiah (2005); Gilroy (2000); Malik (2008).
Anne Phillips
223
When supposed facts of nature become the basis for equality claims, this simultaneously creates alibis for deeming all too many people natural inferiors, for if it is ‘nature’ that makes us equals, then perceived differences of nature can also justify inequality. In contrast, moreover, to the early Christian version where everyone can in principle change, transforming themselves from bad people to good or good people to bad, nature stamps us more irrevocably. Equality becomes more indelibly attached to possession of those properties now regarded as characteristic of or essential to human beings, and the question of who defines the characteristics becomes crucial. As history has demonstrated, those providing the definitions typically conjured up prototypes based on themselves. Centuries of proclamation about ‘our’ natural equality were then characterized by an extraordinary capacity to employ a seemingly universal language yet set aside—in many cases not even notice—the half of humanity who are female, the millions who were enslaved, the hundreds of millions who were subjected to colonial rule. Think of Rousseau, who saw the ‘natural’ inequalities of health, bodily strength, or mental agility as minimal by comparison with the gross social inequalities that came into being with private property, but went on to represent the ‘natural’ differences between the sexes as so extensive that they became a reason to treat men and women almost as species apart.² Think of Kant, who combined a seemingly universal categorical imperative addressed to all humans with a racial taxonomy that expressed contempt for the lesser races and justified their enslavement.³ Think of John Stuart Mill, staunch supporter of women’s equality, staunch critic of slavery, but also defending colonialism as necessary for ‘backward’ peoples. Devin Vartija (2021) has traced part of this story through his study of the eighteenth-century French encyclopaedists, arguing that their fascination with nature simultaneously enabled their social critique and entrenched racial classification. A shared human nature became the key grounding for arguments about human equality; but the very focus on humans as natural beings, and thereby open to scientific investigation, also encouraged the racial taxonomies that classified them according to their physical and presumed moral characteristics. The ambiguities of ‘nature’ have played a major role in the history of ideas of equality.
² Including in Emile: Or on Education, where he advocates distinct forms of education for boys and girls, arguing that ‘the man should be strong and active; the woman should be weak and passive’ (Rousseau, 1792: 359). As so often in arguments about natural difference, it turns out girls and boys are both ‘naturally’ different and must be made so. ³ Towards the end of his life he revised this, and explicitly condemned both colonialism and slavery. Kleingeld (2007).
224
When Equality Needs no Justification
The main argument of this chapter is that contemporary efforts to justify equality remain too much within that original mould, confusing reasons for valuing equality with reasons for thinking of one another as equals, and imagining that the latter must be justified by reference to some property or combination of properties that human beings share. The beings in question are now all humans, not male humans, or white humans, or European humans; and in the philosophical literature at least, no one would now define the intrinsic properties in ways that allowed differences in skin colour or sex organs to be regarded as a basis for unequal treatment. Cognitive capacities do figure in parts of the philosophical literature, with some humans then downgraded as not really ‘persons’,⁴ and this is one area where the property-based account of equality has been subjected to particularly effective critique (as in Kittay, this volume). But most philosophers now treat differences in our physical or intellectual or cognitive properties as irrelevant to our status as equals, and this is reflected in the increased use of the language of ‘moral equality’ to signify our basic equality. In nineteenth-century arguments against slavery or for the enfranchisement of women, people frequently employed the phrase ‘moral and intellectual equals’, arguing that there was no basis in actual differences to justify the subordination of women to men or the enslavement of black people by white. That insistence on intellectual equality was important in challenging assumptions about the inherent inferiority of women and slaves, but it was in other ways a hostage to fortune, seeming to tie the claim to equal status to an empirical claim about intellectual capacities, and thereby leaving it open to disproof. From the 1980s onwards, the predominant language in the philosophical literature has been—more simply—one of ‘moral equals’, and I take it that this is intended to separate the equality claim more decisively from anything involving empirical evidence. Part of my argument is that it fails in this. Remaining too firmly within the justificatory paradigm, it fails adequately to capture the ways in which equality is about enactment rather than possession, about our commitments and our claims. In its emphasis on the ‘moral’ in moral equality, it brings further problems in its wake.
1. Equality as Commitment and Claim My central charge is that arguments about the basis for equality, however attenuated, still make the equality conditional, for in searching for a reason to regard others as our equals, a justification for equality, we make the equality ⁴ For example, McMahan (2002).
Anne Phillips
225
depend on something else. The ‘something else’ is commonly a property or set of properties said to be characteristic of all humans; and in the standard configuration of contemporary argument, it does not matter how much or how little of this particular property any one of us may enjoy. There is continuing debate about this last point, including in this collection. Richard Arneson (2014: 34) is on the sceptical side, and has argued that ‘either the trait will turn out not to matter or variations in it will matter’. In his chapter in this collection, he now offers a threshold account of the rational agency capacities that he sees as providing the necessary grounding for our equality claims, but he acknowledges that this throws up questions regarding the boundaries of personhood that are both politically divisive and intellectually hard to resolve. The boundary question is clearly one of the difficulties with the property-based account; and that the traits in question—moral agency, rational capacity, a sense of one’s life as a life, the capacity for empathy, and so on—are now drawn in broad-brush fashion does not of itself dissolve the problem. Even when the bar is set so much lower than when it included (or implied) characteristics common only to a minority of human beings, even when we insist that continuing variations as regards the property are irrelevant, the argument remains caught up in a logic of justification that looks to some human property to ‘ground’ the equality. The characteristics become a test. We get to be treated as equals because of this human property we are said to share. In challenging this, I argue that equality needs no such justification. We should regard equality more along the lines favoured by Hannah Arendt, in which it is understood as a commitment we make to ourselves and others to regard one another as equals. For Arendt, the commitment is not conditional on anything particular about human nature, but a condition in itself for the freedom and human action she took as necessary features of political life. She described political equality as ‘an equality of unequals who stand in need of being “equalized” in certain respects and for certain purposes’ (1958: 215), something, that is, that we do, rather than something we are. She also famously insisted that we should not confuse this commitment to political equality with any particular stance on social or economic equality. My concern is with the larger notion of our status as equals, which includes but is by no means exhausted by political equality, and I do not believe we can secure this equality of status if we insulate it from questions about the wider distribution of resources and power. My debt to Arendt therefore only goes so far. What I take from her, nonetheless, is that refusal to think of equal status as something owed to us by virtue of ‘something else’. This is the sense in which equality needs no justification. Too much contemporary work on basic equality confuses the questions we do need to
226
When Equality Needs no Justification
ask and debate—why and in what ways is it better to live in a society characterized by equality than one marked by inequality? what does it mean to treat one another as equals?—with a question we should not need to ask— why should we regard other humans as our equals? That last question puts us on the defensive, requiring us to justify to others what should not require justification. It works within the assumption, dating back to at least the seventeenth century, that equality reflects some fact about humans, something that differentiates us from animals, and thereby qualifies us for the status of equal. It is better, in my argument, to regard equality as a commitment and (here going considerably beyond Arendt) a claim: the commitment we make to regard one another as equals, and the claim we make on others who have so far failed to see us in this light. This is a political commitment, not the conclusion of a philosophical argument nor the product of an empirical investigation, and should not be understood as derivative of anything else. As partial illustration, consider Frederick Douglass’s 1852 speech on What to the Slave Is the Fourth of July? In this powerful denunciation of slavery in America, Douglass refuses to argue that slaves are also men, or that as men they are entitled to liberty, or that it is wrong ‘to flay their flesh with the lash, to load their limbs with irons, to sell them at auctions, to sunder their families, to knock out their teeth, to burn their flesh, to starve them into obedience and submission to their masters’. Refusing the work of justification, he asserts, ‘No, I will not. I have better employment for my time and strength … At a time like this, a scorching irony, not convincing argument, is needed’. This is a refusal to justify what should never require justification, in this case that the people enslaved are also human beings. We should not have to demonstrate that, regardless of sex, race, sexuality, or disability, we share common human characteristics that then justify our treatment as equals, any more than we should have to demonstrate that people living in social housing are just as human as those living in mansions. In such contexts, the very act of offering a justification seems to acknowledge that there might be some doubt, and seems then to lend itself to gradations of equality. In Humanity Without Dignity, Andrea Sangiovanni (2017: 3) also rejects the search for ‘some set of natural psychological capacities in virtue of which we all have an infinite, absolute, incommensurable worth’. The respect we owe others, he argues, is not grounded in human worth—in our so-called dignity, for example—but in human vulnerability, the vulnerability in particular to forms of inferiorizing treatment that threaten the ability to maintain an integral sense of self. He thereby shifts the focus from equality to inequality, arguing that it is the rejection of cruelty and inequality that underpins the commitment to equality, and identifying stigmatization, dehumanization,
Anne Phillips
227
infantilization, instrumentalization, and objectification as the five paradigmatic forms of inferiorizing treatment. I agree with much of what he says there: the critique of dignity as a grounding for equality; the larger critique of any property-based account; the shift of focus from equality to inequality. I also agree with much of what he argues in this collection, where he queries whether there is such a thing as moral equality, and suggests it does not matter if there isn’t. I remain unclear, however, to what extent he too still seeks a grounding for equality. In Humanity without Dignity, he does indeed argue that failing some such grounding, we are left with ‘nothing more than blind faith or political expediency’ (3); and his own version seems to ground equality in a psychological claim about what humans need to flourish and survive. All of us, he argues, need a sense of self, a sense of knowing oneself as a self, and knowing one’s life as a life rather than a disconnected series of life fragments, for without that we lose the sense of control over our lives, feel ourselves determined by events or others, and are no longer ‘ourselves’. An integral sense of self is, on this account, ‘a constituent ingredient and structural element of a flourishing life’ (83). This is not, in itself, an implausible claim, and many of us will recognize things that matter very much to us in what Sangiovanni argues, but the structure of the argument still leaves the equality conditional on a set of claims about what humans need to flourish. If some future psychological study were to establish that human beings are perfectly capable of going through life without this integral sense of self, would we then have to abandon our objections to inferiorizing treatment? Or even more troubling, what if the study established that certain kinds of human had less need of this integral sense of self than others? Would we then be justified in saying that they did not require the same level of protection from inferiorizing treatment? Presumably not— and it is indeed clear from Sangiovanni’s arguments that he would reject any such conclusion. My own view is that this means refusing the search for grounds. Refusing the search for grounds does not force one back on blind faith or what Jeremy Waldron (2017: 85–86) has described as simply the ‘decision’ to treat others as equals. Any such description threatens to reduce both commitment and claim to something almost entirely personal—my belief, my decision—as if equality claims have no history, no connection with the multiple instances of sometimes successful, often not, resistance to hierarchies and inequalities, no wider significance than what takes place inside one’s own mind. The language of ‘decision’ abstracts equality from that history, turning it into something more arbitrary and subjective. It seems particularly inappropriate, moreover, as regards the action of claiming to be regarded as an
228
When Equality Needs no Justification
equal, for while one might well make a decision about when and in what circumstances to stake one’s claim, the claim itself surely precedes that. We do not ‘decide’ to regard ourselves as equals, but more simply do so. Indeed, this is one of the reasons why I stress equality as both commitment and claim. A focus on commitment alone might well encourage us to ponder why we should make that commitment, why we should treat others as our equals, for it frames the question as if from the standpoint of those who are already reasonably secure in their own status as equals, but are searching for convincing grounds to extend this status to others. The standpoint of those caught up in the midst of a struggle to claim their equality is somewhat different: they are unlikely to exhibit any doubt as to the legitimacy of their claim to be regarded as equals; they just experience a lot of difficulty in getting those who currently deny it to come round to the same point of view. They might, certainly, be thinking about how best to make an effective argument against the deniers, whether to expose their contradictions, play on their sympathies, or challenge them to explain what makes them so special. Or they might, like Douglass, say one has better use for one’s energy and time. While I share much of Douglass’s impatience with those who still need arguments for human equality, I am not proposing that we should refuse to engage with the deniers. My point, rather, is that we should not mistake any of the arguments we might then wield for justifications, as if the case for seeing people as equals could plausibly fail. Arguments can and do fail, either because we make a poor one or because we lack the skill to make a good one convincing, but no amount of argument (good, bad, or indifferent) should work to convince people that they are not after all equals. The problem with ‘grounding’ equality is that any ground can potentially be swept away. We should have no time for this possibility when it comes to the equal status of all human beings. Some will feel that refusing a justificatory grounding for equality arbitrarily draws a line around humans, declaring them alone as qualifying for this status, yet refusing to give any more substantial grounds for this than mere membership of the human species. But species membership is not that arbitrary (see also Kittay, this volume), and there are good reasons to think of equality as a specifically human innovation and commitment. There are nonhuman animals that engage in highly solidaristic practices; there are nonhuman animals that exhibit qualities of empathy, care, or grieving; and there are many species of nonhuman animals that engage in the kind of strategic planning that used to be claimed as something only humans could do. But while one can make sense of the argument that the world would be a
Anne Phillips
229
better place if all humans treated all other humans as their equals, it would be both arrogant and meaningless to propose equality as the ruling principle for the lives of nonhuman animals. Imagine explaining that to the queen bee and her worker drones, or to the many species of animals that live by preying on others. Our obligations to nonhuman animals should not be based on how similar they are to us, nor should they involve any expectation that nonhuman species come to recognize the value of equality. There is a legitimate distinction to be made between human and nonhuman animals. Drawing that line says nothing about one species being superior to another, and requires nothing more substantial than the very basic recognition that all humans are indeed humans.⁵
2. From Moral Equality to Moral Judgement The language of moral equality is supposed to diminish the problems highlighted above, reducing the qualifying criteria to such a basic level that no one will now be excluded and no one refused their equality. That people continue to debate exactly what are the qualifying criteria suggests that this is far from the case. The problems, moreover, are in some ways exacerbated, for the current language of moral equality introduces at least three further problems. First, it encourages a slippage from arguments about equality to judgements of moral worth, substituting questions about who is a ‘good enough’ human for previous questions about who was human at all. It also enables a distinction between the moral and the material that misrepresents the first stage of equality—the so-called moral stage—as already complete. In making that distinction between moral and material, it further understates the crucial dependence of our status as equals on the reduction of material inequalities, offering us equality on the cheap. Philosophers are, of course, well aware that to talk of moral equality, particularly in association with notions like dignity, worth, or respect, is to engage in a highly moralized language that is, in other contexts, employed to differentiate higher from lower, better from worse, greater from lesser. We use the same language to elaborate an idea of something supposedly attached to all of us without exception—our human rights, our status as moral equals— and to capture something that involves grades of excellence. When deployed ⁵ I discuss the status of nonhuman animals at greater length in Phillips (2021). See also Sangiovanni (this volume) for an argument that the status of nonhuman animals is not best understood as a question about equality.
230
When Equality Needs no Justification
in the second sense, we refer to something people can and do lose. People commit acts that make them unworthy of our respect; they behave in ways that undermine their dignity; they do things that lead us to regard them as more or less moral than their neighbours. Of itself, dual usage is not particularly unusual: the same terms often pop up in different contexts to mean different things, and the dual nature of notions like dignity and respect has been widely recognized in the literature. Jeremy Waldron (2015: 33) notes the early deployment of dignity in a pre-egalitarian discourse that attached ‘dignities’ to hierarchies of rank and social standing; but goes on to argue that the ‘modern notion of human dignity involves an upward equalization of rank, so that we now try to accord to every human being something of the dignity, rank, and expectation of respect that was formerly accorded to nobility’. It is not, however, so easy to eradicate the effects of that prior history, and while philosophical usage has shifted towards the unconditional attribution of dignity to all, popular usage remains considerably more varied. Philosophers commonly address this kind of slippage between meanings by introducing additional distinctions. This, in effect, is Stephen Darwall’s solution (1977) as regards the potential confusion around respect: he adopts the term ‘appraisal respect’ for the kind of respect we feel for those who have excelled in some way; and distinguishes this from the ‘recognition respect’ owed to all of us by virtue of our shared humanity. As a distinction this is clear enough, but it hardly solves the problem. It is not easy to convince people that everyone is equally worthy of Darwall’s recognition respect, when in the vast majority of contexts where we employ the term it is to mark out people we find especially worthy of respect. When Uwe Steinhoff, for example, argues against moral equality, this is precisely what he draws on. He offers vignettes illustrating different degrees of moral worth, with the cruel rapist and his innocent victim figuring particularly large, and concludes from his various examples that a self-evident difference in moral worthiness makes a nonsense of claims about us being ‘moral equals’. The fact of different degrees of moral demerit on the one hand and moral merit on the other fatally undermines egalitarianism, and not only ʻin extreme casesʼ but in ordinary life. If degrees matter, not all persons or [even] all persons who are not murderers or rapists are moral equals; instead, only persons who are moral equals are moral equals. But that is trivial; and it isnʼt egalitarian. (2014: 147)
Anne Phillips
231
We can point out that this is a misunderstanding of what is intended by ‘moral equal’; that Steinhoff gets it wrong when he thinks moral equality means people are equally moral; that he needs to think more along the lines of Darwall’s distinction between appraisal and recognition respect. But his is an important and common misunderstanding, and I take the argument as a particularly revealing illustration of the confusions introduced when a term heavily loaded with comparative evaluation is also deployed to signify a supposedly unconditional equality. If philosophers can fall into this confusion, how much more likely is it that those less trained in careful distinction will do the same? People in general and governments in particular have, as it turns out, been remarkably willing to withdraw what one might have considered basic human rights when people are deemed to fall short of some presumed standard of moral worth. In the UK, convicted prisoners lose their right to vote, even when imprisoned for short periods and not especially heinous crimes; in some states of the USA, they lose this right not just while in prison, but for life.⁶ In such instances, it seems that what has been widely regarded as a basic expression of human equality—the equal right to vote regardless of wealth, gender, race, religion or sexuality—is being made conditional, conditional in this case not so much on being demonstrably human, but on being a ‘good’ human, on being the ‘right kind’ of human. In the wake of the attack on the twin towers, a disturbing number of people, including leading jurisprudence experts, defended the use of torture on people suspected of planning terror attacks.⁷ Over the same period, the targeted killing by drones of people presumed to be terrorists—very often also of their family and entourage—became almost thoroughly normalized in the USA. The practice is still contested, but governments no longer have to hide that they do it, or pretend that their real aim was always to capture and bring to trial. In these instances, substantive moral judgement of our worth is being taken as a reason to deny, or at least attenuate, a previously assumed equal moral status. What seemed axiomatic and unconditional turns out to depend all too significantly on whether one is deemed ‘good’ or ‘bad’. The philosophical preoccupation with what it is about human beings that justifies our equal moral status helps contribute to this way of thinking. In Waldron’s argument, the very seriousness of claims about human equality points to the requirement for a serious and substantial grounding, and he ⁶ Prisoners’ voting rights—House of Commons Library (parliament.uk); Felon Voting Rights (ncsl.org). ⁷ See Waldron (2012) for a powerful critique of this.
232
When Equality Needs no Justification
finds this in notions of human dignity and worth, and what he sees as the astonishing capacities that all individuals possess—whether we exercise them or not—for moral agency, for thinking about what is right and wrong, and trying to make this reflection part of our actions. ‘We are sometimes just blown away by the sheer capacity that people have for the self-application of norms’ (2017: 166). There is a wonderful generosity in that statement, with its refusal of the racialized and classed and gendered stereotypes that stop so many people seeing how widespread these capacities are; and it is in no way Waldron’s intention that his claims about our wondrous human capacities should enable more invidious distinctions. It is, nonetheless, an expression of faith (there is more blind faith here, one might suggest, than in what I am trying to persuade you towards), and as such, it opens the door for those who do not accept that we all reach the necessary minimum of these capacities.
3. The Moral/Material Distinction The second and third worries about the contemporary language of moral equality are closely related: first, that the implied distinction between moral and material encourages a stages view of equality that takes the first phase as more or less complete; secondly, that it understates the extent to which securing the status as ‘moral’ equals depends on reducing material inequalities. There is a marked tendency—marked among political philosophers but also among politicians—to assume that we are all now recognized as moral equals, at least in those liberal democracies that so proudly proclaim equality as one of their core values. Thomas Scanlon (2018: 4), as one exemplar among many, claims that ‘basic moral equality is now widely accepted, even among people who reject substantive egalitarian claims’. The implication is that moral equality is a stand-alone kind of equality (perhaps rather like Arendt’s political equality) that could be fully accomplished even when substantive questions about the distribution of resources remain unaddressed. This suggests a theory of stages, and conveys an overly complacent perception of basic equality as achievable even before the society embarks on the ‘next’ stage. That this view is overly complacent is readily illustrated by the continuing evidence of racism, misogyny, and homophobia, not to mention pervasive class contempt. The most plausible sense in which one could claim a ‘wide acceptance of basic moral equality’ is if one took this to refer more minimally to the equal right to vote. But even this is only accepted within democracies; it only arrived as full adult suffrage in the course of the last
Anne Phillips
233
century; and it remains under threat in a number of jurisdictions (including the UK and USA), not only by restrictions on prisoners’ right to vote, but by initiatives to curtail voter registration or access. The language of moral equality encourages us to think that, whatever our disagreements about other aspects of equality, we are at least agreed on this. This seems very far from the truth. The assumption that moral equality is both a relatively clear notion and more or less achieved underpinned much of the ‘equality of what?’ debates that dominated the egalitarian literature for many years. It was taken as more or less axiomatic that equality mattered, and that people are to be regarded as equals in some basic sense, and the main focus of egalitarian debate was on what kind of distribution, and of what kind of goods, the principle of equality then implied. What exactly should an egalitarian be seeking to equalize: was it resources, welfare, capabilities, or some other candidate? That focus on matters of distribution has since been challenged, and what is described as relational egalitarianism has turned attention more decisively to what it means to stand in relations of equality with others.⁸ One of the strengths of this alternative is that it recalls us to a recognition that the modern world is not characterized by a near universal acceptance of moral equality, that, to the contrary, people face racism, misogyny, and homophobia, and have to battle long and hard to establish themselves as equals. Like many others, I have warmed to the relational repositioning of debate, and the renewed emphasis on challenging the power relations that represent some people as superior and others as inferior. I also warm to the fact that relational egalitarians tend to work with a language of equality rather than ‘moral equality’, signalling, I think, the refusal of a stages account in which the first step towards equality is considered as already accomplished. But as elaborated, for example, by Elizabeth Anderson (2012: 41), relational egalitarianism tends to consider access to resources as only instrumentally necessary for us to be able to live as equals, and no longer considers inequality in the distribution of resources and wealth as an intrinsic object of concern. It is not that the distribution of goods disappears from the picture, but equality in social relations becomes the touchstone for what matters, and the particular distribution of goods is treated as relevant only as ‘conditions for or consequences of this’. This still suggests an overly sharp separation between ⁸ See especially Anderson (1999; 2012), and Scheffler (2003; 2005). The critique of the distributional paradigm owes much to Young 1990.
234
When Equality Needs no Justification
one’s status as a moral equal and one’s access to the resources that enable one to function as such, a separation perhaps heightened by the desire to repudiate what was perceived as the exaggerated focus on distribution in some earlier traditions of egalitarianism. It still insulates what it means to be regarded as an equal more than is appropriate from the distribution of resources and wealth. I noted earlier my reservations about Arendt’s insistence on reserving equality for the political sphere, and her deliberately deflationary understanding of political equality as ‘an equality of unequals who stand in need of being “equalized” in certain respects and for certain purposes’. What matters, in her understanding, is simply the commitment to regard one’s fellow citizens as equals and to engage with them as such, and she saw it as a dangerous perversion of political equality—a reduction of politics to administration, an abdication of specifically political responsibility—to translate this into a matter of resource distribution. I share her focus on equality as enacted rather than given, as commitment rather than natural fact, but I do not follow her in this. Nor do I follow what I take to be Anderson’s view, that we need only a sufficiency of resources, combined with various guarantees of our working conditions, in order to sustain equality of standing. Even if every citizen were ensured a sufficiency of resources—a state of affairs not achieved anywhere in the world—marked inequalities in resources and wealth would continue to engender delusions of superiority, for living in a world of economic inequalities erodes the ability to see others as equals. This is even more the case when richer and poorer live, as they increasingly do, virtually segregated lives. There is a certain kind of class contempt, dismissal of others, that pervades contemporary social interactions, ranging from the derision of the comfortably positioned professional middle classes for those dependent on social security—as depicted for example in Jones (2011)—to Hillary Clinton’s no doubt much regretted phrase about Trump supporters being a ‘basket of deplorables’. My hunch is that these assumptions of superiority would survive even if everyone had the resources necessary to ‘appear in public without shame’— a phrase Anderson (1999) adapts from Adam Smith to capture some of what she means by social equality—for the failure to rise higher would still be taken as evidence of personal inadequacy, and the continuing separation in life experiences would still block perceptions of one another as equals. This is an empirical rather than a logical claim, and in suggesting that inequality in resources, even beyond the sufficiency level, can still undermine equality of status, I implicitly accept that the case for equalizing
Anne Phillips
235
resources is primarily instrumental, rather than a matter of principle. My point, nonetheless, is that we cannot usefully think of equality of status in isolation from equality of resources. One of the worries about the language of moral equality is that it encourages an overly cerebral, insufficiently material understanding of the wrongness of inequality. We would do better just to talk of equality instead. To sum up, justifications of our moral equality fail on a number of counts. They fail, most importantly, because they draw us into a logic of justification that remains bound to the exclusionary practices of the past, puts us on the defensive as regards our status as equals, and leaves open the possibility that our claims to this status might fail. When pursued, more specifically, in the language of moral equality, they encourage a potentially complacent stages account of the progress of equality that overstates the current acceptance of ‘basic’ equality, and understates the extent to which securing the status as equals will depend on reducing material inequalities. They also encourage a slippage into precisely the kind of moral differentiation that an affirmation of our moral equality is supposed to prevent. Talk of moral worth or moral status almost inevitably opens up the possibility that some people have greater moral worth than others, and attaching the qualifier ‘equal’ to these phrases merely begs the question ‘why?’ So long as we continue to engage with equality as if it were a matter of facts or properties about human beings, it will continue to seem a matter of ‘blind faith’ to say that all are equal in respect of these facts or properties. This remains the case even when the properties are drawn in a broad-brush fashion that supposedly includes us all. This not what equality is or should be. Equality is about what we do, not what we are. It is what we make happen when we commit to and claim equality, something we bring into existence, not something we find out.
References Anderson, Elizabeth S. 1999. ‘What is the Point of Equality?’ Ethics, 109/2: 287–337. Anderson, Elizabeth S. 2012. ‘Equality’. In The Oxford Handbook of Political Philosophy, edited by David Estlund, 40–57. Oxford: Oxford University Press. Appiah, Kwame Anthony. 2005. The Ethics of Identity Princeton: Princeton University Press. Arendt, Hannah. 1958. The Human Condition. Chicago: University of Chicago Press.
236
When Equality Needs no Justification
Arneson, Richard. 2014. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Bejan, Teresa. 2022. ‘What Was the Point of Equality?’ American Journal of Political Science, 66/3: 604–616. de Las Casas, Bartolomé. 1992. In Defence of the Indian. DeKalb, Illinois: Northern Illinois University Press. Darwall, Stephen. 1977. ‘Two Kinds of Respect’. Ethics 88/1: 36–47. Douglass, Frederick. 1852. ‘What to the Slave is the Fourth of July?’ Reproduced in Frederick Douglass: Selected Speeches and Writings, edited by Philip S Foner, 188–206. Chicago: Chicago University Press, 1999. Gilroy, Paul. 2000. Against Race. Cambridge, MA: Harvard University Press. Jones, Owen. 2011. Chavs: The Demonization of the Working Class. London: Verso. Kleingeld, Pauline. 2007. ‘Kant’s Second Thoughts on Race’. The Philosophical Quarterly, 57/229: 573–592. Malik, Kenan. 2008. Strange Fruit: Why Both Sides are Wrong in the Race Debate. London: OneWorld Publications. McMahan, Jeff. 2002. The Ethics of Killing: Problems at the Margins of Life. Oxford: Oxford University Press. Phillips, Anne. 2021. Unconditional Equals. Princeton: Princeton University Press. Rousseau, Jean-Jacques. 1792. Emile, Or, On Education. New York: Basic Books, 1979. Sangiovanni, Andrea. 2017. Humanity Without Dignity: Moral Equality, Respect, and Human Rights. Cambridge MA: Harvard University Press. Scanlon, T. M. 2018. Why Does Inequality Matter? Oxford: Oxford University Press. Scheffler, Samuel. 2003. ‘What is Egalitarianism?’. Philosophy and Public Affairs, 31/1: 5–39. Scheffler, Samuel. 2005. ‘Choice, Circumstance, and the Value of Equality’. Politics, Philosophy, and Economics, 4/1: 5–28. Steinhoff, Uwe. 2014. ‘Against Equal Respect and Concern, Equal Rights, and Egalitarian Impartiality’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 142–172. Oxford: Oxford University Press. Vartija, Devin. 2021. The Color of Equality. Philadelphia: University of Pennsylvania Press. Waldron, Jeremy. 2012. Torture, Terror and Trade-offs: Philosophy for the White House. Oxford: Oxford University Press. Waldron, Jeremy. 2015. Dignity, Rank, and Rights. Oxford: Oxford University Press. Waldron, Jeremy. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge MA: Harvard University Press.
Anne Phillips
237
Wynter, Sylvia. 2003. ‘Unsettling the Coloniality of Being/Power/Truth /Freedom: Towards the Human, After Man, its Overrepresentation—An Argument’. The New Centennial Review, 3/3: 257–337. Young, Iris Marion. 1990. Justice and the Politics of Difference. Princeton: Princeton University Press.
PART III
SCOPE Inclusion and Marginalization
10 The Basis of Childrenʼs Moral Equality Giacomo Floris
1. Introduction In recent years, moral and political philosophers have paid increasing attention to the question of the basis of moral equality, namely, the question of what makes human beings fundamentally each other’s moral equals, such that it is appropriate to consider and treat them accordingly. However, much of the literature has focused on working out a theory of the basis of persons’ moral equality—typically understood as fully competent adults who hold sophisticated cognitive capacities¹—whereas less has been said about what grounds the equal moral status of those human beings who are not fully competent adults, such as severely cognitively disabled human beings and children.² The aim of this chapter is to contribute to filling this gap by providing a novel theory of the basis of children’s moral equality with respect to one another.³ The chapter is structured as follows. In Section 2, I argue that the basis of moral equality is to be found in a fitting, basic, and independent moral attitude which provides a coherent and plausible explanation for why the variations in the degree to which the status-conferring property (or properties) is held above the threshold for moral status do not matter. Next, I offer a critical discussion of what is arguably the most influential attitude-based account of the basis of moral equality: Ian Carter’s (2011) opacity respect
¹ See, for example, Carter (2011), Christiano (2015), Miklosi (2022), and Rawls (1971). ² For notable exceptions, see Kittay (2005; this volume: 261–286), Jaworska and Tannenbaum (2014), and Waldron (2017). ³ Two points are worth noticing. First, for the purposes of this chapter ‘children’ refers to those nondisabled human beings who do not possess the relevant agential capacities up to the threshold for moral personality. For reasons that will become clearer below, my argument does not apply to those children who are severely cognitively disabled and, as such, do not have the potential to become moral persons. Second, it is important to observe that the question of the equality of moral status among children themselves is different and independent from the question of the equality of moral status between children and adults, for the former neither amounts to, nor entails, the latter. In this chapter, I focus only on the former; I address the latter in Floris (2023a), and Floris and Spotorno (forthcoming). Giacomo Floris, The Basis of Children’s Moral Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0011
242
The Basis of Childrenʼs Moral Equality
view. I argue that while Carter’s view provides a plausible explanation for persons’ moral equality, it does not have the theoretical resources to account for children’s equal moral status. This is because either opacity respect is not a fitting response to children’s moral status, or it is not a basic requirement of what is owed to them qua moral status-holders. Instead, I contend that the basis of children’s moral equality is to be found in another kind of respect— what I call positive respect—which requires providing children with those social conditions that facilitate the acquisition of a proper and stable capacity for moral personality (Rawls, 1971: 505). In brief, this is because unequal consideration and treatment is a significant obstacle to the provision of some goods that are necessary to obtain a proper and stable capacity for moral personality; hence, it is incompatible with what is owed to children as a matter of respect for their moral status. The positive-respect-based argument for a positive conception of children’s moral equality—which does not only demand the avoidance of inferiorizing consideration and treatment of children,⁴ but requires considering and treating them as equals—is elaborated in Sections 3 and 4. In Section 3, I argue that moral inferiority is detrimental to the development and maintenance of a robust sense of self-respect, the possession of which is a necessary precondition for holding the capacity for moral personality. Therefore, the avoidance of moral inferiority is a fundamental social base of a child’s sense of selfrespect. In Section 4, I contend that moral superiority is also a significant obstacle to the cultivation of a proper and stable capacity for moral personality because it precludes, or at least obstructs, children’s access to those social conditions that favour the formation of integrated friendships, which are essential to their developmental process into well-adjusted adults capable of moral personality. In Section 5, I conclude by addressing two objections so as to clarify and strengthen the argument.
2. Moral Equality, Respect, and Children Hardly anyone denies that (nearly) all human beings have equal moral status and therefore are entitled to equal consideration and treatment. But if human beings ought to be considered and treated as equals, this must be because there is something about human beings which makes them equals in some fundamental sense. However, as many have observed, when we look at the valuable properties that are shared by human beings—such as the capacity ⁴ For a prominent example of a negative conception of moral equality, see Sangiovanni (2017: Ch. 2). For a critique of Sangiovanni’s negative conception of moral equality, see Floris (2020: 413–414). See also footnote 18.
Giacomo Floris
243
for rational agency (Christiano, this volume: 149–173; Kant 2002[1785]) or the possession of a subjective perspective (Sher 2015)—they appear to be held to unequal degrees. Human beings, in fact, are more or less capable of acting rationally, and some have a richer subjective perspective on the world than others. This then generates the following problem: if human beings’ moral status is based on the possession of a status-conferring property X, then it seems reasonable to maintain that the degree of their moral status should vary according to the degree to which they hold X. But if this is true, then why do those human beings who possess X to a higher, or lower, degree not have a superior, or inferior, moral status? This is known as the ‘variations objection’ or the ‘variation problem’ (Floris 2019; Kirby and Floris this volume: 1–32).⁵ To overcome this challenge, we need a principled and compelling explanation for why the variations in the degree to which the status-conferring property is held above the relevant threshold do not generate differences in degrees of moral status. And, as I argue elsewhere, such an explanation is to be found in the proper way of valuing the valuable property that grounds moral status. More precisely, the basis of moral equality lies in a (i) fitting, (ii) basic, and (iii) independent moral attitude owed to some beings qua moral statusholders, which offers a coherent and plausible rationale for why what matters is that they hold some valuable properties within a certain range, regardless of the unequal degree to which these properties are held above the threshold for moral status.⁶ I have called this the attitude-based account of the basis of moral equality (Floris 2023b). My aim here is to develop an attitude-based account of the basis of children’s moral equality. To do this, it will be helpful to start by analysing what is arguably the most influential attitude-based view of the basis of moral equality: Ian Carter’s opacity respect view.⁷ According to Carter, moral persons—that is, individuals who hold the capacity for a conception of the good and the capacity of a sense of justice up to a sufficient minimum— are entitled to opacity respect. This kind of respect requires avoiding to ‘look inside persons’, thereby preventing us ‘from evaluating any of the variable capacities on which moral personality supervenes’ (Carter 2011: 551; emphasis in the original). In other words, respect for persons consists in refraining from taking into account the degree to which they hold the subvenient agential capacities above the threshold for moral personality. Therefore, opacity ⁵ For a different approach to the question of moral equality which grounds human beings’ equal moral status not in the equal possession of a status-conferring property, but in the wrongness of treating others as inferiors, see Sangiovanni (2017). For a critique of this relation-first approach, see Floris (2019). ⁶ The idea of the range property as the basis of moral equality was first developed by Rawls. See Rawls (1971: 505–512). ⁷ For another account that grounds moral equality in the appropriate response to the status-conferring property, see Miklosi (2022).
244
The Basis of Childrenʼs Moral Equality
respect is an independent moral requirement that explains why what matters is that persons hold some agential capacities within the range of moral personality, regardless of the different degrees to which these capacities are possessed above the threshold for moral personality. Accordingly, persons’ moral equality is entailed by a commitment to a particular kind of respect that is owed to them qua moral persons. In my view, Carter’s theory offers a plausible explanation for persons’ equal moral status (at least in certain contexts).⁸ However, in what follows, I argue that it does not have the theoretical resources to account for the equal moral status of children. In brief, this is because opacity respect is neither a fitting nor a basic moral attitude that is owed to children qua moral status-holders. Let me explain. To begin with, it will be instructive to say something more about why it is appropriate to treat persons as opaque. To explain this, Carter invokes Strawson’s distinction between the ‘participatory’ perspective and the ‘objective’ perspective (Carter 2011: 559; this volume: 127–147), where the former consists in taking one’s agency as given, whereas the latter amounts to viewing a human being as someone ‘to be managed or handled or cured or trained’ (Strawson 2008: 9). According to Carter, opacity respect is the appropriate response to the possession of agential capacities because it expresses a particular form of the participatory perspective towards persons qua agents: it entails ‘assuming [a person’s] integrity (in the sense of wholeness or completeness), not measuring [their] capacities or dismantling them into so many reactions to stimuli’ (Carter 2018: 828). Respect for persons, therefore, requires abstaining from looking inside them, thus refraining from questioning their status as fully competent agents by regarding them as beings who need to be ‘managed, or handled, or cured or trained’. But if this is why opacity respect is a fitting or appropriate response to the moral status of persons, we can see that there are at least two reasons to maintain that an appeal to opacity respect is unable to justify children’s moral equality. First, it seems reasonable to hold that opacity respect is not an appropriate attitude towards children simply because children are in some fundamental way not yet developed agents, but they are in the process of developing (Schapiro 1999: 716). Accordingly, it is hard to see why we should take children’s agency (in the sense of wholeness or completeness) as given by showing opacity respect towards them when in fact children are not fully realized moral agents and, as such, their agency is not yet fully developed. Indeed, precisely for this reason, an ‘objective’ perspective is usually considered appropriate in the case of children: children are particularly ⁸ For criticisms of Carter’s view, see Arneson (2015: 44–48) and Christiano (2015: 57–58).
Giacomo Floris
245
vulnerable beings, and adults have a duty to satisfy their basic needs by nurturing, protecting, and educating them. It might be objected that a form of opacity respect is owed to those (older) children who have developed agential capacities up to at least some degree, despite not having reached the threshold for moral personality. However, even if we concede that some children are indeed entitled to a form of opacity respect, it is difficult to see why this should be considered a basic requirement of what is owed to them. On the one hand, there are very limited circumstances, if any, in which we might have a duty to assume that children are capable of elaborating coherent plans and forming and acting on a sense of justice, without inquiring into whether they in fact hold the necessary cognitive and emotional capacities up to a sufficient degree. On the other hand, and most importantly, it seems reasonable to maintain that there are other more fundamental obligations that are owed to children which would often trump the requirement of treating them as opaque. For example, it is hard to deny that the duty to promote children’s welfare or the duty to foster their development into well-adjusted adults are more fundamental requirements of what is owed to children than the duty to take their agency as given. Accordingly, in those cases in which opacity respect is incompatible with the fulfilment of these more fundamental duties, the latter should take priority over the former. Therefore, even if opacity respect is sometimes an appropriate response to the moral status of children, it is not a basic requirement of what is owed to them. Hence, it does not have enough normative weight to ground a significant commitment to children’s moral equality. Opacity respect is thus neither a fitting nor a basic moral attitude that is owed to children qua moral status-holders. Accordingly, an appeal to opacity respect is unable to account for children’s moral equality. Therefore, we need to identify another moral attitude which is a fitting and basic response to children’s moral status and that can offer a coherent and convincing rationale for why when assessing children’s moral status what matters is that they hold a status-conferring property (or properties) within a certain range. Now, moral philosophers disagree about what significant property (or properties) grounds children’s moral status: some suggest that children have moral status because they are self-conscious beings; others contend that children’s moral status is grounded in their capacity to flourish; still others maintain that children’s moral status is based on their being in relation with other human beings.⁹ However, whatever conceptions of children’s moral status one subscribes to, few would deny that one of the most basic ⁹ For overviews on the question of the basis of children’s moral status, see Floris (2023a) and Jaworska and Tannenbaum (2019).
246
The Basis of Childrenʼs Moral Equality
requirements of what is owed to children consists in assisting them in becoming well-adjusted adults who hold a proper and stable capacity for moral personality.¹⁰ Hence, at least part of what is owed to children as a matter of respect for their moral status is to provide them with those social conditions that foster the developmental process of obtaining a proper and stable capacity for moral personality. Doing this, however, does not require maintaining a proper distance and refraining from assessing any of the variable properties upon which children’s status-conferring property supervenes. On the contrary, it requires ‘looking inside’ children—thus violating opacity respect—to ascertain what they need to become capable of moral personality. Call this form of respect for children, positive respect.¹¹ In what follows, I argue that positive respect is a fitting and basic moral attitude owed to children qua moral status-holders, which offers a principled and compelling explanation for why the variations in the degree to which children hold their status-conferring property (or properties) above the moral status threshold should be considered as irrelevant when assessing their moral status. In short, this is because unequal consideration and treatment is a significant obstacle to the provision of some goods that are necessary to acquire a proper and stable capacity for moral personality and, therefore, it is incompatible with the fulfilment of the duty of positive respect. Children’s moral equality, then, is a constitutive requirement of what is owed to them qua moral status-holders.¹²
3. Moral Inferiority and Self-Respect Consider a fundamental right that children have qua moral statusholders: the right to health care. As we saw earlier, children’s standard status-conferring properties—such as the capacity to flourish or selfconsciousness—are scalar. The challenge, therefore, is to explain how any of ¹⁰ For different justifications of a child’s right to be assisted in acquiring the capacity for moral personality, see Brighouse (2002), Eekelaar (1986), and Schapiro (1999). ¹¹ A critic may observe that the moral imperative to provide children with the social conditions that facilitate their developmental process into well-adjusted adults is not a matter of respect, but concerns other modes of valuing, such as care for children’s welfare. In reply, I believe that respect is the appropriate mode of valuing because this positive duty is justified not only if and because it is compatible with and entailed by a commitment to care for children’s welfare. That said, my argument does not hinge on the nature of the mode of valuing underpinning the duty to ensure that children have access to the social conditions that promote their developmental process: what matters is that this is a fundamental requirement of what is owed to children simply by virtue of their moral status and, as such, it has enough normative weight to ground a significant commitment to children’s moral equality. ¹² In this chapter, I do not directly address the question of the scope of positive respect, that is, the question of who owes positive respect to children. I assume that positive respect is an appropriate attitude especially in the relationship between political institutions and children as citizens. Therefore, the positive-respect-based argument justifies children’s moral equality in the eyes of the state, at least.
Giacomo Floris
247
these scalar status-conferring properties can ground children’s equal moral status and, thus, justify their equal right to health care. For instance, suppose that there are two groups of children, A and B, and that the children of group A hold a scalar status-conferring property X to a lesser degree than the children of group B. What reason do we have to maintain that the children of groups A and B have equal moral status and, thus, an equal right to health care, if the former hold X to a lesser extent than the latter? In this section, I argue that the principle of positive respect can help us explain why we ought to ignore the variations of the degree to which the status-conferring property X is held above the relevant threshold when assessing children’s moral status. This is because moral inferiority is detrimental to the development and the maintenance of a child’s robust sense of self-respect which, in turn, is essential to obtain a proper and stable capacity for moral personality. Therefore, considering and treating children as moral inferiors with respect to one another is incompatible with the fulfilment of a duty of positive respect to ensure that children have access to those social conditions that facilitate the acquisition of the capacity for moral personality. Let us then begin by examining what self-respect is and why it is an essential precondition of the capacity for moral personality. Following Rawls, self-respect is commonly defined as ‘a person’s sense of his own value, his secure conviction that his conception of his good, his plan of life, is worth carrying out’, as well as ‘a confidence in one’s ability, so far as it is within one’s power, to fulfil one’s intentions’ (Rawls, 1971: 440).¹³ Having self-respect, then, is clearly a prerequisite for the possession of a proper and stable capacity for moral personality. On the one hand, a person who believes that nothing is worth doing—or that, even if something is worth doing, they are not capable of pursuing their own intentions—holds an impaired capacity to formulate and revise their conception of the good, and to carry out their life plan: either they are unable to elaborate a conception of the good, or their set of possible conceptions of the good is unduly limited due to a lack of confidence in their ability to pursue a wide range of life plans. On the other hand, as Thomas E. Hill Jr. observed, a person who does not respect himself ‘fails to acknowledge fully his own moral status because he does not fully understand what his rights are, how they can be waived, and when they can be forfeited’
¹³ To be precise, the Rawlsian notion of self-respect encompasses both recognition self-respect and evaluative self-esteem, where the former involves recognizing one’s moral standing in relation to others, whereas the latter consists in a positive evaluation of one’s life plan as well as confidence in one’s ability to carry it out (Dillon 1997: 229–231). In this section, I argue that inferiorizing consideration and treatment is a significant obstacle to the development of recognition self-respect and evaluative self-esteem, which are both necessary to hold a proper and stable capacity for moral personality.
248
The Basis of Childrenʼs Moral Equality
(Hill Jr. 1991: 9). Accordingly, a lack of a robust sense of self-respect constrains persons’ capacity to recognize and face the challenges against their own moral worth. This, then, makes them less prone to stand against the injustices that they are victims of, and therefore less able to contribute to the cause of those who suffer from the same injustice (Schemmel 2019: 635). Once we have clarified why the possession of a sense of self-respect is necessary to hold a proper and stable capacity for moral personality, the question that needs to be addressed is the following: what do children need to develop and maintain a robust sense of self-respect that can be retained even in the event of adversity? In other words, what are the social bases of children’s sense of self-respect? While a detailed answer to this question goes beyond the scope of this chapter,¹⁴ for our purposes it is sufficient to observe that being considered and treated as a moral inferior in the upbringing is widely recognized to be a significant obstacle to the development of a robust sense of selfrespect. For instance, feminist scholars have powerfully argued that a maledominant society—which rests on a socio-political construction that conveys the message that women are dominated by, and dependent on, men—greatly impedes women’s development into adults who are self-aware of their moral worth and their abilities to execute their life plan (Okin 1989). As Robin Dillon puts it, ‘where the basal framework codes inferiority due to deep and longstanding forms of social oppression, the result of self-construction is a diminished self: women become the lesser beings the dominant worldview defines us to be’ (Dillon, 1997: 246).¹⁵ Racial segregation is another paradigmatic example of how inferiorizing consideration and treatment is detrimental to children’s sense of self-respect, for it imposes an overwhelming obstacle on children—who belong to the race, or ethnic group, considered morally inferior—to become adults with a strong sense of what is owed to them qua moral status-holders and a resilient confidence in their agential abilities to fulfil their life plan. As the US Supreme Court in Brown vs Board of Education famously declared: ‘To separate [black children] from others of similar age and qualifications solely because of their race generates a feeling of inferiority as to their status in the community that may affect their hearts and minds in a way unlikely to ever be undone’ (Brown vs Board of Education 1954).¹⁶
¹⁴ For further discussion, see Ryan (2023) and Spotorno (2021). ¹⁵ See also Benson (1991) and Meyers (1987). ¹⁶ To be sure, I am not suggesting that all women or persons of colour who have not been brought up as equals do not have a robust sense of self-respect. Rather, the point is that since moral inferiority is a
Giacomo Floris
249
The avoidance of inferiorizing consideration and treatment, then, is a fundamental social base of children’s sense of self-respect: to assist children in acquiring a robust sense of self-respect, a just society ought to ensure that children have access to those social conditions that reinforce in them the awareness that they are not morally inferior to others as a fundamental motivational and epistemic resource to the development and maintenance of a proper and robust sense of self-respect. I began this section by presenting the case of two groups of children who possess a status-conferring property X to an unequal degree. The challenge, I have observed, is to provide a coherent and compelling answer to the variations objection: if children of group A hold X to a lower degree than children of group B, what reason do we have to maintain that they have equal moral status and, therefore, an equal right to health care? We are now in the position to better appreciate why the principle of positive respect offers a powerful response to this question: considering and treating children as moral inferiors with respect to one another is incompatible with the duty to provide them with those social conditions that facilitate the acquisition and maintenance of a robust sense of self-respect. Hence, it is inconsistent with what is owed to children as a matter of respect for their moral status. In particular, a society that considers the children of group A as moral inferiors, thereby (i) denying them a right to health care, or (ii) ascribing them a less stringent right to health care than children of group B—by, for example, granting the former access to worse hospitals and less comprehensive patient care services, or prioritizing the latter’s right to health care in cases of scarce resources and conflicting claims—imposes a significant obstacle to the development and maintenance of a robust sense of self-respect for the children of group A. For this reason, then, a just society ought to abstain from considering the variations in degrees to which children of groups A and B possess a status-conferring property X above a significant threshold as relevant for the purpose of assessing their moral status, as a matter of positive respect. The principle of positive respect thus supplies us with a plausible moral requirement that explains why, when reasoning about how children ought to be considered and treated, what matters is that they hold a significant property within a certain range, regardless of the variations
significant obstacle to the development and the maintenance of a robust sense of self-respect, it is incompatible with the fulfilment of the duty of positive respect to provide children with the social conditions that facilitate the acquisition of a proper and stable capacity for moral personality. Therefore, the validity of my argument does not rest on empirical evidence about whether children end up acquiring a robust sense of self-respect in adulthood despite having been considered and treated as moral inferiors in their childhood.
250
The Basis of Childrenʼs Moral Equality
above the relevant threshold.¹⁷ In other words, the avoidance of moral inferiority is a constitutive requirement of what is owed to children qua moral status-holders.
4. Moral Superiority and Friendship In the previous section, I have argued that moral inferiority is incompatible with the fulfilment of a duty of positive respect towards children because it hinders the development and maintenance of a robust sense of self-respect, which is a precondition of a proper and stable capacity for moral personality. While this is sufficient to justify a negative conception of children’s moral equality that demands the avoidance of inferiorizing consideration and treatment, it is insufficient to ground a positive conception of children’s moral equality whereby children should be considered and treated as moral equals. To justify the latter, an argument that explains why superior consideration and treatment is wrong in its own right is necessary.¹⁸ Thus, in this section, I argue that moral superiority is also detrimental to the development of the capacity for moral personality, thereby being incompatible with the satisfaction of a duty of positive respect. To begin with, it is important to observe that having access to relational resources is necessary to develop and exercise the capacity for moral personality. On the one hand, relational goods, such as trust, emotional support, ¹⁷ It is important to notice that this is not to say that children’s level of status-conferring properties should be considered irrelevant for all purposes. On the contrary, the level of a status-conferring property may be relevant when determining what is required to satisfy a right. Here is an example. Assume that children’s moral status is grounded in the possession of developmental agential capacities, and consider Luke, a child whose developmental agential capacities are impaired due to some mental health issues, and Stephanie, a child who does not suffer from any internal impairments. A commitment to positive respect entails that Luke and Stephanie ought to have an equal right to health care, regardless of their unequal level of developmental agential capacities. Presumably, however, society must allocate more resources to Luke than to Stephanie to satisfy their equal right to health care, precisely because Luke’s level of developmental agential capacities is inferior to that of Stephanie. Accordingly, while positive respect justifies ignoring children’s level of the status-conferring property when assessing the degree of their moral status, it allows— indeed it may demand—considering children’s levels of the status-conferring property as morally relevant when determining what the satisfaction of their equal rights requires. ¹⁸ To be clear, I am not suggesting that the superior consideration and treatment of some can occur without the inferior consideration and treatment of others. Rather, what I am suggesting is that moral superiority is also bad for those who are considered and treated as superiors, independently of, and in addition to, being bad for those who are considered and treated as inferiors, as a result of it. This is important for two reasons: first, because only an account of positive equality allows us to capture all the wrongs qua violation of equal moral status (Floris 2020: 143–144). Second, because one may object that if considering and treating children unequally is only bad for the allegedly morally inferior children but good for the allegedly morally superior children, then either (i) we do not have conclusive reasons to claim that children ought to be considered and treated as equals, at best, or, (ii) we do have conclusive reasons to maintain that children ought to be considered and treated as unequals, at worst, for the interests, or the claims, of children with superior moral status should count more. By showing that unequal consideration and treatment is bad for all children, a positive conception of moral equality avoids this pressing objection.
Giacomo Floris
251
care, friendship, and social influence are crucial for persons to maintain and develop a stable capacity for a conception of the good. On the other hand, the opportunity to establish social relationships and participate in associations is of particular importance to acquire a sense of justice as well as to discharge one’s duties of justice (Cordelli 2015b). Now, at least some of these relational resources play a fundamental role also in children’s developmental process into becoming healthy adults. Hardly anyone, in fact, would deny that trust, emotional support, care, and friendship are necessary to foster the acquisition of a proper and stable capacity for moral personality. Most importantly for our purposes here, it is generally held that friendship is an essential motivational and epistemic resource for the cultivation of proper and stable capacities for a conception of the good and of a sense of justice. Indeed, besides being an invaluable source of emotional support, friendship facilitates the development of the capacity for a conception of the good in three different ways: first, it allows broadening the range of ends that are regarded to be worth pursuing and therefore enriches the set of conceptions of the good among which one can choose (Friedman 1993: 207). Second, it promotes the ability to collaborate with others by justifying one’s solutions, and elaborating and criticizing others’ proposals (Azmitia and Montgomery 1993). Finally, being a source of social criticism, friendship also contributes to the development of the capacity to engage in critical deliberation with others who have a different perspective about what constitutes a flourishing life (Friedman 1993: Ch. 8). In short, then, friendship is an essential resource for the cultivation of the capacity to formulate, carry out, and revise one’s conception of the good. The developmental value of friendship in relation to moral growth has also been widely defended. First, as Rawls pointed out, ties of friendship are essential to generate association guilt, namely, the guilt that one feels as a result of failing to do their part in a scheme of cooperation. More specifically, friendship promotes moral sensibility to engage in fair cooperation by fostering a disposition to promote others’ interests—rather than exploiting others for one’s own benefit—and a sense of reciprocity and commitment to others (Rawls 1999: 103). Second, friendship also enables the widening of the range of moral perspectives, thereby enriching the set of reasons worth considering when assessing the validity of a specific moral principle (Friedman 1993: 200). Since friendship is a crucial motivational and epistemic resource for the acquisition of a proper and stable capacity for moral personality, a just society has a duty to ensure that children have access to those social conditions that facilitate establishing relationships of friendship, as a matter of positive
252
The Basis of Childrenʼs Moral Equality
respect. This has significant implications for the wrongness of moral superiority because a well-functioning friendship is a paradigmatic instance of an egalitarian relationship: a genuine friend, in fact, does not claim to have power over their friend, thus wielding greater authority concerning decisions made within the context of this relationship. Nor do they assert that their interests are worthy of special consideration—or that they are entitled to a special treatment that their friend does not have an equal claim to—by, for example, expecting their friend to be supportive of their projects while being indifferent towards their friend’s efforts to pursue their goals. More generally, a well-functioning friendship is an egalitarian relationship wherein the participants do not place a higher or lower moral value on themselves, thereby showing equal respect and concern towards each other (Aristotle 1984: 3925; Scheffler 2015: 32–33; Viehoff 2019: 10). Accordingly, if friendship is a relational good the lack of which obstructs children’s developmental process to obtain a proper and stable capacity for moral personality, and if friendship is accessible only among moral equals, then it follows that moral superiority is a significant obstacle to the acquisition of a proper capacity for moral personality. Hence, positive respect requires the avoidance of moral superiority as a fundamental social base of children’s access to establish relations of friendship. At this point, however, it might be objected that, even if this argument against moral superiority is correct, it is unable to justify a general prohibition against superior consideration and treatment. To appreciate this, consider the following example. Imagine a society that takes intelligence as a parameter to evaluate the degree of children’s moral status and assigns a superior moral status to the group of the ‘most intelligent’ children. If the wrongness of moral superiority lies in being an obstacle to children’s access to relations of friendship, then it is not clear what is wrong with equal moral superiority. After all, if the ‘morally superior’ children are each other’s moral equals, then they can be one another’s friends. Hence, the positive-respect-based argument against moral superiority, so the objection concludes, does not have the theoretical resources to justify a society in which no child is considered and treated as a moral superior; at best, it can condemn the superior consideration and treatment of very few children. In response to this objection, however, it should be noticed that friendship segregation along the lines of moral superiority is detrimental to children’s development into healthy adults capable of moral personality. The reason for this is twofold: first, as Emily Buss pointed out, interaction with a wider range of peers fosters ‘a more thorough exploration of [one’s] identity which, in turn, could produce a sense of self-understanding and self-authorship
Giacomo Floris
253
associated with more autonomous decisionmaking’ (Buss 2000: 1275). More generally, a diversified group of friends facilitates the development of the capacity to revise one’s conception of the good in light of social criticism and the capacity to critically engage with established social norms that influence one’s sense of justice (Cordelli 2015a; Moody 2001). Second, and most importantly, it is widely recognized that friendship segregation along the lines of moral superiority promotes significant epistemic vices, such as epistemic laziness and close-mindedness: specifically, by incentivizing ‘morally superior’ children to maintain their privilege, moral superiority hinders their willingness and ability to discover and critically question those social norms that underpin their privileged position in society (Medina 2013: 33–36). For example, as feminist scholars have forcefully argued, the inequality of gender that dominates the family poses a serious obstacle to the development of children’s ability and willingness to engage in fair deliberation with individuals of the opposite gender when they become adults (Okin 1989: 237). As a result of superior consideration and treatment during childhood, some adult men believe that their sense of justice need not be revised in light of feminist criticism. In addition, they are epistemically lazy insofar as they are not interested in critically engaging with patriarchal norms, and they are close-minded in that they are not open to being challenged by those values that may destabilize their own privileged position.¹⁹ Not all forms of friendship nurture children’s developmental process into well-adjusted individuals. In particular, friendship segregation along the lines of moral superiority prevents children from exploring a wide range of conceptions of the good and moral values, and obstructs the cultivation of important epistemic virtues—such as awareness and criticism of one’s own privileged position, epistemic curiosity, and open-mindedness—which are crucial to the possession of a proper capacity for moral personality. Accordingly, the positive-respect-based argument has the theoretical resources to condemn moral superiority as such: superior consideration and treatment is incompatible with the fulfilment of the duty of positive respect because it precludes, or at least is a significant obstacle to, having access to that kind of friendship that is an essential motivational and epistemic resource for the acquisition of a proper and stable capacity for moral personality, namely: integrated friendship. To conclude, a duty of positive respect towards children requires providing them with those social conditions that facilitate the acquisition of a proper ¹⁹ For further discussion of how moral superiority is bad for the moral superiors, see also Fourie (2022).
254
The Basis of Childrenʼs Moral Equality
and stable capacity for moral personality. In this section, I have argued that superior consideration and treatment is irreconcilable with this duty because it is a significant obstacle to children’s access to integrated friendship, which plays a crucial role in their development into well-adjusted adults capable of moral personality. Hence, a just society has a duty to avoid superior consideration and treatment as a fundamental social base of children’s access to integrated friendship.
5. Objections If children hold their status-conferring property (or properties)—for example, the capacity to flourish or the potential for rational agency—to an unequal degree, how come they should be considered and treated as each other’s equals? I have argued that the answer to this question lies in a fitting, basic, and independent moral attitude which is owed to children qua moral status-holders and that provides a coherent rationale for why the degrees to which children hold the status-conferring property (or properties) above the threshold for moral status are morally irrelevant. In particular, the basis of children’s moral equality lies in a duty of positive respect, which requires providing children with the social conditions that foster their developmental process into healthy adults who hold a proper and stable capacity for moral personality. This is because unequal consideration and treatment precludes, or at least obstructs, children’s development of a robust sense of self-respect and their ability to form integrated friendships. Hence, moral inequality is incompatible with what is owed to children as a matter of respect for their moral status. Therefore, an appeal to a duty of positive respect offers a coherent and convincing explanation for why when reasoning about how children ought to be considered and treated what matters is that they hold some relevant properties within a certain range. In this final section, I consider two objections that can be raised against this argument.
5.1 The question-begging objection A critic might object that a commitment to positive respect does not offer an independent explanation for why variations above the threshold for moral status are morally irrelevant. On the contrary, the positive-respect-based argument begs the question of the basis of children’s moral equality, for a commitment to children’s moral equality comes before a commitment to
Giacomo Floris
255
a duty of positive respect for children. Thus, for instance, in Section 3 I have argued that inferior consideration and treatment is incompatible with the duty of positive respect because moral inferiority hinders the development and maintenance of a robust sense of self-respect. But a critic might observe that this is true only if one presupposes an egalitarian conception of self-respect, whereby a person respects themselves as an equal in society. After all, it seems plausible to maintain that members of inegalitarian societies can have a sense of self-respect according to which they understand their ‘station’ and their ‘duties’, and fully carry out their duties in that role, come what may. If this is true, so the objection goes, then the argument against moral inferiority rests on an implicit commitment to moral equality which grounds a duty to consider and treat children as each other’s equals. Hence, the positive-respect-based argument fails to provide an independent explanation for children’s moral equality. In reply, I argue that a commitment against inferiorizing consideration and treatment of children need not presuppose an egalitarian conception of a child’s sense of self-respect; rather, it rests on an assumption in favour of an egalitarian conception of an adult’s sense of self-respect. Hence, the positiverespect-based argument does not beg the question of the basis of children’s moral equality. To see this, let us briefly recall the argument against moral inferiority: (1) Children are entitled to be provided with the social conditions that facilitate their developmental process into well-adjusted adults capable of moral personality, as a matter of positive respect. (2) Inferiorizing consideration and treatment is a significant obstacle to the development and maintenance of a robust sense of self-respect. (3) A robust sense of self-respect is a necessary precondition for holding a proper and stable capacity for moral personality. (4) Hence, inferiorizing consideration and treatment is incompatible with the fulfilment of a duty of positive respect. We can now see that there is no need to appeal to children’s moral equality to justify a duty to avoid treating children as moral inferiors with respect to one another. This duty is entailed by (i) a commitment to a form of respect that is owed to children qua moral status-holders which requires assisting them in becoming well-adjusted adults capable of moral personality, and (ii) the claim that a robust sense of self-respect is necessary for holding a proper and stable capacity for moral personality. What I am arguing therefore is not that children should be respected equally because they are moral equals, but that
256
The Basis of Childrenʼs Moral Equality
they ought to be considered and treated as equals because this is part of what is owed to them as a matter of respect for their moral status. An example may help to illustrate this point. Consider a person, John, who grew up into being a servant—that is, he has been considered and treated as a servant throughout his childhood—and, as a result of this upbringing, he respects himself qua a good servant who fulfils his duties in all circumstances. Now, it seems reasonable to hold that, far from having a robust sense of selfrespect, John has a broken sense of self-respect, insofar as he is incapable of considering himself as an equal in society, who has the right to revise his conception of the good and to challenge the kind of treatment that society deems him to be worthy of. Crucially, however, the wrongness of treating John as a moral inferior throughout his childhood need not rest on a prior commitment to the equality of moral status between John and the other children. Rather, the impermissibility of inferiorizing consideration and treatment is entailed by (i) the claim that John as a child is entitled to have access to those social conditions that foster his developmental process into a well-adjusted adult capable of moral personality, and (ii) the claim that inferiorizing consideration and treatment is incompatible with the satisfaction of (i) because it constrains the development of a robust sense of self-respect, whereby John is aware of his own moral worth and his own basic rights. For this reason, then, I conclude that the positive-respect-based argument for children’s moral equality does not beg the question: children’s moral equality does not entail but is entailed by a commitment to positive respect. The former is a requirement of the latter. An appeal to a duty of positive respect, therefore, provides an independent explanation for children’s moral equality.
5.2 The sidestepping objection Even if one accepts that the positive-respect-based argument for children’s moral equality does not beg the question, one may suspect that it merely sidesteps the problem rather than solving it. If we accept the positive-respectbased argument, so the objection goes, then we cannot conclude that children are each other’s moral equals, but we must simply adopt an agnostic attitude towards this question. This is because this argument only supplies us with a reason to maintain that society ought to ignore the unequal levels of children’s status-conferring property (or properties) when assessing their moral status, but it is unable to justify the claim that children are one another’s moral equals.²⁰ ²⁰ A similar objection has been raised against Carter’s opacity respect view. See Arneson (2015: 45).
Giacomo Floris
257
The positive-respect-based argument, however, does not merely entail that society cannot ascertain the levels of children’s status-conferring property (or properties) and therefore it is compelled to consider them as moral equals even if they may not be. Rather, society ought to regard and treat them as such because this is a constitutive requirement of what the principle of respect for children demands. In other words, the positive-respect-based justification is not an epistemic argument about the impossibility of determining the level of the scalar properties that ground children’s moral status. Rather, it is a normative argument about the impermissibility of considering the level of the scalar properties above the threshold for moral status as morally relevant when assessing children’s moral status, because this is inconsistent with what is owed to them qua moral status-holders. Thus, the principle of positive respect offers a coherent and plausible rationale for why when reasoning about how children ought to be considered and treated what matters is that they hold a relevant status-conferring property within a certain range, regardless of the variations in degree to which this property is held above the moral status threshold. Therefore, it explains why children are equal in the possession of the range property, thereby providing a normative basis for their equal moral status. Hence, the side-stepping objection fails to undermine the positive-respect-based argument for children’s moral equality.
6. Conclusion Much of the literature on the basis of moral equality has focused on what makes persons—understood as fully competent adults—one another’s equals. Less attention, instead, has been given to the question of the equal moral status of those human beings who are not fully competent adults. In this chapter, I have contributed to filling this gap by developing a novel theory of the basis of children’s moral equality. I have argued that the basis of children’s moral equality lies in a fitting, basic, and independent attitude of positive respect which is owed to children qua moral status-holders. This kind of respect requires providing children with the social conditions that foster their developmental process into well-adjusted adults capable of moral personality. But unequal consideration and treatment precludes, or at least is a significant obstacle to, the provision of some goods that are crucial to the cultivation of proper and stable capacity for moral personality, namely: a robust sense of self-respect and access to integrated friendship. Therefore, it is inconsistent with a duty of positive respect towards children. Hence, children’s moral
258
The Basis of Childrenʼs Moral Equality
equality is a constitutive requirement of what is owed to children qua moral status-holders.²¹
References Aristotle. 1984. ‘Nicomachean Ethics’. In The Complete Works of Aristotle, edited by J. Barnes, 3718–4009. Princeton: Princeton University Press. Arneson, R. J. 2015. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Azmitia, M., and Montgomery, R. 1993. ‘Friendship, Transactive Dialogues, and the Development of Scientific Reasoning’. Social Development, 2/3: 202–221. Benson, P. 1991. ‘Autonomy and Oppressive Socialization’. Social Theory and Practice, 17/3: 385–408. Brighouse, H. 2002. ‘What Rights (If Any) Do Children Have?’. In The Moral and Political Status of Children, edited by D. Archard and C. M. Macleod, 31–52. Oxford: Oxford University Press. Brown vs Board of Education. 1954. 347 U.S. 483. Buss, E. 2000. ‘The Adolescent’s Stake in the Allocation of Educational Control Between Parent and State’. University of Chicago Law Review, 67: 1233. Carter, I. 2011. ‘Respect and the Basis of Equality’. Ethics, 121/3: 538–571. Carter, I. 2018. ‘Equal Opportunity, Responsibility, and Personal Identity’. Ethical Theory and Moral Practice, 21/4: 825–839. Christiano, T. 2015. ‘Rationality, Equal Status, and Egalitarianism’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 53–75. Oxford: Oxford University Press. Cordelli, C. 2015a. ‘Distributive Justice and the Problem of Friendship’. Political Studies, 63: 679–695. Cordelli, C. 2015b. ‘Justice as Fairness and Relational Resources’. Journal of Political Philosophy, 23/1: 86–110. Dillon, R. S. 1997. ‘Self-Respect: Moral, Emotional, Political’. Ethics, 107/2: 226–249. Eekelaar, J. 1986. ‘The Emergence of Children’s Rights’. Oxford Journal of Legal Studies, 6/2: 161–182. ²¹ Earlier versions of this chapter have been presented at the 11th Braga Meetings on Ethics and Political Philosophy at the University of Minho; the 14th Conference of the Italian Society for Analytic Philosophy at the University of Messina; and the Politics and Philosophy Seminars at the University of Genoa. For their questions and comments, I am grateful to the participants at these events. I would also like to thank Ian Carter, Agnieszka Jaworska, Nikolas Kirby, Matt Perry, Liam Shields, and an anonymous Oxford University Press reader for very helpful written comments. Finally, I gratefully acknowledge financial support from the European Union’s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101060448.
Giacomo Floris
259
Floris, G. 2019. ‘On the Basis of Moral Equality: A Rejection of the Relation-First Approach’. Ethical Theory and Moral Practice, 22/1: 237–250. Floris, G. 2020. ‘Two Concerns about the Rejection of Social Cruelty as the Basis of Moral Equality’. European Journal of Political Theory, 19/3: 408–416. Floris, G. 2023a. ‘Are Adults and Children One Another’s Moral Equals?’. Journal of Ethics, 27/1: 31–50. Floris, G. 2023b. ‘Contingency, Arbitrariness, and the Basis of Moral Equality’. Ratio, 36/3: 224–234. Floris, G., and Spotorno, R. Forthcoming. ‘What Does It Mean to Be Moral Equals? The Special Case of Children’s Moral Status’. Social Theory and Practice. Fourie, C. 2022. ‘How Being Better Off Is Bad for You: Implications for Distribution, Relational Equality, and an Egalitarian Ethos’. In Autonomy and Equality: Relational Approaches, edited by N. Stoljar and K. Voigt, 169–194. New York: Routledge. Friedman, M. 1993. What Are Friends For? Feminist Perspectives on Personal Relationships and Moral Theory. Ithaca and London: Cornell University Press. Hill Jr., T. E. 1991. Autonomy and Self-Respect. New York: Cambridge University Press. Jaworska, A., and Tannenbaum, J. 2014. ‘Person-Rearing Relationships as a Key to Higher Moral Status’. Ethics, 124/2: 242–271. Jaworska, A., and Tannenbaum, J. 2019. ‘The Moral Status of Children’. In The Routledge Handbook of the Philosophy of Childhood and Children, edited by A. Gheaus, G. Calder, and J. De Wispelaere, 67–78. New York: Routledge. Kant, I. 2002. Groundwork for the Metaphysics of Morals. New Haven and London: Yale University Press. Kittay, E. F. 2005. ‘At the Margins of Moral Personhood’. Ethics, 116/1: 100–131. Medina, J. 2013. The Epistemology of Resistance: Gender and Racial Oppression, Epistemic Injustice, and the Social Imagination. New York: Oxford University Press. Meyers, D. T. 1987. ‘Personal Autonomy and the Paradox of Feminine Socialization’. Journal of Philosophy, 84/11: 619–628. Miklosi, Z. 2022. ‘The Problem of Equal Moral Status’. Politics, Philosophy & Economics, 21/4: 372–392. Moody, J. 2001. ‘Race, School Integration, and Friendship Segregation in America’. American Journal of Sociology, 107/3: 679–716. Okin, S. M. 1989. ‘Reason and Feeling in Thinking about Justice’. Ethics, 99/2: 229–249. Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Harvard University Press. Rawls, J. 1999. ‘The Sense of Justice’. In Collected Papers, edited by S. Freeman, 96–116. Cambridge, MA: Harvard University Press.
260
The Basis of Childrenʼs Moral Equality
Ryan, N. 2023. ‘Self-respect & Childhood’. Journal of Ethics, 27/1: 51–76. Sangiovanni, A. 2017. Humanity without Dignity. Moral Equality, Respect, and Human Rights. Cambridge, MA: Harvard University Press. Schapiro, T. 1999. ‘What Is a Child?’. Ethics, 109/4: 715–738. Scheffler, S. 2015. ‘The Practice of Equality’. In Social Equality: On What It Means to Be Equals, edited by C. Fourie, F. Schuppert, and I. Wallimann-Helmer, 21–44. Oxford: Oxford University Press. Schemmel, C. 2019. ‘Real Self-Respect and Its Social Bases’. Canadian Journal of Philosophy, 49/5: 628–651. Sher, G. 2015. ‘Why We Are Moral Equals’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 17–29. Oxford: Oxford University Press. Spotorno, R. 2021. ‘Homophobes, Racists, and the Child’s Right to Be Loved Unconditionally’. Critical Review of International Social and Political Philosophy, early view, 1–24. Strawson, P. F. 2008. Freedom and Resentment and Other Essays. London: Routledge. Viehoff, D. 2019. ‘Power and Equality’. In Oxford Studies in Political Philosophy, edited by D. Sobel, P. Vallentyne, and S. Wall, 3–38. Oxford: Oxford University Press. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: The Belknap Press of Harvard University Press.
11 Basic Human Moral Equality Eva Feder Kittay
1. Introduction ‘Whereas recognition of the inherent dignity and of the equal and inalienable rights of all members of the human family is the foundation of freedom, justice and peace in the world …’. So begins the UN Declaration of Universal Human Rights. Its assertion of human equalityechoes numerous philosophical and religious treatises in the Abrahamic religions, Buddhism,¹ Hinduism,² as well as those of many indigenous belief systems. Some religions not only stress the equality of humans, but also maintain that nonhumans and our natural environment have equal moral importance. Basic equality posits the equal moral worth of all human beings. Philosophers have attempted to make sense of claims to basic human equality, given the extent of human variability. The idea of equality has had its critics. Some argue that many evident differences among human beings are not given by nature, but are created by unjust conditions. The presumed inherent inferiority of entire populations is reflected in the bloody course of human history. Those who affirm human equality despite human difference will view the claim as a normative one about how we should treat each other—all should be treated as equals. Most opposition to the idea of human equality has come from rigidly aristocratic, authoritarian, or fascist regimes. However, criticisms of equality as a goal of liberation movements also come from progressive quarters. Who, they ask, do aspirants want to be equal to? The question exposes a norm defined by heterosexual men’s bodies and men’s lives—hegemonic norms that do not fit women and other marginalized groups well. Nonetheless, an assumption of a fundamental moral equality is not discarded. ¹ For Buddhist beliefs see Sarao and Long (2017). ² In Hindu sacred texts, we find passages such as ‘No one is superior, none inferior. All are brothers marching forward to prosperity’ (Rig Veda V.60.5), and I regard them to be perfect yogis who see the true equality of all living beings and respond to the joys and sorrows of others as if they were their own (Bhagavat Gita 6.32). Eva Feder Kittay, Basic Human Moral Equality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0012
262
Basic Human Moral Equality
There is another challenge that comes from an unexpected source: some proponents of animal welfare or animal rights. It is perhaps currently the most potent objection to the idea that all human beings are equal. The claim to human equality renders nonhumans our moral inferiors. In the face of such criticism, I want to assert the moral equality of all human beings born in a condition compatible with life. That is not to claim that being human is both necessary and sufficient for the full robust moral status we attach to humans. The claim of human equality instead should be understood as a sufficiency condition, one compatible with the extension of moral equality to nonhumans. But, in the view of these critics of human equality, as we will see, even the sufficiency condition is not enough to escape their critique. They object that a belief in human equality is a mere prejudice, ‘the human prejudice’, as Bernard Williams (who argued that it is entirely justifiable), called it. The critics instead argue that moral solidarity among human beings needs to be replaced by one based only on the possession of an individual’s intrinsic morally relevant attributes. The appropriate question, they maintain, is not: does the individual belong to the human species? It is: does the individual sentient being, regardless of its species, have intrinsic (not relational) relevant moral properties? Those who possess the morally relevant properties are persons.³ Highlighting intrinsic morally relevant properties is a philosophical dogma that emerged among thinkers who abandoned a theological conception of human equality and adopted a naturalistic one.⁴ Another long-standing philosophical position has been that the ability to reason is required for full moral status. It is the characteristic used repeatedly to distinguish humans from all other creatures and to place them at the top of the scala naturae, just below the angels and above the beast. The assumption is that it is the unique ability to reason that is the source of our singular moral status. It is similarly the basis of our dignity. The absence of reasoning capacity in some humans, when they are acknowledged in philosophical texts, has rendered them less than human, or, at least, lesser humans. Locke compared them to the ‘changeling’ who takes on a human form but is either a creature of the devil or not fully human. Reason and reflection and the sameness of self from one time to another is what defines the Person (or Moral Man, as Locke says). Mere membership in ³ I prefer to say ‘moral persons’ as ‘person’ does not always refer to moral standing, for example, the term as it is used in a legal context. ⁴ The Christian conception held that differences among human beings were of indifference to God. It is within a naturalistic conception that equality came to be characterized by the similar possession of intrinsic properties such as rationality. For a fascinating discussion of the historical emergence of concepts of equality, see Bejan (this volume) and Phillips (this volume).
Eva Feder Kittay
263
the human species is not significant. Locke declared: ‘Were these to be found in a monkey or parrot, we would have to view it as a moral equal’.⁵ Should we then speak of the equality of persons, rather than humans? The Lockean position, excluding some humans from personhood but not from humanity, is adopted in contemporary discussions. People with significant cognitive impairments are ‘on the margins’ of personhood. Nonhuman beings, such as very highly evolved primates, might now be considered for full moral status. Other beings with a lesser capacity for reason (or other morally significant intrinsic psychological capacities) have different degrees of moral status. The concept of humanity itself is stripped of any moral significance not just as a necessary condition, but even as a sufficient condition.
2. The Challenge of Moral Individualism James Rachels has been influential in asserting that we should look to the individual’s intrinsic properties in assigning moral status, regardless of species membership. How should we understand species membership and its relevance to intrinsic properties? On one view, ‘species’ references intrinsic properties, in particular a being’s genetic makeup. On this view, what Rachels excludes from moral consideration are intrinsic properties that are speciesspecific, and the claim is that species-specific intrinsic properties are not morally relevant. The standard evolutionary definition of species, however, is based not on intrinsic properties, but relational ones. Philosopher of science David Hull writes: ‘If species are to be units of evolution, they need not be composed of similar organisms⁶; instead they must be made up of organisms related by descent (Hull, 1976)’.⁷ That is, relational attributes such as species do not refer to the capacities of any individual, but to the interrelations among individuals. ⁵ See the discussions in (Locke, 1690, Bk III, 16, 218 and Bk II, 27, 8). Locke’s interspecies reference was aimed at excluding those humans lacking these, rather than granting moral equality to nonhuman animals. He did not appear to seriously countenance the possibility that nonhuman animals might be persons. His rather skeptical discussion of the talking parrot in Bk 2, is, I think, indication enough. ⁶ He continues: ‘Taxonomists do not impose this requirement on the phenomena; rather it follows from the nature of the evolutionary process itself ’ (Hull 1976: 2). Organisms are characterized by intrinsic properties such as genetic makeup (Hull 1976: 2). The most current view is that there are multiple species concepts, and determining which one to use depends on what you want to do with it (Elisabeth Lloyd, personal correspondence). Relying on the intrinsic genetic property would, I believe, still be dismissed by Rachels, and other moral individualists, as merely a scientific identification carrying no moral significance. The intrinsic properties that moral individualists are interested in are those that are morally significant. At the end of the chapter, we will pursue the question of what makes a property morally significant, and who decides what is morally significant. ⁷ See a further discussion of this point in Section 4.
264
Basic Human Moral Equality
The moral individualist claims that only those properties that belong to individuals as such have any moral claims on others. The one exception is those moral obligations we owe to those with whom we are in a special relationship, such as parent to child. We owe something to our child merely because it is ours—not because of the child’s intrinsic capacities. The capacity to reason is regularly invoked as a morally significant intrinsic property. Its moral relevance is independent of any claims of its being a uniquely human capacity. Furthermore, it is evident in other animals and apparently absent in some human beings.⁸ Reason is neither a mark of humanity, nor should full moral status be accorded exclusively to humans. To claim that moral individualism is applicable to all sentient beings (and perhaps to supersmart robots or space aliens), one must believe not only that species membership is morally irrelevant, but also that morally relevant features cross species boundaries, are comparable, and can be hierarchically ordered. With respect to cognitive capabilities, for example, one must hold that they are just the same in human and nonhuman animals (and perhaps in robots, or Martians) and may be greater in some individuals and inferior to others. Moral individualism both allows us to count possible nonhumans as having full moral status and requires us to single out human beings without the requisite capacities as not equal to other human beings.⁹ Being human should not accord an individual a higher moral status than nonhumans with comparable cognitive (or other morally relevant) capacities. Adopting this position, Singer asks why it should be acceptable to kill a nonhuman animal to obtain organs for transplant, and wrong to kill a severely impaired infant, as he judges the latter to have even less of the cognitive abilities that count as morally relevant properties? In his most provocative book, Should the Baby Live?, he and coauthor Helga Kuhse state quite baldly: ‘We think that some infants with severe disabilities should be killed’ (1988). Singer does acknowledge that parents may object, and in deference to the parents who have full moral status, the killing could be disallowed in such a case. Jeff McMahan treads on the same ground, challenging our intuitions about the relative wrongs of killing certain human beings, claiming in The Ethics of Killing (henceforth EOK) that it is less bad to kill a ‘congenitally severely
⁸ Korsgaard (2009: 23–44) distinguishes the intelligence of animals from rational agency. But, as she acknowledges, how higher animals use perception to navigate the world—and whether they are self-conscious in the way humans are—is a matter of investigation. ⁹ Were there a property that all humans share with some nonhumans, such as subjectivity, as Sher (this volume) argues, then we could grant that such a property is sufficient to confer some moral status. But we would need to know if such a property were to confer full and equal moral status.
Eva Feder Kittay
265
mentally retarded [sic]’¹⁰ human being than to kill ‘one of us’, and equivalent to the wrong of killing a nonhuman animal of comparable cognitive capacities. He, too, acknowledges the force of special relations but insists that such a child (or adult) has no moral claim on ‘the rest of us’ (Kittay, 2010).
3. Is Species Membership Morally Irrelevant? Is it reasonable to dismiss species membership as morally irrelevant? Can we speak of individuals without reference to their species membership? Rachels himself explains the harms that animals endure in terms of their species. For example, he speaks of having to choose between killing a rhesus monkey and swatting a fly: ‘If we compare the two, we find that the life of the monkey is far richer and more complex than that of the fly, because …’ (1999). Rachels is not speaking of a particular chimp, or of a particular fly, but of chimps and flies, and he characterizes them by species-specific features. In a particularly vivid example, Rachels informs us that veal calves are kept throughout their lives in pens too small to allow them ‘to turn around or even to lie down comfortably, making it impossible for them to groom themselves’ (1999: 212). The passage speaks of an ability to ‘turn around’ or ‘lie comfortably’. He concludes, not unreasonably, that this possibility is what a calf would want—a want that we appropriately regard as important from a moral point of view. As they are deprived of their mothers’ teats to suck, ‘they can be seen vainly trying to suck the sides of their stalls’. He refers here to conditions that also would cause humans to suffer. He continues: … [T]hey are fed a liquid diet deficient in both iron and roughage. The calfʼs craving for iron becomes so intense that, if allowed to turn around, it will lick at its own urine, although calves normally find this repugnant. The tiny stall, which prevents the animal from turning, solves this problem. (1999: 212)
What Rachels describes is awful enough to protest such treatment. But how does he characterize the harm? He speaks of what calves do and compares how like they are to young human beings in certain regards—again speaking in species terms. A human similarly confined would not be drawn to licking ¹⁰ Both Singer and McMahan use the term ‘mentally retarded’ not as derogative term, but as a designation. Today the term is viewed as no less offensive than a racial slur. When discussing their views, I will use their term, but I will put quotation marks around the phrase to distance my terminology from theirs.
266
Basic Human Moral Equality
their urine for iron. That is a behaviour specific to bovines. A moral individualist could counter that such talk of a species-specific harm or benefit can be introduced, but only as one factor. A utilitarian would hold that the calculations are based on the harms and benefits to individuals, not to the individuals as belonging to a given species.¹¹ One can agree that to cause creatures who are sentient to needlessly suffer—regardless of species membership—is morally wrong, on utilitarian (but not only on utilitarian) grounds. But what is typically painful or pleasurable for one species may not be—or may be less so—for another. Let’s suppose a flea is sentient. A flea, with no need to suck or groom, might find such a calf ’s pen spacious and very comfortable—especially if its companion were a calf to feed on. Assessing harms and benefits to an individual requires us to refer to the entity’s species.¹² Now suppose a farmer claimed that his individual calf doesn’t mind the confinement. Would Rachels then approve of the practice for such an outlier? I wouldn’t, and I suspect Rachels’ moral individualism would not reach that far.
4. The Importance of Empirical Knowledge for Moral Assessments of Other Human Beings If moral individualists want to discount the importance of species membership of individuals, one would assume that they know a good deal about the individuals in question. Thus, it is particularly shocking how those who propound similar views have little knowledge of the human beings they exclude from the equal moral worth they attribute to persons. This is evident in a provocative example where Singer (1994) asks us to think about a ‘special institution’ for the ‘retarded’ in the Netherlands. Here individuals are confined but live a life without many of the usual constraints of such institutions. They are free to wander about, free to form associations with one another, engage in sexual activity, have children, and raise them when they result from sexual encounters. They pick leaders and get elderly females to help with raising the little ones. None of the residents have language but indicate their desires and wants with grunts and gestures and ¹¹ This is regardless of the nature of the utilities counted, but most certainly so for hedonic utilities. ¹² Agnieszka Jaworska has remarked that to assess that a creature has been harmed still does not establish the creature’s moral status. She points out that we need to determine if he has the right not to be harmed in this species-specific manner. Nonetheless if harm is species-specific, the project of assessing the creature’s right not to be harmed cannot get off the ground with taking species membership into account. (Agnieszka Jaworska, personal communication. However, moral individualists who are also hedonic utilitarians may see this as one distinction too many.)
Eva Feder Kittay
267
seem to communicate quite well by these means. The passage concludes with information that should one be killed, such deaths are viewed by their supervisors as less significant than the killing of other residents in the Netherlands. Nor are they given the same medical care as other residents. In other institutions like this one, they are sometimes used in lethal experiments. Singer ends with the question of how this treatment should be viewed: as justifiable or as a moral outrage. Although we initially are led to believe that the passage is about humans with impaired cognitive abilities—indeed, that is how the passage is introduced—after the description there is a big reveal: it is not a description of humans, but of chimpanzees in the Arnhem Zoo in the Netherlands.¹³ On the supposition that the audience will believe that the residents are impaired humans, Singer concludes that as both populations share the same morally relevant attributes, we should agree that they have the same moral standing, and so any moral outrage we would feel about their deaths should apply equally to both populations. To deny this is not to affirm that chimpanzees should be used in lethal experiments, nor that their killing should be viewed as unimportant. But the argument using this example yields a question-begging conclusion. The hypothetical example assumes that the capacities—and so the moral status—of the two populations are the same. In this way, the conclusion—that is, what we should think about their relative moral status—is built into the construction of the example. But why should we take Singer’s word that they are alike? It is true that Singer makes it clear that this is a description of chimpanzees, not human cognitively impaired individuals—so why should he be faulted for not getting the picture right? Yet to construct the example already supposes that chimps and ‘retarded individuals’ are the same, morally speaking. And, to believe that this is an effective subterfuge, as he calls the example, he depends on his reader’s ignorance of its subject. It depends, for example, on believing that within each community residents are alike in their capacities, that they function at much the same level, that they are free of physical impairments that would limit their ability to meet their needs, and that they are capable of the self-organization evident in the example. All this is true of the chimp community, and this is true because they are non-impaired chimps. But it is ¹³ The presumed comparability would be risible were it not morally repugnant. To understand just how odious the comparison is, recall Joel Grey in Cabaret Fosse (1972) declaring his love for a gorilla who is dressed in satins and silks. He asks the audience to ‘see her through my eyes’. The punch line goes: ‘If you could see her through my eyes’, followed by a whisper, then, ‘she wouldn’t look Jewish at all’. The simile is turned around, but the force is the same. In Cabaret the comparison is used to slyly expose the grotesque views of the Nazis; here it is supposed to enlighten us. (The scene is available at https://www.youtube. com/watch?v=q9XdGS5to8s.)
268
Basic Human Moral Equality
not true of a group of significantly cognitively impaired humans. To borrow from Tolstoy, ‘all unimpaired humans are alike; each human with (severe) impairments is impaired in its own special way’. The example does not so much challenge a reader to revise their views about the moral standing of chimpanzees as it reinforces prejudices that distort the reality of the lives of people with severe disabilities—making them appear more chimp-like than human. If Singer is dubious, I submit, it is only because he has never gotten to know the human populations of which he speaks. How are they different? I have already remarked on the fact that great diversity of cognitive disabilities would make such an ‘unusual institution’ unlikely. This diversity is not simply a question of ‘more or less’ (say) of cognitive capacities. Instead, it is a question of which capacities are intact, and even perhaps heightened, and which capacities are impaired. One writer who has worked with people with severe cognitive disabilities writes that when you’ve seen one person with cognitive disabilities, you’ve seen one person with cognitive disabilities. There are abilities that are undeveloped, underdeveloped and underestimated even by those who are close to the person¹⁴. I can give many examples of what one would consider ‘morally relevant’ capacities. But, as with the attempt to find a property that is unique to humans and not found in nonhuman animals, one is likely to find the capacity in perhaps an attenuated form in nonhuman beings. More importantly, I do not want to claim that the moral status of human beings is a matter of capacities as such—as I make clear later in this chapter. By eliciting our moral outrage at how cognitively disabled people in the hypothetical example are treated, Singer hopes to show that we should feel the same moral outrage at how chimpanzees are treated. Singer’s point is to say that we should treat animals better than we currently do. Charitably, we can suggest he doesn’t mean that we should treat humans as poorly as we treat animals today, but in making the comparison, he does seem to think we should treat them as well as we should treat nonhuman animals. I submit that Singer fails to appreciate the extent to which cognitively disabled human beings have, in many regards, already been treated as nonhumans. This has happened with little hue and cry. He remarks that killing one of them is not viewed by the supervisors or the townspeople to be as serious as killing one of us, that these residents do not get the same medical care as other inhabitants of the Netherlands, and that they are used in lethal experiments. To convince us that these are morally indefensible when
¹⁴ My use of the word ‘person’, rather than individual, is intentional.
Eva Feder Kittay
269
applied to chimps, he asks us to consider that were this treatment applied to the severely cognitively disabled, such treatment would be condemned. My reply is that all these treatments are readily found in the histories of cognitive disability and in current-day newspapers; they have been advocated by thinkers such as D. H. Lawrence and Virginia Wolff. D. H. Lawrence wrote: If I had my way, I would build a lethal chamber as big as the Crystal Palace, with a military band playing softly, and a Cinematograph working brightly; then Iʼd go out into the back streets and the main streets and bring them all in, the sick, the halt and the maimed: I would lead them gently, and they would smile me a weary thanks; and the brass band would softly bubble out the Hallelujah Chorus.15
Nearly ten years later, Virginia Wolff wrote: On the towpath we met & had to pass a long line of imbeciles. The first was a very tall man, just queer enough to look at twice, but no more; the second shuffled, & looked aside; and then one realised that everyone in that long line was a miserable ineffective shuffling idiotic creature with no forehead, or no chin, & an imbecile grin, or a wild suspicious stare. It was perfectly horrible. They should certainly be killed.16
Not only do Black mothers with non-disabled sons have to fear that they will be gunned down by the police: Black, brown, and white mothers have to worry about the autistic son or daughter who ventures out alone into the world.¹⁷ The example of lethal experiments Singer mentions were, in fact, practices at state institutions such as Willowbrook. Residents there were intentionally injected with the hepatitis virus, on the assumption that they would likely die of hepatitis anyway as the disease was rampant in the institution. The practice was viewed as defensible by the medical community. It is offered up in bioethics texts as a quandary about consent for medical experimentation—the subjects themselves were unable to consent or to refuse participation—instead of as an instance of a morally indefensible practice, and of morally indefensible conditions in the state hospital. The denial of the same medical care for the severely cognitively disabled is not unusual. Some doctors look askance at a mother like me who fights like hell for the best treatment for her severely disabled daughter, spending all day and night in ¹⁵ D. H. Lawrence in a letter to Blanche Jennings, 9 October 1908. (1979 1. 81) ¹⁶ Woolf (1915) ¹⁷ For a recent example, see ProPublica’s reporting of an Illinois residential facility. https://www. propublica.org/article/illinois-choate-mental-health-abuse-911-staff ?utm_source=sailthru&utm_ medium=email&utm_campaign=majorinvestigations&utm_content=river.
270
Basic Human Moral Equality
her hospital room to assure herself that her daughter is not being disregarded because of her disabilities. People with much less severe disabilities speak of the many times they are not afforded the care offered to the nondisabled. During the COVID-19 pandemic, whether they were awaiting treatment or vaccines, people with cognitive disabilities were relegated to the end of line, sometimes denied treatment altogether. To attempt to elicit moral outrage at the ill-treatment of chimps by comparing them to a group of humans who have failed to elicit moral outrage at their mistreatment would seem to be an ineffective way to advocate for chimps. Unfortunately, it is an effective way to reenforce prejudices that lead to the mistreatment of the concerned humans in the first place. Singer is right that we oughtn’t treat animals as mere things to be disposed of. (Nor should we neglect the importance of relationships in their lives.) But Singer’s hypothetical example is deeply flawed and tells us little about the moral comparability of (on the one hand) people with significant cognitive disabilities and (on the other) normally functioning chimpanzees, a comparability that isn’t already evident in any comparison between chimps and people without disabilities. A chimp is not a human being minus some sophisticated psychological abilities. The difference is a species difference.
5. Cognitive Ability, Speciesism, and Moral Standing Taking species membership out of the moral equation does have a certain attraction. We could accept the widespread view that species is a merely biological classification (a fact) and therefore has no normative significance (a value).¹⁸ However, a sharp fact-value distinction is neither possible nor sound (Putnam 2002). Hilary Putnam points out that we have to dismiss well-established and productive scientific practices rooted in values rather than a fact, as, for example, looking for a ‘beautiful’ solution. Moreover, being a member of the species homo sapiens is never a bare fact—all human beings are born into a human community, and all require nurturance from socialized and enculturated human beings to survive, much less thrive. This value-laden fact already points to morally significant attributes of human beings. Nonetheless, as it is difficult to find a single human intrinsic capacity with moral significance that we can say definitively is not found in animals, it remains appealing to strip species membership of moral importance. ¹⁸ See footnote
Eva Feder Kittay
271
We could alternatively invoke a morally relevant feature whether or not it belongs to humans. Kant does away with the requirement that a creature be human, as long as it is rational. Kant however does insist that all human beings have a rational nature. Even if, at any given time, one may not seem rational, Kant maintains that one’s nature is still rational and one remains an end, not merely a means to someone else’s end. For Kant, then, being human is sufficient for equal moral consideration. Korsgaard, a Kantian, thinks that the one thing that Kant gets wrong is that no other natural creature is an end. She argues that all our fellow creatures are ends. Strikingly, however, what draws other beings into the moral circle—that all beings are ends and not mere means—reintroduces species membership. On her view, all creatures are subjects of their own experience, and so have a point of view from which they value things. That alone makes them ends. That end is tied to its species membership.¹⁹ What is distinctive of humans is that we have a special sort of agency: a collective agency (Korsgaard 2018). She draws on Marx’s conception of a species-being.²⁰ One does not create vaccines for one’s own needs alone, but for humans one will never meet, and those who have yet to be conceived. The grieving parent who becomes an activist after losing a child to gun violence cannot bring her child back; she acts so parents (and not only those with cognitively able children) will not suffer such grief. All humans are included within this collective. It matters not that some humans do not appear to be rational. One may insist that other creatures are capable of collective agency, but Korsgaard’s claim that ‘being moral is the human way to be an animal’ makes the point more sharply. If ‘being moral’ is the human way to be an animal, then of course our species membership matters morally—as does the species membership of other animals.
6. Substituting Hierarchies Even while denying the moral priority of humans over nonhumans, moral individualists in fact create another category to supplant the importance of species: (moral) ‘personhood’. They espouse a two-tiered morality: one ¹⁹ One might say that this is species membership not in a biological sense. Instead, it identifies a member’s species by a salient moral property, their end. But I find it difficult to parse the biological from the moral here insofar as the end is just what is the biologically given—the moral understanding that the being has an end (in the moral sense) seems to me to be affirming the moral significance of their biological strivings. ²⁰ See discussion in Czank (2012: 316–323).
272
Basic Human Moral Equality
for moral persons, another for nonpersons. Reason and other higher psychological capacities, now not tied to the human species, become criterial for full and equal moral status, and their lack is a disqualification. It’s not okay to be speciesist, but it is okay to be personist, to use a neologism. This preferred status is, for them, not mere prejudice because these capacities are ‘morally relevant’. A nonperson presumably has only his present interests that matter morally.²¹ The person who acts in part with respect to future interests (and past ones) has, to put it crudely, more interests to be satisfied or frustrated. Persons who fashion a narrative of their past and future are beings who would lose not only the good of the moment, but other goods associated with it. Consider, for example, that I have an interest in writing a book I have been working on for a year or more. My interest is not merely in the immediate moment of writing. Should I be prevented from writing this moment—say I am killed by a lightening flash—I would lose not only the immediate interest in writing, but also the interest I have in realizing the labours of the past, in the completion of the book, and in the satisfaction of having, in the future, an audience who may benefit from my work. All these interests are at stake. On the other hand, a creature who lives entirely in the moment has presumably only the interests of that moment. If it is killed, it is denied only that one interest. These beings, according to this view, are nonpersons. So, crudely speaking, creatures who are persons have more to lose if they are killed than nonpersons. (As we will see, McMahan puts a fine point on this idea). This absolves Singer of being speciesist and allows him to assert that the killing of the impaired infant and the pig (both equally nonpersons)—for the purposes of obtaining an organ to transplant into a person—is morally equivalent. Both individuals here are nonpersons. Note the contrast between the personist view and the idea of a human being as a species-being. To think of a human being as a species-being is to consider the individual human in relation to other human possibilities and other human beings (past, present, and future). The personist asserts the moral priority of persons because the person can conceive its individual present self in the past and future. As we have already noted, the moral consequence of ignoring species membership, and substituting equality of interests for human moral equality, falls ²¹ This may be a mischaracterization of nonhuman animals. Birds build nests for their future young. Chipmunks gather acorns in the spring which they store in their underground larders, or steal them from others if they have been too imprudent to gather acorns in time. Elephants return to the site of their dead comrades. Geese assign one goose to look after multiple broods of goslings so that the mothers can fish for food. These are all signs of their awareness of their future, their past, and other members of their species. The difference between our own and other species may be one only of degree.
Eva Feder Kittay
273
hardest on people with severe cognitive disabilities, especially when these are congenital. Singer brought this point home in an exchange with Alice Crary and me (Kittay 2018). He wrote that when he is lecturing at a university or to the public, he quotes a passage from Bentham that asks why we give humans a moral importance that we deny to animals. The quotation ends with the lines: ‘a full-grown horse or dog is beyond comparison a more rational, as well as a more conversable animal, than an infant of a day, or a week, or even a month, old’ (1789). Addressing his students’ objection that an infant normally eventually does develop rationality, he resorts to the ‘obvious reply’: ‘[T]here are some humans who … do not have that potential’, namely, congenitally severely impaired human beings. Singer asks rhetorically: ‘Should I forego this response in order to avoid a comparison to which some take offense, and that may lead to reduced concern for people with disabilities?’²² Well, yes. The comparison is not merely ‘offensive’. It is genuinely harmful—an outcome he appears to accept in that last sentence. The answer that is far more appropriate is the one that Bentham himself gives in his very next sentence: ‘The question is not, can they reason? Nor, can they talk? But can they suffer?’ How odd that Singer doesn’t avail himself of that reply when challenged that infants can eventually develop reasoning capacities. But that reply seems to leave people like himself and his readers (that is, persons) on the same footing as nonhuman animals. Singer is not prepared to go that far (Singer, 1999). Nor is Jeffrey McMahan, the other currently prominent moral individualist.
7. Grounding Moral Individualism in a Metaphysics While Singer highlights the moral evil of species prejudice, McMahan is most interested in defending moral individualism within a comprehensive ethics of killing. The question of moral status becomes: who is more justifiably killed than others? To begin, he asks: “What is it about ‘us’ that makes killing one of us a great wrong?” Among the ‘us’ he includes himself and his readers. ‘We’, he says, are persons. What is it about persons that makes it worse to kill a person than a nonperson (2003: vii)?²³ Human beings who are not ‘us’ but ‘them’ ²² Cited in Kittay (2018); see also Brison (2021: 99–122) for a discussion of why agreement could not be reached. ²³ There is a puzzle that McMahan’s metaphysical grounding is meant to address. If we say with Singer that equality should be understood as equality of interests, and interests are, in a sense, free-floating— untethered to the individual to whom they belong—then hunger is hunger, whether it is the hunger of my
274
Basic Human Moral Equality
are presumably the cognitively disabled who, like nonhuman animals, lack that ‘rich psychological architecture structure’ that both makes ‘us’ persons and is requisite for a good life. As their lives have less goodness than that of a person, it is less bad to kill one of ‘them’ than one of ‘us’. McMahan muses that this result is not so bad even for ‘them’, since capacity for goodness in their lives is already so diminished that the benefit they receive from continuing to live is less than a person would receive. McMahan’s conclusion is as troubling as Singer’s claim that not all human beings are equal. Elsewhere I have attempted an elaborated refutation of McMahan’s position (Kittay 2006). I will confine myself to addressing the position McMahan calls Convergent Assimilation Since we do not treat animals as well as we should, we should raise their moral status, and correspondingly lower that of the congenitally severely cognitively disabled. His argument for Convergent Assimilation hinges partially on his claim that if the goodness of a life is sufficiently subpar, its loss and its moral worth are diminished. Let us set aside this troubling conclusion, as well as all the empirical data that shows that people with disabilities report the same level of goodness in their lives as non-disabled people (Albrecht and Devlieger 1999), and ask, is it even coherent to compare the goodness of very different sorts of lives? Such comparison requires that the capacity goodness be comparable and measurable across many different sorts of lives, and, furthermore that the goodness of one’s life is relatively fixed by intrinsic capacities. They appear to hold that by confining themselves to intrinsic properties, we are provided with such criteria. Now, consider an individual with a learning disability, an intrinsic impairment that limits the person’s ability to learn in standard ways. She fails at school, and all assume such failure is due to her incapacity (or lessened capacity) to learn. And so it is, were there only one prescribed way to learn. But a skilled teacher or therapist intervenes, helps her use different learning techniques, and demands accommodations from the school. She becomes an A-student and eventually a heart surgeon. Our learning-disabled individual may feel more fulfilled once her learning disability is properly dealt human neighbour or a skunk that has taken up lodging in my backyard. We would have an equal obligation to feed both. Neither Singer nor McMahan believe this. McMahan means to give us a plausible account of why the two situations are not equal. He introduces the idea of time-relevant interests—interests we have at any given time. To the extent these are strongly connected with the ‘time-slices’ of our former and future selves, as they are in persons, they count for more than that of the nonperson in whom the interests at any given time (that is, time-relative) are discounted as theirs are not. This is a metaphysical ground for what we have already seen in Singer. Hunger may be hunger regardless of who is hungry, and equality of interests supposes that we should treat these two hungers the same. But if we don’t feed the skunk in our backyard, we frustrate only its present concerns. If our neighbour is left hungry, the effect extends to her ability to engage in future projects.
Eva Feder Kittay
275
with—maybe even happier; but was she any less morally worthy before than after? Clearly not. Had she had these difficulties when people misunderstood learning disabilities, and did not realize that accommodations would help or that the impairment was as reparable as a heart defect,²⁴ would she then have had less moral worth because of this historical contingency? Clearly not. What about those congenital conditions that have a genetic origin—a pathological variant on a gene or set of genes? The assumption is that the potential for the ‘rich psychological architecture’ is absent. Such a belief is in line neither with current medical knowledge nor practice. It ignores the elasticity of the brain, as well as advances in surgical and genetic technology such as CRISPR/Cas-9, with its capability to allow physicians to alter a mutated gene—if not now, then eventually.²⁵ Furthermore, the ‘fixes’ needed are likely ones that transform an ableist society into one that is fully inclusive and more appreciative of human variability. Human capacities may not be infinitely alterable, but there is not much that is written in stone. What is possible for one human being is within the realm of possibility of any human being. (What is written in stone is that a nonhuman cannot, even with genetic manipulation and transgenic advances, become a human being.) If levels of goodness are mutable, then we are basing gradations of moral status on shifting sands. It could scarcely be a measure of moral worth. Transgenic comparisons of goodness—saying the life of a cognitively disabled individual is comparable in goodness, as well as cognitive capacity, to a nonhuman animal—are still more questionable. John Stuart Mill wrote that he would rather be Socrates unsatisfied than a pig satisfied. But Mill spoke as a man, not as a pig. Korsgaard makes this point eloquently: ‘A non-human animal, I believe, lives in a world that is in a deep way her own world, a world that is for that animal …’. (2018: 146). Transgenic comparisons depend on analogies and homologies that result from the evolutionary process. However, evolution did not yield a singular line, but a tree with many branches. Frans de Waal writes: ‘[E]very species has a different cognitive story to tell. Each organism has its own ecology and lifestyle’ (2017: 267). If we can say that an octopus has a brain in each of its tentacles, why not say that a dog’s sense of smell is a seat of canine intelligence?²⁶ Would it enhance our wellbeing if we acquired the dog’s sense of ²⁴ I do not want to be misunderstood as saying that it is only if or when people with cognitive impairments have such a ‘fix’ would they be moral equals. I am putting forth such considerations only to argue on the turf of the moral individualist. ²⁵ This discussion is not meant to endorse the desirability of so ‘fixing’ a disabled person, but to indicate that our capacities and ability to have goodness in our lives are not immutable. ²⁶ Dogs have more than 100 million sensory receptor sites, while humans have 6 million. The canine brain size devoted to the olfactory sense is forty times larger than the comparable part of the human brain.
276
Basic Human Moral Equality
smell? Would the dog be better off were he to acquire an ability to read? A dog’s sense receptors would make human life unbearable—the world is too odoriferous. A dog’s olfactory sense gives him the information he needs for his umwelt (De Waal 2017). What good would reading do? If we lack solid grounds to put the thumb on the scale for a person, then whether the dog’s capability for a rich life is greater or lesser than that of a human being with cognitive disabilities is of little moral importance. In a conference paper delivered at The Challenge of Cognitive Disability to Moral Theory conference in New York in 2008, McMahan listed the various ways that nonhuman animals were intellectually superior, if not on par, with people with severe cognitive disabilities. He compared the productivity of domesticated animals to the apparently low productivity of the severely cognitively disabled. Is our measure then productivity? Shall we call unproductive humans ‘nutzloser Essers’, ‘useless eaters’? The Nazi regime paved the way for the extermination of mentally disabled people by calling them nutzloser Essers. None of us, not Singer nor McMahan nor other moral individuals, want to go down that rabbit hole. Neither intelligence nor productivity is the source of moral worth.
8. Intrinsic Properties Versus Relational Properties Moral individualists want to make an individual’s worth contingent only upon their intrinsic capacities. To appeal to one’s usefulness to others relies on a being’s relational properties—a nonstarter for the moral individualist, as for many other philosophers. Special relations, we noted above, are regarded as an exception, and species membership, they claim, is not a special relation like a parent-child relation. Moral individualists insist that appealing to a special relation we have with members of our own species is speciesist. Singer, replying to Bernard Williams’s (2006) defence of ‘the human prejudice’, insists that Williams is saying nothing different than a white supremist who believes one is justified to have a prejudice for your own white race. McMahan makes a similar point with respect to nationalism. It seems to me that the waters here are muddier than these philosophers want to make out. If Singer and McMahan are right, then in the AngloEuropean context at least, not only is it wrong for whites (the dominant McMahan and Singer may deny that the canine sophisticated olfactory architecture is intelligence. But why not? What is processed is not sheer sensation. It includes memory and problem-solving accomplished through the sense of smell. See Kokocińska-Kusiak et al. (2021: 2463).
Eva Feder Kittay
277
group) to privilege whites, but it would be equally racist for a Black person to feel special obligations²⁷ to improve conditions for Blacks. However, while we consider the white’s claim to be racist, we view the Black’s to be not only justifiable, but commendable. The same moral messiness applies to nationalism. While it is surely wrong to hoard vaccines in the name of ‘Country First’, it is less clear that a leader ought not first to hand out vaccines to the nation he serves. And while serving in aggressive wars is at best morally ambiguous, it cannot be morally wrong to want to defend one’s own country against attacks. This muddle applies to all special relations. It is certainly not wrong to feed your child first, but it is wrong to hoard food when another child needs it. It is certainly right to vigorously defend your child when bullied or attacked by another child; but it is not wrong to side with the other child, if it is your child that is doing the bullying. The morality of special relations comes into play, argues Robert Goodin (1985), when another is vulnerable to your actions, or when you are in a special position to help a vulnerable person. By becoming a parent, you make a child vulnerable to your actions. A Black person in a racist society is uniquely positioned to help other Blacks fight back against racist oppression. We are in special relations with nonhumans as well, especially those whom we have made dependent on our actions, such as pets and livestock. Special relations do not absolve one from injurious behaviour towards another outside the relationship if the other’s actions are not harmful to those with whom you have special relations.²⁸ Furthermore, the moral claim to protect the interests of particular individuals is justified when others are also in such special relations. We take care of our own child, not every child, on the expectation that other children have parents (or some persons) to correspondingly care for them. Likewise, we have many special relations at any given time. I not only have a special relation to humans, but I also have special relations with nonhuman animals and a natural environment that we all depend upon. However, as special relations and obligations based on our relations to others are not agent-neutral, they bind no one except those in the relationship. Moral worth, which needs to be recognized by those not in relation, then cannot be based on relational attributes. McMahan, at the conference mentioned ²⁷ For a distinction between racism and racialism, see Appiah (2005). ²⁸ A relevant suggestion is found in Qi (2016): ‘Special obligations are positive obligations, as opposed to negative obligations … For instance, our special obligations to our beloved ones should not be met at the cost of unjustly frustrating others’ comparable interests’. He claims this can direct us to a ‘mitigated speciesism’. An ethic of care can be characterized, mostly as an ethic of special relations, but in Kittay (2019) I argue that in a fully normative ethic of care we are to caringly attend to another, in ways that do not impede the ability of another to do likewise.
278
Basic Human Moral Equality
above, said to me something like the following: ‘Eva, you may have moral obligations to your congenitally severe cognitively disabled daughter, but what about her gives her a claim on the rest of us?’ McMahan grants that society might have a responsibility to me (as a moral person) to treat my child well. His supposition, of course, is that the child does not have equal moral status, so why should she have any moral claim on ‘us’? My response is that one cannot both hold that my relationship to my daughter is the special relation of mother to child without also recognizing the disabled daughter as having the same moral status as any other daughter. I, as a parent, have obligations to fulfil towards all my children, regardless of capacities. And to meet my obligations and responsibilities I require resources, many of which are provided by the larger society: protective regulations to help ensure the child’s safety, schools to ensure that my child can develop whatever her capacities may be, a social world into which she can be integrated and find acceptance, respect from others and the wherewithal to develop self-respect. No child is simply the parent’s own private matter. If my relationship to my child—that is, the parent-child relationship— is acknowledged as a morally important one, then how can it be consistent to deny me the means to fulfil my morally significant parental obligations (obligations that I have to all my children regardless of their intrinsic features) because my child is not recognized as an equal? It is inconsistent (even incoherent) to grant the special relationship I have with my daughter, and then to turn around and say: but that daughter has no moral hold on anyone but her parent. Her parent cannot act responsibly as a parent—that is to raise a being to be accepted by the community of moral equals—unless others also have an acknowledged moral responsibility to the child, a moral responsibility on par with that they have to anyone’s child. And it is not that I, as a parent, want people to care about my daughter because I care about her. I want and need to have my child recognized for her sake. That is the nature—one may say the core—of the parental relationship. I cannot give her the care it is my duty to provide if others do not respect her as a being worthy of the same care as is due to any child. Each of us has needed the care of another, who in turn is supported—at least minimally— by a larger society that recognizes a child’s need for another’s attention to survive and thrive—all of which are relational properties, the primary one being that we are all ‘some mother’s child’. As that is so, we cannot do without relational properties in considering moral worth—it is baked-in, given the human condition. Equal moral worth is just the equal consideration to every ‘mother’s child’.
Eva Feder Kittay
279
9. What About ʻUsʼ is Morally Relevant? At the start of McMahan’s inquiry, he asks: what about ‘us’ is morally relevant? Philosophers want to give an answer acceptable to a secular audience and so eschew answers such as ‘we are all equally children of God’. As they turn to those attributes in ourselves that ‘we’ recognize as morally significant, they encounter the difficulty that the assignment of any attribute is either too exclusive or too inclusive. In secular discussions it is rationality that traditionally gets invoked. For the moral individualist, not all humans are rational, or have associated higher cognitive abilities, whereas some nonhumans do. As reason is crucial to being a moral agent, only those who are like us in that sense are persons. But many, including me, refuse to leave certain humans to the side—to argue that they are not of equal moral worth. So, shall we continue to look for the morally relevant features of ‘us’ that people such as my daughter does have and that other beings lack? Unless we identify these, we are being speciesist—merely privileging our own. I have attempted to show that ‘privileging our own’ is not always morally problematic.²⁹ Jaworska and Tannebaum (henceforth J&T) have tried to avoid the difficulty of requiring rationality. They summarize their view as follows: [T]he basic tenet of our proposal is that if a certain cognitively sophisticated capacity, what we called SSP [self-standing person] capacity, can justify a beingʼs higher moral status, then the capacity to incompletely realize this cognitive sophistication can also justify a beingʼs higher moral status. (2015)
The capacity they have in mind is being able to participate as a rearee in a care-rearing relationship. They emphasize that this is an intrinsic, not a relational, capacity. The rearee must be able to participate in ways that aim at the goal of becoming a self-standing person, even if the goal may never be completely realized.³⁰ J&T have crafted their definitions to be both non-speciesist and yet inclusive of all—or almost all—human beings. People with Alzheimer’s, up until near their very end, are persons as they have had the capacity to participate ²⁹ In other work (Kittay 2016: 715–741) I make the further case that ‘privileging our own’ is in itself not the pernicious aspect of group-based bias. To be pernicious, such bias requires that the in-group claims for itself all valued attributes, denying these to other groups and attributing to them all the devalued and despised properties. ³⁰ They insist also that this must be the same rearee—that is, nothing in their metaphysical status has changed.
280
Basic Human Moral Equality
in such rearer-rearee care relations. However, they will not be self-standing persons. The definition would also include those who are in a coma and perhaps those in a persistent vegetative state. It should exclude all foetuses except those who are viable outside the pregnant woman’s womb.³¹ They mean to include people such as Ashly X, a severely cognitively and physically disabled girl, who gained notoriety when her parents chose growth attenuating treatment to keep her small.³² The only exclusion is those humans born with a genetic mutation. J&T presume that Ashley’s condition is not genetic. But what if it turns out to be genetic—does the previous candidate of SSP, and so a moral equal, cease to be one? My own daughter, whom we raised lovingly and in a manner that would take full advantage of her capabilities, was diagnosed at age 48 with a de nova genetic variant—a single mutation on one allele of the PURA gene. This results in an pur-alpha protein insufficiency, a condition responsible for all her cognitive and physical impairments. Our first reaction to the news was ‘this changes everything’, but we quickly saw that it changed nothing. Our daughter Sesha was still the wonderful daughter she had been prior to the diagnosis. It changed not a hair of how we thought of her, considered her, or loved her. What’s more, we also realized that there were aspects of her functioning that were either not affected by the genetic condition, or—given the plasticity of the brain—were compensated for by other parts where the pur-alfa protein played a lesser role. There were even possible gains—such as her remarkably sweet temperament, her lack of aggressiveness, and her incredible soft yet healthy skin. While J&T’s criteria for full moral status succeeds in being more inclusive than those based on cognitive capacities, it still divides human beings into those with and those without such status. And while the exclusion may be small, it may be enough to allow the camel’s nose under the tent. The camel’s nose is the pernicious exclusion of some human beings based on criteria decided by a group with the power to make delineations between the truly human and the marginally human, between those with full moral status and those whose moral status is reduced to that of nonhuman animal. That camel’s nose is not a mere hypothetical worry. Increasingly scholars are recognizing that the Nazi murder of approximately 200,000 mentally
³¹ Although it does leave open the door for someone to argue that the foetus in another’s womb is already in a rearer-rearee relationship. Though I cannot see how the argument could be conclusive as it would be just as easy to deny that this is not such a relationship until the foetus is born—and it is not clear what could settle such a dispute. ³² See Kittay (2011: 610–631).
Eva Feder Kittay
281
disabled people served as a prelude to the murder of 6 million Jews, and half a million Roma and Sinti people. A historian of the Nazi killings writes that: According to Wagner, [the Führer of the National Socialist Physicians League] the doctrine of equality was even worse in its effects than the Russian revolution insofar as it led one ʻto value the sick, the dyingʼ and the unfit on a par with the healthy and the strong. Wagner maintained that while much of the world clung to the ʻinsane idea of equalityʼ Germany had begun to recognize ʻthe natural and God-given inequality of menʼ. (Proctor, 1988 181 Emphasis mine).
Singer rejects a slippery slope argument and maintains that the genocides of the Nazis including the attempt to eliminate those ‘unfit’ was an aberration, a confluence of historical events that are unlikely to occur again. I get little comfort that racial and religious exclusions do not resemble (to our contemporary mind) the denial of the equality of the severely cognitively disabled. I do not share Singer’s optimism that an equally pernicious confluence of forces is highly unlikely. Notably, both the racial genocides and the murder of the mentally disabled (regardless of their Aryan pedigree) were justified by writings taken from the eugenics movement in England and the US at the time (Kittay 2016). Eugenics had already linked the two populations as inferior beings. We have seen too many genocides since the Nazi’s, and we have witnessed, just recently, the way in which those with mental disabilities have been relegated to the end of the list in triage conditions during COVID-19. To deny human equality does not lead us along a slippery slope; it is a walk along a precipice. A misstep easily leads to the abyss. The idea of the moral equality of all human beings is a guardrail against the worst sorts of brutality, albeit a rickety one. Only the full-throated commitment to basic human equality can secure that guardrail—can respond to the real-life threats to the vulnerable.
9.1 Who is the ʻUsʼ? Still the question remains: is there a basis for the commitment to basic human moral equality other than the historical and ongoing threats of discrimination and genocide? I believe there is. Consider how McMahan determines what traits are morally relevant to personhood (that is, full moral equality): what is it about ‘us’ that makes us persons? That’s a reasonable way to begin. But who is ‘us’? What if the edifice of moral status gradations intended to
282
Basic Human Moral Equality
be universal is built on too few data points, and so cannot be universally valid—or valid at all? Newtonian physics was based on how entities appear to move on earth. But with Einsteinian and quantum physics, we understood that they are valid only for midsized objects on this planet. Just as a theory that is adequate to our planet with its mid-sized objects does not give us the fundamental laws of matter, so the ‘we’ McMahan invokes is not an adequate measure when that ‘we’ is limited to only a sample of human beings. How then should we proceed? I suggest we start with the whole of humanity—those born humans with whom we share our world. Why privilege humanity? Because it is possible to have epistemic access, and so moral access, to human beings that we can have only derivatively in the nonhuman realm. By moral access, I mean access to what is valuable to the other in a way that makes us recognize the other’s right to flourish. Some (not complete) epistemic access is a prerequisite to such an understanding. But how, might the philosopher ask, can I have such access to someone who doesn’t speak, with whom I share nothing but the bare fact of our common membership in the human species? This is one response: as remote as it may seem, should I become so disabled, I then would know what it is like to be that individual. As I could never become a nonhuman animal, even such possible knowledge is opaque to me. The moral world as we know it is not a creation of that nature that is ‘red in tooth and claw’.³³ It is human beings who create a conception of morality for themselves. Our epistemic access to the life of people who do not look or think like ‘us’ may be limited. And yet we can become any one of them at least in the sense that there is some near possible world in which ‘we’ are ‘them’.³⁴ This is especially true in the case of disabling conditions. The most intellectually brilliant of the ‘us’ can be struck by a truck, rendering us as cognitively disabled as my daughter. The etiology may be different, but the consequence is similar. Similarly, although I am a privileged white woman in a racist society and my skin cannot turn black, under different circumstances I may find myself among those shunned as a despised minority. In the time of the Irish famines, the Irish, despite their pale complexion, were disparaged as the Negroes of Europe (Engle 2001). Of course, we share experiences with ³³ The idea that nature is such a violent world of survival of the fittest is being challenged as we learn more about our fellow creatures. We have come to recognize the evolutionary advantage of cooperation within species and even across species. We have come to recognize the extent to which animals care for injured or impaired others. For a fascinating discussion, see Peña-Guzmán (2023). ³⁴ Note how this view differs from Kagan’s (2016: 1) possible world morality. His supposition is that they can become us. While this is not an unreasonable supposition, it holds fast to the idea that the philosopher’s ‘we’ is the moral measure. My view is that we can become them. This eschews the idea that we already know what is and isn’t morally relevant until we take a broader view of who the ‘we’ may be.
Eva Feder Kittay
283
nonhumans: death, pain, pleasure, reproduction—but even these remain epistemologically opaque to us. What do we know of a dog’s misery when deprived of a rich olfactory environment—what could we know? Therefore, although other human beings are often epistemologically opaque to us, we have a moral access we lack with other beings. We have a sense of what it is to harm or benefit someone like us, but we need to enlarge our understanding of what it is to benefit or harm someone who is not like us—but someone whose circumstances we could inhabit. To depend only on what we learn from the ‘us’ means we will get a lot of things wrong. To build more equitable political and social institutions, we need to understand the morally important needs and wants of people of different sorts. We have gotten many things wrong. We overvalue the significance of a calculative rationality and independence, just as we undervalue inevitable human dependency. We misunderstand the value and significance of distant others. These are all big claims, but they are consonant with that philosophical thought that recognizes both the limitations in our Western philosophical canon and our incomplete understanding of the human organism. Consider how much has been learned about ‘normal human functioning’ by looking at what we deem pathological. Moreover, if we find ourselves no longer included in the philosopher’s ‘we’ and are cast into the moral shadow world—the world of the marginal humans—we will have justified being treated as one with lesser moral worth. If life outside the ‘we’ is a possibility for ‘us’, and if we, or our loved ones, should then be viewed as having lesser moral value, then we have no justifiable complaint to make. We would have created a self-defeating theory, been guilty of a contradiction in practical reasoning. There is an obvious objection to my argument. McMahan would reply that when I find myself in radically different circumstances, ones in which I lack those morally relevant properties, ‘I’ am not the same individual I was before. My prudential relations to that earlier self are weakened, and so I should have less concern for that self. Therefore, there is no practical contradiction—the ‘I’ who judges a given level of cognitive function to be morally necessary for moral personhood is not the same self who is now without the necessary capacities. This metaphysics has some theoretical coherence but deviates in important ways from how we do (and I believe ought to) treat each other. If my mother has advanced Alzheimer’s, she is still the person who reared me, and whatever my love and my obligations are to my mother, all tattered by the disease, they are still directed at this individual who mothered me. The ‘I’ that we are remains, even as we lose those capacities on which we hung our identities.
284
Basic Human Moral Equality
Were I a great beauty, so had tied my identity to my physical appeal, and then contracted smallpox that left my face severely scarred, I would have to look at myself anew, find new meaning in my life. However, it is still me who is doing the recalibration. This is no less true of a brain injury or a personalityaltering trauma. If my mother died of advanced Alzheimer’s disease, and I constructed a narrative of her life, I would say that it was the Alzheimer’s that killed her, my mother, not some other individual. This is no less true of a child born with congenital disabilities. That child is still some mother’s child, and if it is my child, she remains my daughter—a relationship just as important as that another mother has to her daughter. If any other daughter is—morally speaking—a person, I cannot view her as my daughter without viewing her as having the same moral standing I and other daughters have. Her being born human,³⁵ a human daughter, must suffice for us to give her full moral status. Still more important, unless we regard each human infant (born in a condition that sustains life) as an equal in moral worth, we will not understand the full range of morally relevant attributes and relations. Without this understanding, we fail to truly understand ourselves.³⁶
References Albrecht, G. L., and Devlieger, P. 1999. ‘The Disability Paradox: High Quality of Life Against the Odds’. Social Science and Medicine, 48: 977–988. Appiah, A. 2005. The Ethics of Identity. Princeton, N.J.: Princeton University Press. Bentham, J. 1789. An Introduction to the Principles of Morals and Legislation. Oxford: Clarendon Press. Brickell, C., and Munir, K. 2008. ‘Grief and its Complications in Individuals with Intellectual Disability’. Harv Rev Psychiatry, 16/1: 1–12. Brison, S. J. 2021. ‘Valuing the Lives of People with Profound Intellectual Disabilities’. Philosophical Topics, 49/1: 99–122. Clute, M. A. O. 2017–2011. ‘Living Disconnected: Building a Grounded Theory View of Bereavement for Adults with Intellectual Disabilities’. Journal of Death and Dying, 76/1: 15–34. ³⁵ I argue in Kittay (2021: 79–103) that the sufficiency condition for full moral status is being born a human being in a condition that is compatible with life. Note this is a sufficiency condition. To argue for full moral status of foetuses (especially prior to viability) or an anencephalic neonate, we would need another argument. ³⁶ I would like to thank Peter Singer, the undergraduate students in the Human Values seminar, and the graduate students of his class, Peter Singer and his Critics, for our discussions. I feel confident that I have represented his views better in this chapter than in previous articles and chapters.
Eva Feder Kittay
285
Czank, J. M. 2012. ‘On the Origin of Species-Being: Marx Redefined, Rethinking Marxism’. Journal of Economics, Culture and Society, 24/2: 316–323. Deutsch, H. 1985. ‘Grief Counseling with the Mentally Retarded Clients’. Psychiatric Aspects of Mental Retardation Reviews,4(5): 17–20. Engle, A. 2001. ‘Depictions of the Irish in Frank Webb’s “The Garies and Their Friends” and Frances E. W. Harper’s “Trial and Triumph”’. MELUS, 26/1: 151–171. Fosse, B. 1972. Cabaret. United States: Allied Artists Pictures. Goodin, R. E. 1985. Protecting the Vulnerable: A Reanalysis of Our Social Responsibilities. Chicago: University of Chicago Press. Harper, D. C., and Wadsworth, J. S. 1993. ‘Grief in Adults with Mental Retardation: Preliminary Findings’. Research in Developmental Disabilities, 14/4: 313–330. Hull, D. L. 1976. ‘Are Species Really Individuals?’. Systematic Biology, 25/2: 174–191. Jaworska, A., and Tannenbaum, J. 2015. ‘Who Has the Capacity to Participate as a Rearee in a Person-Rearing Relationship?’. Ethics, 125/4: 1096–1113 Kagan, S. 2016. ‘What’s Wrong with Speciesism? (Society of Applied Philosophy Annual Lecture 2015) (Report)’. Journal of Applied Philosophy, 33/1: 1. Kittay, E. F. 2006. ‘On the Margins of Moral Personhood’. Ethics, October: 166: 100–131. Kittay, E. F. 2010. The Personal is Philosophical is Political: A Philosopher and Mother of a Cognitively Disabled Person Sends Notes From the Battlefield. Oxford: Blackwell. Kittay, E. F. 2011. ‘Forever Young: The Strange Case of Ashley X’. Hypatia, 26/3: 610–631. Kittay, E. F. 2016. ‘Deadly Medicine: The T4 Project, Disability, and Racism’. Res Philosophica, Special Issue on Disability, 93/4: 715–741. Kittay, E. F. 2018. ‘Comments on Alice Crary’s The Horrific History of Comparisons between Cognitive Disability and Animality (and How to Move Past It) and Peter Singer’s Response to Crary’. Zemo. Kittay, E. F. 2019. Learning from My Daughter: The Care and Value of Disabled Minds. Oxford; New York: Oxford University Press. Kittay, E. F. 2021. ‘Why Human Difference is Critical to a Conception of Moral Standing: An Argument for the Sufficiency of Being Human for Full Moral Status’. Journal of the Philosophy of Disability, 1/1: 79–103. Kokocińska-Kusiak, Agata, et al. 2021. ‘Canine Olfaction: Physiology, Behavior, and Possibilities for Practical Applications’. Animals: An Open Access Journal from MDPI, 11/8: 2463. Korsgaard, C. M. 2009. ‘The Activity of Reason’. Proceedings and Address of The American Philosophical Association, 83/2: 23–44. Korsgaard, C. M. 2018. Fellow Creatures: Our Obligations to the Other Animals, 1st ed. Oxford: Oxford University Press.
286
Basic Human Moral Equality
Lawrence, D. H. 1979. The Letters of D. H. Lawrence. Vol 1, The Cambridge edition of The Letters and Works of D H Lawrence. Cambridge: Cambridge University Press. Locke, J. 1690. Essay on the Human Understanding. London. McMahan, Jeff. 2003. The Ethics of Killing: Problems at the Margins of Life. Oxford; New York: Oxford University Press Peña-Guzmán, D. M. 2023. When Animals Dream: The Hidden World of Animal Consciousness. Princeton University Press. Proctor, Robert. 1988. Racial Hygiene: Medicine Under the Nazis. Cambridge, Mass: Harvard University Press. Putnam, H. 2002. The Collapse of the Fact Value Distinction. Cambridge: Harvard University Press. Qi, E. X. 2016. ‘Special Relations, Special Obligations, and Speciesism’. Bangladesh Journal of Bioethics, 7/3: 12–22. Rachels, J. 1999. Created from Animals: The Moral Implications of Darwinism. Oxford, UK: Oxford University Press. Sarao, K. T. S., and Long, J. D. E. eds. 2017. Buddhism and Jainism. Dordrecht: Springer. https://doi.org/10.1007/978-94-024-0852-2_1894. Singer, P.,Kuhse, H. 1988. Should the Baby Live?: The Problem of Handicapped Infants. Oxford; New York: Oxford University Press. Singer, P. 1994. Rethinking Life and Death: The Collapse of Our Traditional Ethics. New York: St. Martin’s Press. Singer, P. 1999. ‘Reflections’. In The Lives of Animals, edited by A. Gutmann, 85–92. Princeton: Princeton University Press Speece, M. W., and Brent, S. B. 1984. ‘Children’s Understanding of Death: A Review of Three Components of a Death Concept’. Child Development, 55, 1671–1686. De Waal, F. 2017. Are We Smart Enough to Know How Smart Animals Are? New York, London: W. W. Norton &Company. Williams, B. A. O. 2006. ‘The Human Prejudice’, in Philosophy as a Humanistic Discipline. Princeton, N.J.: Princeton University Press, 135–155.
12 First Among Equals George Sher
It is widely agreed that all normal humans have equal moral standing, but there is no comparable agreement about why this is so. In a previous paper,¹ I attributed our moral equality to the fact that each has a subjective perspective of his own—that we are, each of us, complete worlds unto ourselves. However, many animals also have subjective perspectives, so this proposal appears to imply that cows, cats, and rats, and perhaps also vultures, garter snakes, and carp, have the same moral standing as humans. That implication would be hard to accept if it meant that we are never permitted to favour humans over animals; but in what follows, I will argue that we can accept the implication without having to draw any such conclusion.
1. Moral Standing When I say that a being has moral standing, what I mean is that its interests are among the factors that determine what morality requires of us. Although moral standing can also be glossed in more deontic terms—for example, as involving the possession of certain non-derivative rights²—this way of putting things will only be a genuine alternative if it is possible to specify the relevant rights without alluding to the interests they protect. Because I view that as unlikely,³ I will stick with the simpler ‘having interests that matter’ formulation. Under this interpretation, two beings will have the same moral standing if their interests count equally towards what morality requires. ¹ That paper, entitled ‘Why We Are Moral Equals’, first appeared as chapter 5 of my 2014 volume. Slightly altered versions of it also appear in Steinhoff (2015: 17–29), and in my 2017 volume, at 31–46. ² This possibility is raised by Richard Arneson in his (2015). ³ Even if we accept some version of the will theory of rights, we will have to explain why our normative control over the activities of others takes the form it does; and it is hard to see how we can do this without alluding to the interests that the relevant forms of control protect. When Arneson discusses the rights approach, he telling speaks of ‘my interest in freedom of movement [generating] a right on my part to freedom of movement’ (2015: 50). George Sher, First Among Equals. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0013
288
First Among Equals
Although the moral equality of persons is an axiom of most contemporary moral philosophy, it has proven hard to defend. To justify it, we will presumably have to locate some feature that (1) confers moral importance on the interests of its possessors, and (2) is possessed to an equal degree by every normal human. This last requirement is problematic because people notoriously differ along every known empirical dimension. We are not all equally smart, equally strong, equally clever, equally nice, equally moral, equally rational, or equally anything else. Thus, if any one of these traits, or any combination of them, were what determined a person’s moral standing, then it would apparently follow that all people do not have equal moral standing. To avoid this difficulty, some have proposed that what determines a person’s moral standing is not how high he scores in any given dimension, but only whether his score in some dimension surpasses a crucial threshold.⁴ By advancing this proposal, these philosophers have sought to convert some relevant scalar property—intelligence, perhaps, or responsiveness to reasons—into one whose possession is all-or-nothing. However, to be convincing, their proposal would have to be backed by an explanation of why the difference between the points just above and below the proposed threshold is any more important than the difference between any other two points on the continuum. We may also wonder why we should focus exclusively on this ‘range property’ and not at all on whichever scalar properties constitute its base. As far as I can see, these questions have never been adequately answered. Thus, pending some further development, we are thrown back on the search for a property that is genuinely all-or-nothing—a property that everyone who has it must have to the same degree because it only comes in one degree. Within the public world, the only properties that fit this description appear to be generic properties such as being embodied, which seem incapable of bearing the argument’s weight.⁵ But it is surely significant that the thing whose loss we fear the most—our continued existence as a subject of consciousness—is not a part of the public world at all (although of course it depends on many things that are). Because the continuity of each person’s subjectivity matters so greatly to him, and because so much else of what matters to him requires its continuity, it is natural to wonder whether the all-or-nothing property in virtue of which each person’s interests have equal moral weight might also be located within his subjectivity. ⁴ See, for example, Rawls (1999: 444) and Waldron (2017: Ch. 4), and passim. ⁵ Cf. Arneson (2015: 41).
George Sher
289
Moreover, as soon as we make this inward turn, we encounter a property— namely, that of having an interiority at all—that does seem all-or-nothing in just the relevant sense. Although we can indeed describe a person as ‘not fully conscious’ when he is badly hungover or half asleep or just coming out of anaesthesia, even a person in one of these states has an inner world of a kind that a rock or a robot never has, and that he himself lacks when he is out cold or in a deep coma or dead. Even when someone’s consciousness shrinks to the narrowest point—even when he has entirely lost track of where he is, and is aware only of light and sound and pain—his subjectivity remains a complete thing that is distinct both from everything in the public world and from anyone else’s subjectivity. Because the property of having a subjectivity is all-or-nothing, it is possessed to an equal degree by every person who has it at all. Moreover, although there is ongoing controversy about how a person’s interests are related to his desires and other subjective states, it is far less controversial to say that the fulfilment of whatever interests a person does have would matter much less (if indeed it mattered at all) if he lacked an inner life. The ‘interests’ of a being that has never had a subjectivity, and so has never been in a position to care what happened to it, do not seem capable of providing anyone with a weighty reason to act.⁶ And, because of this, it seems safe to conclude that the property of occupying a subjective perspective is at least among the necessary conditions for having moral standing. But even if this is so, it does not follow that every being that satisfies this condition has the same moral standing. There are two reasons for this: first, that to have moral standing at all, a being might have to have not only an inner life, but also some further property that some beings with inner lives lack, and, second, that even if moral standing itself does require only an inner life, the weight of a being’s interests might still vary with some further property that is not all-or-nothing. I think, in fact, that neither of these things is terribly likely, but I need not pursue them because my aim is not to establish that all normal humans are moral equals, but only to specify a property in virtue of which they could be. Like most others, I simply assume that each person’s interests count equally, and so my question is only whether there is anything about us that might explain why that is so. To fill this explanatory role, a feature must both belong to each person to an equal degree and be intimately linked to the interests whose moral weight is in question.⁷ As far as I can see, the only feature that satisfies both desiderata is that of having an inner life. ⁶ Although this appears to be the standard view, there are some who reject it; see, for example, Kagan (2019: 16–30). ⁷ For related remarks, see Waldron (2017: 135–137).
290
First Among Equals
Thus, my explanation of why each person’s interests matter equally is simply that each of us is a world unto himself.
2. Humans and Animals But just as all normal humans have inner lives, so too, as far as we can tell, do all the normal members of many animal species. Because no being’s subjectivity is penetrable by others, we cannot hope to gain direct access to the inner life of any animal; but just as the obvious physical and behavioural similarities between us and other humans support the view that they too have subjectivities, so too, albeit to a lesser extent, do the physical and behavioural similarities between us and many animals. Even if a person cannot coherently imagine what it would be like for him to be a cat, he can still have good reason to believe that there’s something it’s like for the cat itself. Thus, if what grounds the moral equality of normal humans is simply that each inhabits a distinct subjectivity, then every feline will presumably be the moral equal of every human. Just as the innumerable differences between the thoughts, actions, and capacities of individual humans have no impact on the moral weight of their interests, neither will the far more profound differences between the thoughts, actions, and capacities of humans and those of various animals. At first glance, the claim that some nonhuman animals are our moral equals may appear to imply that we have the same moral duties towards them as we do towards our fellow humans. This implication, if it holds, will be deeply at odds both with many of our intuitive beliefs and with many of our familiar practices. There are indeed many who believe that it is just as wrong to eat a cow as to eat a person, but there are far fewer who would be as exercised by the destruction of a vole’s den to clear a field for a tract house as they would by the razing of someone’s tract house to provide habitat for a vole. Far from intervening on behalf of slaughtered feedstock, dispossessed wildlife, or sacrificed knockout mice, most of us manage to blink even the well-documented cruelties of factory farming. I mention this not as a defence of our treatment of animals—the mere fact that we engage in these practices obviously doesn’t make them right—but only to bring out how much our lives would have to change if we abandoned all of them. But does the proposed explanation of the moral equality of humans really require such drastic changes? Does it really support both the implication that many animals are our moral equals and the further implication that we therefore have all the same duties towards those animals that we have towards one
George Sher
291
another? In the remainder of this chapter, I will consider a number of possible ways of resisting this conclusion. In my earlier paper, I suggested that we might do this by refining our description of the all-or-nothing property that grounds our moral equality. The key to the refinement is the fact that every normal human has both (a) a subjectivity, and (b) a subjectivity that is organized around certain fundamental assumptions. In particular, each of us takes it for granted [T]hat the world is temporally as well as spatially ordered, that [he] himself is an embodied subject who has existed in the past and will exist for at least some time in the future, that various courses of action are open to him, that the world gives him reason to do some things and refrain from doing others, and that he is, within limits, capable both of finding out what reasons he has and of acting on them.8
Because every normal human thinks in these terms, there is room for a view that attributes the moral equality of humans not merely to the fact that they all have inner lives, but rather to the fact that they all have inner lives that are structured around just these assumptions. If this refinement can be sustained, then any animal whose inner life lacks this structure—roughly, any animal that is not self-aware, has no concept of time, and does not inhabit the space of reasons—will not have the feature that accounts for the equal moral standing of humans. Hence, by this criterion, cats and cows will almost certainly not be our moral equals. I had my doubts about this proposal when I advanced it, and I’m even less sure about it now. One obvious problem is that each additional specification of the property that is said to ground our moral equality represents an additional dimension in which that property might turn out to be scalar. Although it is hard to deny that the bare property of having some sort of inner life is all-or-nothing, it becomes progressively easier to deny this when we add in such further elements as reflexivity, reasons-responsiveness, and a sense of time. The case for saying that animal subjectivity entirely lacks these elements is that each element involves one or more concept—that of the future, of oneself, of a reason, etc.—that animals simply don’t possess. However, the case against accepting that case is that the boundaries of these concepts seem fluid enough to leave open the possibility that at least some animals do have rudimentary versions of them. That, at any rate, is one conclusion that we might draw from the facts that all animals engage in purposive activity, that many appear to form and execute simple plans, and that others again appear ⁸ Sher (2014: 82).
292
First Among Equals
to recognize themselves in the mirror or act in ways that bespeak shame or embarrassment. It is of course possible to discount the concepts of temporality, reflexivity, and reasons that inform this behaviour on the grounds that they fall below some threshold of complexity or sophistication; but any move in this direction would open up the possibility that some humans might also fall below the threshold, and so would threaten my explanation of why all normal humans are moral equals. Moreover, quite apart from this, any such attempt to modify the proposed explanation would have the wrong feel. Even if having a subjectivity that is structured around the concepts of time, selfhood, and reasons were just as all-or-nothing as having a subjectivity, period, the division between the beings that have and that lack a subjectivity with this structure would remain far less fundamental than the division between the beings that have and that lack a subjectivity at all. The idea that the world contains billions of private realities seems at once undeniable and irreconcilable with the equally compelling idea that there is a single all-encompassing public world: the project of resolving this tension has preoccupied us for centuries. In comparison, even the undeniably important division between those subjectivities that are organized around such concepts as time, selfhood, and reasons and those others that are not seems relatively trivial. And, for this reason, I remain inclined to attribute the moral equality of humans to the bare fact that each of us has his own inner life—an inclination which if correct will leave intact the implication that animals are our moral equals.
3. Human and Animal Interests But even if they are, the inference that we have all the same duties towards animals that we have towards one another will remain a further step. It is, moreover, a step that the greater complexity of our own inner lives may block; for even if the structure of a being’s subjectivity has no bearing on the weight of its interests, it may significantly affect what those interests are. This last fact is important because the constraints of morality appear to be dictated precisely by the nature of the interests that they protect. In view of this, it would be surprising if any systematic differences in the interests of animals and humans did not show up as differences in the kinds of treatment we owe them. Those who regard animals as our moral equals often exploit the connection between a being’s interests and our duties towards it to defuse the more
George Sher
293
radical implications of their position. That, for example, is what Peter Singer appears to be doing when he writes that It is not arbitrary to hold that the life of a self-aware being, capable of abstract thought, of planning for the future, of complex acts of communication, and so on, is more valuable than the life of a being without these capacities … the value of a life is affected by these other characteristics. Normally this will mean that if we have to choose between the life of a human being and the life of another animal, we should choose to save the life of the human …9
By thus acknowledging that the reflexivity, temporality, and rationality of human thought affect the content of the interests that humans have—that, I take it, is what Singer means by saying that they make human lives ‘more valuable’—he is in effect repurposing these features in a way that enables him to justify treating humans and animals differently without abandoning his commitment to their moral equality. I think this general strategy is a significant step forwards, but I also think Singer’s use of it does not go far enough. As long as we restrict our attention to the ways in which the structure of a being’s consciousness affects its interests, there will not be enough daylight between our own interests and those of animals to accommodate all of the apparent differences in our duties towards them. Interestingly, versions of this problem arise both when the interests that the divergent duties protect are ones that humans and animals have in common and when they are not. Like all normal humans, the members of many animal species have significant interests in remaining alive, being free, not suffering, and securing the wellbeing of their offspring. However, although the interests of those animals are in these respects comparable to our own, the corresponding duties that we acknowledge towards them are not. The standard prohibition against killing humans is far more stringent and wide-ranging than the prohibition against killing animals, we don’t keep people in zoos, we are willing to separate cows from their calves but not parents from their children, and we take our duty to prevent animal suffering to give way whenever it conflicts with our duty to prevent equivalent human suffering. Thus, the question we must now ask is whether it is possible to account for the differences in these duties by arguing that the temporality, reflexivity, and rationality of human thought can somehow intensify the interests that the duties protect. ⁹ Singer (1975: 21–22).
294
First Among Equals
The answer, I think, is that this suggestion does have some appeal when applied to our interest in remaining alive, but that it becomes far less plausible when we apply it to the other interests that we share with animals. On the one hand, our interest in remaining alive is at least partly a reflection of our interest in retaining whatever good things our lives contain. Thus, if the greater complexity of human thought increases the number or quality of the good things in our lives, then the resulting intensification of our interest in not dying may indeed explain why it is more wrong to kill humans than animals. Yet even if it does, the impact of our greater mental powers on the other interests that we share with animals will remain less clear; for a human being’s greater powers can either intensify or mitigate a setback to one of these interests. A person’s pain can be made either harder to bear by his awareness that there is no end in sight or easier to bear by his understanding of the need for the medical procedure that has caused it, while a criminal’s confinement can be made either worse by his resentment or self-pity or less bad by his constructive attitude. Because it is hard to know which effect is more pronounced, it is also hard to know whether the overall impact of our greater mental capacities is to strengthen or to weaken the duties that protect these interests. Moreover, there is also a deeper problem with the suggestion that the greater mental capacities that support such higher-order reactions as resentment and humiliation can thereby magnify setbacks to our interests; namely, that even when some such capacity does make a setback harder to bear, what is generally worsened is not the setback itself, but only the subjective quality of our attitude towards it. A person who burns with resentment at being imprisoned may be more unhappy, or be unhappy in a different way, than an animal in a cage of the same size, but the animal is no less unfree; and the same holds, mutatis mutandis, for a human and an animal each of whom is suffering debilitating pain or has been separated from her offspring. In each case, the human’s greater mental capacity affects only the way he experiences the setback to his interests, but not the magnitude of the setback itself. Thus, at least where interests like these are concerned, the difference between the mental powers of humans and animals seems unlikely to show up as a difference in the strength of the duties that protect the interests. But are these the only interests that might have a differential effect on our duties? Don’t human beings also have an interest in avoiding the unhappiness that their greater mental powers make possible? And mightn’t it be their ability to suffer setbacks to this interest that supports the claim that we owe more to them than we do to animals? With this suggestion, we have moved from a case in which the interests of humans and animals are the same to a case in which they are not. Such
George Sher
295
differences are very common, and many of them are linked to our capacity for language. It has been argued, for example, that because a pig cannot learn to read, we have no duty to educate it.¹⁰ Taking our cue from this, we may be tempted to conclude that even if animals are our moral equals, the duties that we owe to them must be far more limited, and thus must be far easier to discharge, than our duties to our fellow humans. However, for at least two reasons, this inference would again be problematic. The first and less important problem is that the inference will only work if animals have fewer overall interests than humans, and it is far from clear that this is the case. Just as humans appear to have interests that animals lack, so too do many animals appear to have interests that humans lack. For example, Christine Korsgaard, quoting Temple Grandin, has noted that pigs appear to have an unending appetite for rolling around in straw, an activity that holds only limited appeal for most humans.¹¹ When this and the many other interests that are peculiar to animals are factored in, they may turn out to have just as many distinct interests as humans do. The second and more important problem with the inference is that even if animals do have fewer interests than humans, that difference will be of little significance if our duty to protect whichever interest ranks highest in an animal’s hierarchy is just as stringent as our duty to protect whichever interest ranks highest in the human hierarchy, and if the same is true of our duties to protect the interests that rank the second highest, the third highest, and so on, within each hierarchy. Because this formula does not prioritize the interests of any particular species, it looks like a good fit with the idea that humans and animals have equal moral standing. However, if we accept the formula, then we will have little reason to suppose that our duties to animals are significantly less demanding than our duties to one another. This, therefore, is yet another reason to resist the inference from the premise that humans have greater mental powers than animals to the conclusion that we owe them more.
4. Treating Moral Equals Differently But neither can we infer, from the premise that animals are our moral equals, that we always owe them the same treatment that we owe to humans. If we did accept that inference, then we would also have to agree that because all humans are moral equals, we are morally obligated to treat each of them ¹⁰ Singer (1975: 6). ¹¹ Korsgaard (2018: 68–69).
296
First Among Equals
equally. This would require us to deny (among other things) that when we promise to do something for someone, we acquire an obligation to him that we do not have to others; that we may punish the guilty but not the innocent; that we are obligated to help the destitute but not the well-off; that we should reward the deserving but not the undeserving; and that we owe many things to our friends and loved ones that we do not owe to strangers. Because we all take the moral equality of persons to be consistent with at least some such disparities in our duties towards them, we are hardly in a position to insist that the moral equality of humans and animals means that our duties to them are always the same. Moreover, by getting clear about why the moral equality of humans does not require that they all be treated equally, we will also gain a new perspective on our duties towards animals. As I noted earlier, I accept (what I take to be) the standard view that a being’s moral standing is measured by the moral importance of its interests. This view clearly implies that if two people have equal standing, then their interests should be given equal weight in our moral deliberations. But at which level of deliberation does this dictum apply? Because morality is action-guiding, the most straightforward answer is ‘at the level at which we try to decide what to do’. But because moral decision-making involves appeals to principles, we cannot make decisions at this level without presupposing some such principles, and it is to our logically prior deliberations about these principles that the moral equality of persons is usually thought to apply. When philosophers assert that each person’s interests count equally, what they generally have in mind is that each person’s interests should be given equal weight in whatever reasoning we use to arrive at our moral principles. Philosophers notoriously disagree both about the form that reasoning should take and, albeit to a lesser extent, about the principles that should emerge from it; but for present purposes, none of that matters. For us, the crucial point is simply that each competing line of reasoning begins by asserting that persons are moral equals but ends by endorsing principles which allow or require us to treat them differently. So, for example, when an act-consequentialist (somehow) moves from the premise that each person’s interests count equally to the conclusion that we ought to maximize the satisfaction of people’s interests (or that we ought to maximize some quantity that is closely related to those interests), he thereby endorses a principle that requires us to benefit person A but not person B whenever our doing so would bring the greatest increase in the relevant total. When a rule-consequentialist moves from the same premise to the conclusion that we ought to follow the rules whose general adoption will maximize the satisfaction of people’s
George Sher
297
interests, he similarly implies that we should benefit A but not B whenever that is what the rules call for. And while a contractarian will take the fact that each person’s interests count equally to require not that we maximize anything, but rather that each person be given a veto over the principles that are to govern our interactions, the principles that emerge—the ones that no one can reasonably reject—will again include many that allow us to do things that burden some people more than others. It is harder to reconstruct the role of other people’s interests in Kant’s more complicated account, but however we do it, we can hardly deny either that he takes our practical reason to commit us to acting in ways that treat others’ interests as no less important than our own or that some resulting duties, such as the imperfect duty to render aid, will allow us to treat different people differently, while others, such as the perfect duty to keep our promises, will positively require this. By thus relocating the point at which a person’s interests enter our moral deliberations—by treating his moral standing simply as what qualifies his interests for inclusion among the ones that our moral principles seek to reconcile—we make it possible to analyse moral justification in a way that explains why moral equals need not be treated equally. As so analysed, each complete justification of an action has three distinct stages, the first an argument from certain facts about persons to the conclusion that their interests are of equal importance, the second an argument from the equal importance of their interests to a set of principles to govern our behaviour when those interests conflict, and the third an argument from some such principle, together with certain factual premises about a given individual, to the conclusion that a certain way of treating him is called for. Although each such justification appeals to facts about persons at both its first and its third stage, the reason we remain able to say both that all persons are moral equals and that we may often treat persons unequally is that the facts about them that enter at the different stages are very different. For, on the one hand, to ground the claim that all persons have moral standing, and thus also the stronger claim that they all have equal standing, we must to appeal to one or another feature that they share—on my account, their each being a separate centre of subjectivity, on others, their each having some scalar property that surpasses some relevant threshold. By contrast, to justify a claim about how we are obligated or permitted to treat any particular individual in any given situation, we must appeal to whichever facts about him or his situation are relevant to whichever principles we accept. We must appeal to facts about his crimes to justify punishing him, facts about his virtue or effort or achievement to justify rewarding him, facts about his level of wealth or wellbeing to justify aiding him, facts about our promises
298
First Among Equals
to him or his relation to us to justify favouring him over others, and so on. Because people differ greatly in each of these dimensions, there is no reason to expect that the ways in which our moral principles allow or require us to treat them will always be the same. And, because of this, there is simply no conflict between the claim that all persons are moral equals and the claim that we need not treat them all equally.
5. Treating Humans and Animals Differently And neither, similarly, is there any conflict between the claim that the members of certain animal species are the moral equals of humans and the claim that we may treat those animals differently from humans; for here again, the facts that establish that the animals are our moral equals will enter only at the stage at which we identify the beings whose interests our moral principles must reconcile, while the facts that determine how we may or must treat those beings will enter at the later stage at which we draw on our principles to decide what to do. Just as any schema for the reconciliation of the equally important interests of different human beings must allow our treatment of them to vary with the facts about them and their circumstances, so too must any schema for the reconciliation of the equally important interests of the members of different species. With these observations, we arrive at an answer to our original question. To explain how we can both ground the moral equality of persons in the fact that each occupies a separate subjectivity and at the same time justify treating persons differently from the innumerable cows, grackles, and garter snakes which presumably also have separate subjectivities, we need only hold fast to the idea that our moral principles regularly and predictably allow our treatment of moral equals to vary in accordance with the ways in which they differ.¹² However, if we were to leave things here, then our answer would tell us nothing either about which factual differences between humans and animals might justify us in treating them differently or about which forms of different treatment those factual differences might license. Because these are the things we really want to know, it would be unsatisfying to leave these ¹² In her contribution to this volume, Teresa Bejan distinguishes between the ancient Greek and Roman ideas of equality, which entails commensurability, and parity, which does not. Of these two notions, the second, parity, is more flexible, in that two beings can be on a par—can be peers—even though they may in some contexts be treated differently. In these terms, my account treats humans and animals as equals at the initial stage of moral thought at which it assigns their interests equal weight, but merely as on a par at the later stage at which it views them as inhabiting as single moral universe within which humans are nevertheless favoured.
George Sher
299
questions entirely unaddressed. Thus, to end, I will take a few hesitant steps into this vast problem area. The first thing to note here is that there is a significant discontinuity between the human-human and the human-animal cases. When our moral principles allow or require us to treat different people differently, the reason is often that those people have not all acted in the same way. So, for example, to explain why B but not A should be punished, we cite the crime that only B has committed; to explain why A but not B should get the promotion, we appeal to A’s steadier work history; and to explain why it is not wrong for A to have more money than B, we point out that A has invested while B chose to consume. In each such case, our explanation presupposes that the parties are in some sense responsible for what they did. But cows and grackles and garter snakes are never responsible for what they do, so it cannot be their actions that justify us in treating them differently from humans. Thus, if we are to explain why we owe them less, it will have to be on the basis of some further differentiating features. But which other features might these be? The natural answer, I think, is that they are likely to be precisely the features that are commonly taken to set animals apart—their apparent inability to conceive of themselves as selves, for example, or their associated inabilities to project themselves into the future, make plans, think in abstract terms, or step back from their impulses and ask what they really have reason to do. Because the way we take these intellectual limitations to affect our duties towards animals will depend on which moral principles we accept, and because that in turn will depend on the reasoning we use to arrive at our principles, there is ample room for disagreement about the content of the relevant duties. Thus, instead of trying to reach any substantive conclusions, I will simply examine a few of the ways in which the intellectual differences between humans and animals might affect either the reasoning through which we arrive at our moral principles or the further reasoning through which we invoke those principles to identify our duties in particular cases. Consider first Peter Singer’s observation that because human beings have greater intellectual capacities than animals, they also have a broader range of interests, and so have more both to gain and to lose. As Singer deploys this observation, it combines with his act-consequentialism to yield the conclusion that we are often justified in favouring humans over animals. So, for example, when we must choose between saving (or taking) the life of a human or a nonhuman, it is often preferable to sacrifice the nonhuman because the life it stands to lose is less rich in the sorts of goods that require awareness and reflection. Along similar lines, an act-consequentialist might argue that we should generally seek to benefit a human in preference to an animal because
300
First Among Equals
the human’s greater mental abilities will generally enable him to gain more from our efforts. When act-consequentialists say things like this, they’re not going back on their claim that the interests of animals matter as much as those of humans because they would just as readily urge us to favour animals over humans if our doing that would yield the greatest bang for the consequentialist buck. In either case, the facts that these consequentialists take to decide the issue will enter their thinking not at the early stage at which they arrive at their maximizing principle, but only at the later stage at which they apply that principle to the decisions at hand. But when other theoretical approaches take the greater mental powers of humans to justify us in favouring them in certain ways, they may invoke those powers at a different and earlier stage of their reasoning. Consider, for example, Thomas Scanlon’s contractualism, which permits us to do whatever is allowed by at least one system of rules that no one could reasonably reject.¹³ Although Scanlon argues that even systems of rules that maximize utility can sometimes be reasonably rejected by those who are unduly burdened by them, he acknowledges that the case for reasonable rejection diminishes as those burdens become more tolerable.¹⁴ This means that even if animals or their proxies are among the contracting parties, as Scanlon thinks they are,¹⁵ the mental differences between animals and humans that make a certain mode of treatment less tolerable to humans than it is to animals may make it reasonable for humans but not animals to reject any principle that subjects them to it. This may be true, for example, if the ability of humans to pursue temporally extended projects gives them a significant additional interest in remaining alive, or if the human capacities for resentment and indignation greatly amplify the harms that humans suffer when imprisoned. If these claims are true, and if the differences they mark are great enough, then there may be principles which allow us to sacrifice animal lives for human lives, or to cage animals but not humans, to which no member of either group can reasonably object. Because Scanlonian contractualism sometimes forbids us to perform the acts that would have the best consequences, the ways in which it allows us to favour humans over animals are not likely to coincide with the ones that are allowed by act-consequentialism. However, even if the permitted ways of favouring humans did entirely coincide, the greater mental powers of humans would still play a very different role in the reasoning that justifies them; for ¹³ The canonical statement of Scanlon’s view appears in his 1998 volume. For an influential earlier statement, see his (1982). ¹⁴ Scanlon (1982: 123). ¹⁵ See Scanlon (1998: 177–187).
George Sher
301
whereas the act-consequentialist does not invoke the additional interests to which our greater mental powers give rise until he brings his maximizing principle to bear on particular cases, the Scanlonian contractualist will invoke those additional interests at the earlier stage at which he arrives at his moral principles themselves. And while it would be tedious to repeat the argument, I think it is clear that what holds for contractualism will also hold for ruleconsequentialism. Like the contractualist but unlike the act-consequentialist, the rule-consequentialist who views animals and humans as moral equals will consider whatever additional interests humans have in virtue of their greater mental powers primarily at the stage at which he decides on his principles. He will appeal to those additional interests not when he is deciding what the rules whose general acceptance would bring the best consequences tell us to do in particular cases, but rather when he is trying to decide which rules would bring the best consequences if they were generally accepted. The fourth and final theoretical approach that I want to consider is Christine Korsgaard’s unorthodox Kantianism.¹⁶ Because Kant himself believed that animals lack the forms of reason that account for the moral standing of humans, his official reason for opposing cruelty to animals was simply that it degrades the character of human beings. Thus, for Kant himself, the problem of squaring the moral equality of humans and animals with the apparent differences in our duties to them simply did not arise. By contrast, although Korsgaard expresses some general doubts about the concept of moral standing, she endorses an interpretation of Kant’s theory which in effect implies that all sentient creatures do have it. Although she stops short of maintaining that animals and humans have the same moral standing, she comes close enough to saying this to justify an examination of the role that the greater mental powers of humans play in her reasoning. Kant argues that when we exercise our rational agency in pursuit of an end, we assume both that that end is worth seeking and that we who seek it have a different kind of worth—that we are ends in ourselves. He also argues that practical reason requires its possessors to act only on principles that they can will to be laws for all other rational agents. Because animals are not rational agents, Kant holds that these arguments do not apply to them, but Korsgaard thinks he is only partly right about this. As she sees it, Kant is right to say that our rationality requires us to act only on principles that we can will as laws for other rational agents, but is wrong to say that a principle can pass this test without assigning worth to non-rational beings and their ends. Because ¹⁶ Korsgaard (2018: part II), and passim.
302
First Among Equals
any animal that pursues an end is also assuming that what it seeks is good in some absolute sense, our rationality requires us to treat its ends as no less important than those of any rational creature. By prising apart these aspects of the Kantian position, Korsgaard opens up a new way of understanding how the greater mental powers of humans can affect the scope of our duties to animals. On the one hand, because animals are not autonomous agents—because they lack the capacity to evaluate principles with an eye to deciding which ones to adopt as laws—Korsgaard concedes that we are under no obligation to respect their autonomy. Although we are forbidden to interact with humans in ways that pre-empt or bypass their own judgement, we have no comparable duties towards animals. However, on the other hand, because both humans and animals view their own ends as absolutely good, we are obligated to act only on principles which take their ends as seriously as our own. As Korsgaard understands it, this last obligation requires us to rethink our treatment of animals in areas ranging from the management of predatory species to medical research and the keeping of pets. Korsgaard’s treatment of these issues is characteristically deep and resourceful, and I obviously cannot go into the details here. My point in mentioning her approach is simply to highlight the structural differences between it and the others we have considered. Like the proponents of the other approaches, Korsgaard agrees that the moral standing that we share with animals does not mean that we must always treat the members of the two groups equally, and like them also, she traces the differences in our duties to the greater mental powers of humans. However, unlike the other theorists, she appeals to our greater mental powers not to explain how certain modes of treatment affect humans differently from animals (act-consequentialism), nor yet to show that such differences in impact are relevant to the rules that we ought to adopt (contractualism, rule-consequentialism), but rather to distinguish one aspect, or ‘moment’, of the practical reason from which all of our duties flow that does affect what we owe to animals from another that does not. Of the different approaches just mentioned, each has much to recommend it, each makes different use of the empirical differences between us and the animals, and each can be expected to yield different conclusions about our obligations to animals. Thus, even with a clear understanding of how the fact that animals are our moral equals can be consistent with our having different obligations towards them, our views about the content of those obligations will depend not only on the factual information that is available to us, but also on our answers to some of the most fundamental questions of moral theory.
George Sher
303
Here as elsewhere, the cost of solving one problem is to open up a new and daunting array of others.¹⁷
References Arneson, Richard. 2015. ‘Basic Equality: Neither Acceptable nor Rejectable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 30–52. Oxford: Oxford University Press. Kagan, Shell. 2019. How to Count Animals, More or Less. New York: Oxford University Press, 2019. Korsgaard, Christine. 2018. Fellow Creatures: Our Obligations to the Other Animals. New York: Oxford University Press. Rawls, John. 1999. A Theory of Justice, revised Edition. Cambridge, MA.: Harvard University Press. Scanlon, Thomas. 1982. ‘Contractualism and Utilitarianism’. In Utilitarianism and Beyond, edited by Amartya Sen and Bernard Williams, 103–128. Cambridge: Cambridge University Press. Scanlon, Thomas. 1998. What We Owe to Each Other. Cambridge, MA: Harvard University Press. Sher, George. 2014. Equality for Inegalitarians. Cambridge: Cambridge University Press. Sher, George. 2017. Me, You, Us: Essays. New York: Oxford University Press. Singer, Peter. 1975. Animal Liberation. New York: Avon Books. Steinhoff, Uwe. 2015. Do All Persons Have Equal Moral Worth? New York: Oxford University Press. Waldron, Jeremy. 2017. One Another’s Equals. Cambridge, Mass.: Harvard University Press.
¹⁷ Emily Fox Gordon and Vida Yao have offered helpful comments on this chapter; thanks to both of them. Thanks too to Teresa Bejan, Andrea Sangiovanni, and the editors of this volume, all of whom have offered constructive suggestions.
13 Basic Equality, Rational Agency Capacity, and Potentiality Richard Arneson
The idea of basic equality is elusive and not transparently clear. The rough thought is that all human persons share a fundamental equal moral status, the content of which depends on what are deemed to be fundamental moral principles applying to all persons. But the idea is not purely formal, to the effect that whatever fundamental moral principles there are apply to all persons equally. There is a substantial component to the basic equality ideal. One can interpret the basic equality idea as contrasting ancients and moderns and repudiating ancient doctrines that are accepting of naked hierarchies of birth and military power. Jeremy Waldron (2017) associates the idea of basic individual human dignity, a version of the basic equality idea, with taking the claim of special worth and status ascribed to high-caste individuals, kings, and nobles, and asserting that (some rendering of ) this special high status properly attaches to each and every human person regardless of actual conventional social rank and standing. This opposition to caste and class is an opposition to naked hierarchy, as contrasted with functional hierarchy, possibly justified by its operating to advance the common good. So understood, the idea of basic equality opposes egoism as well as doctrines that say the nobles and elites are genuinely nobler than the rest of us, superior not just by virtue of possession of some valuable traits that some others lack, but in some all-things-considered nobility that merits special adulation, deference, non-reciprocal sacrifice, and the like. The doctrine of egoism says that each individual always has decisive reason to do whatever is most to her own advantage, at any cost to others. Basic equality says that each person has decisive reason to do whatever would be giving due consideration to the interests of all other persons, each person’s interests being equally as valuable as the identical interest of any other. Due consideration might allow some partiality towards those near and dear to us, including ourselves, but such partiality must be justified by impartial principles binding on all of us. Richard Arneson, Basic Equality, Rational Agency Capacity, and Potentiality. In: How Can We Be Equals?. Edited by: Giacomo Floris and Nikolas Kirby, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192871480.003.0014
306
Basic Equality, Rational Agency Capacity, and Potentiality
A substantial basic equality doctrine requires a satisfactory grounding basis. What justifies the idea that all persons share a fundamentally equal moral status? A natural thought is that basic equality is a status that human persons have, by virtue of some traits they possess. You have basic equality, and your perhaps beloved cat does not have it. The cat lacks the status by virtue of lacking the traits. Basic equality thus separates most human individuals from other animals with which we have come in contact. These may have traits that make them morally considerable, owed something by virtue of possessing animal sentience or other capacities. But being morally considerable does not bring you to basic equality status. For that, some further eligibility condition must be met. We classify as persons those beings that share the trait, possession of which qualifies a being as having basic equality status. As is by now well known, interpreting the basic equality idea in this naïve way immediately introduces formidable questions that so far look as though they do not admit of satisfactory answers. We are off on what G. A. Cohen once called the ‘wild goose chase for defining characteristics’ (Cohen 2013: 194). That is, we are looking for some trait X (or cluster of traits) such that all persons (all those we are sure should be classified as persons) equally possess X, where the traits render the bearer respectable or estimable or worthy of special recognition, and since we all have equal X, we are all owed equal basic respect, estimation, recognition. Cohen, along with many others, senses that this search for defining characteristics is futile, since any remotely plausible candidates for the role of trait X, such as intelligence, capacity for language use, the capacity to resist acting on an immediate felt desire or subjective impulse to action (and instead consider whether there is reason to put one’s will behind acting on the desire), the capacity for reciprocity or for treating others with moral consideration and a concern for fair dealing, the possession of a subjective point of view on the world, and so on, will all vary by degree. For example, if possession of subjectivity is singled out as the candidate trait that suffices for basic equality status, it turns out that some persons have more of it than others, or a richer, more complex subjective outlook (but cf. Sher 2015). They differ not only in developed subjectivity, but also in capacity for that. In the case of this particular candidate basis for equal personhood, there is the further difficulty that other animals that intuitively we judge to be appropriately classified as nonpersons exhibit capacity for subjectivity and developed subjectivity (but see Sher, this volume). More generally, although one can define a binary trait term, so one either possesses the trait or not, underlying any such trait will be a scalar trait or
Richard Arneson
307
traits, which better capture our valuations. ‘Has the capacity to play baseball or not’ is binary, but ‘has capacity to play baseball skilfully’ is scalar. For any candidate trait X, there is no plausible line of sufficiency, such that whether or not the person possesses a level of X beyond the sufficient level, and to what extent, is of little or no concern (Arneson 1999; 2015). Just as important, it seems a hard fact that people do differ in their amounts of any plausible candidate defining characteristic X, above and well above any line that anyone might deem as the line that marks sufficiency. If an individual has some combination of X traits that passes the line of sufficiency, and on this basis has equal basic moral status, why does not the fact that above this line, the individual has more X compared to others, entitle the person to greater basic moral status? There are by now familiar attempted escape routes from this impasse. I can’t here examine them seriously, but I will simply register my belief that they fail (but cf. Carter 2011, and for discussion, Arneson 2015). One possibility is partial comparability. It is plausible that trait X is multidimensional, and the different dimensions are qualitatively diverse. It is not clear that there is a fact of the matter, given Tom has this array of developed trait X relevant abilities, and Sally has this different array, which of them has more of developed trait X abilities all things considered. The trouble with this gambit is that unless there are limits to partial commensurability, it threatens our basis for holding that normal adult human persons have trait X and other animals known to us do not. My cat probably has a more keenly developed subjective appreciation of mice as prey than I do, but this does not dent my conviction that taking all the relevant defining characteristics, all things considered, I qualify as a person and my cat does not. Another possible response to the impasse is to hold that it is not that people really are fundamentally moral equals in some basic sense, but instead that we ought to regard each other and act towards each other as though that were so, so far as we can. In defence of an ‘as if ’ equality norm, we might point to several considerations. A plausible standard for assessing people as fundamentally equal is bound to be multidimensional. How to measure scores along the different dimensions—and even more, how to weight the dimensions to arrive at all things-considered moral status scores—is bound to be uncertain and contentious. Any attempt to classify individuals on such a basis would give rise to squabbling and impede social cooperation. Better to eschew this grading and sorting enterprise altogether. However, at the end of the day, it is counterintuitive to hold that while we don’t really believe that people are fundamental equals entitled to equal basic moral rights, it is pragmatically convenient to pretend that this is so, for
308
Basic Equality, Rational Agency Capacity, and Potentiality
the sake of minimal social harmony. We seek to find a justified secular nonmisogynist construal of proclamations of equal moral status shared by all humans such as that found at the ‘all men are created equal’ start of the United States Declaration of Independence. According to the desired paraphrase, the proclamations will qualify as true, not merely useful. Another possibility is that we regard and treat people as equals not primarily because we recognize that they possess traits that call for recognition of them as equals. Rather, we seek to establish and sustain relating as equals. But seeking to relate as equals to ordinary humans is apt, and to ordinary dogs and cats, horses, and even gorillas, inapt. It’s hard to avoid the thought that we should regard and treat as equals because we are, in some important way, equals. What way is that? A more blanket dismissal of the basic-equality-and-its-basis problem is available if fundamental moral principles regarding what we owe to all sentient beings do not require for their application any notion of a person. The most prominent view of this sort is utilitarianism, which says what anyone morally always ought to do is whatever would maximize the total sum of happiness (or more broadly, good attained). So long as one can know, for any actions one might perform, which will bring about the greatest sum of good, one does not need to know how that good is apportioned across persons and other sentient beings, or even whether or not there is a coherent notion of person. In this spirit Peter Singer (2011) affirms that all animals (sentient beings) are equal. The discussion that follows assumes that this feature of utilitarianism renders it unacceptable, but no backing for this assumption is offered here, and the reader may find it needs justification. There is available a simple, natural, and popular, albeit puzzling, view that at present strikes me as most defensible, or better, least indefensible. This is the view that there is a threshold level of possession of certain traits, such that if you attain that threshold, and your possession of the traits falls within a certain wide range, you qualify for basic equality status. You are a person among others, all equally persons, despite the uncontroversial further fact that above the threshold, at least within a certain range, all of us basically equal persons are unequal in possession of the very traits possessing enough of which makes us basically equal. The traits in question are rational agency capacities (RAC)—cognitive, affective, and volitional—that together enable one to discern and assess reasons for belief and for choice of action and to be moved by recognition of the overall balance of reasons to decide what to believe and do and to resolve to act on one’s choices so far as one is able to do so. In terms familiar from discussions of what can make one responsible and accountable, having
Richard Arneson
309
reasons-recognition capacities and reasons-responsive capacities at a threshold level or above makes one a responsible agent. Following discussions on related topics by John Rawls (1999: 442) and Robert Nozick (1974: 49) in the 1970s (and earlier discussions throughout the history of philosophy), it is important to stress that the capacity to recognize reasons that is fully agencyconstituting includes a capacity to recognize moral reasons, those having to do with what any agent owes to others by way of consideration and concern, and also affective capacity to be moved to desire and choose and act by the reasons one discerns. A cognitively competent extra-terrestrial predator might recognize the reasons there are including moral reasons, but have no capacity to respond empathetically to the plight of any agents, or any agents except oneself and perhaps some few others with whom one happens to identify. The smart extra-terrestrial so characterized would not be a full person, on the familiar view I am rehearsing. In this spirit Rawls suggests that it is ‘moral persons’, having a capacity for moral personality, who ‘are entitled to equal justice’ (Rawls 1999: 442). Rawls explicitly observes that the capacities that constitute moral personality can vary by degree from one person to another, and just reaffirms that these variations do not affect the judgement that all individuals who possess moral personality at or above the threshold level and within a wide range are all equally owed justice (the moral rights that are the essentials of what we owe to one another) and equally owe justice to one another. Rawls can hardly be oblivious of the fact that there is an anomaly in the position he is staking out. The justification for adhering to the simple RAC account despite the anomaly appeals to our considered judgements in an epistemic state that is as close as we can get to reflective equilibrium in the current state of the discussion. After considering as best we can what moral beliefs to accept, we find that our conviction that all beings possessing moral personality capacities at or above a certain threshold are all (1) equally entitled to the same moral rights and (2) equally required to respect those moral rights in their conduct towards each other is stronger than any convictions that conflict with this claim. The main conviction in conflict with basic equality is that if greater overall possession of RAC below the threshold makes one morally more considerable, it is arbitrary to deny that the same holds for greater overall possession of RAC above the threshold. Lacking a satisfactory explanation for why this is not so, we are in a state of perplexity. But this epistemic unease is not a good reason to reject basic equality. Our moral progress is limited and halting. In almost any area of morality, views in which we are confident are shadowed by puzzles and discrepancies we do not know how to resolve. This messy state of affairs is never in itself good reason to scrap the
310
Basic Equality, Rational Agency Capacity, and Potentiality
beliefs in which after reflection we repose most confidence for others less well supported. Any position one might take on basic equality, including its flat denial, will be problematic or counterintuitive in some respect. The most promising and likely correct doctrine is the one that is least problematic all things considered. The remainder of this chapter explores the RAC account of basic equality and the moral status of personhood and offers a modest and preliminary defence of it. Section 1 clarifies the RAC account. Section 2 locates its plausible source of appeal in the fact that it provides a simple unified explanation, or the key building block for such an explanation, for several moral convictions that most of us share. Section 3 sketches a controversial apparent implication that may yet be defensible and acceptable. Section 4 raises a seemingly devastating difficulty that feminist writers on the moral permissibility of abortion (the first being Warren 1973; 1984) have also flagged as needing attention. Section 5 proposes a modification of the account, to the effect that potential for threshold RAC in combination with actually developed RAC determine a being’s moral considerability up to basic equality. The proposal eases the seemingly devastating difficulty. Section 6 briefly discusses other views.
1. The RAC Account To reiterate: the RAC account of what confers basic equality status on a being says that possession of developed rational agency capacities (cognitive, affective, and volitional) at or above a threshold level, within a certain range, renders one a person entitled to the same consideration as any other person. One has the same fundamental rights as all other persons; one’s interests count the same as those of any other person’s in determining what ought to be done. One question that immediately arises is where the threshold that marks the lower boundary of personhood is located. Another is how the three kinds of rational agency capacity—cognitive, affective, and volitional—are weighted against each other and combined to yield a being’s overall level of RAC capacity. To the latter question, we might hold that to be a person, one must have each of cognitive capacities, volitional capacities, and affective capacities at a threshold level, not just a combined score of all three that is sufficiently high to meet a combination threshold. For example, if there were giant sloths that were found to have enormous developed affective capacity but hardly any cognitive capacity or volitional capacity, we would not, I think, now
Richard Arneson
311
judge that the enormously affectionate sloths are persons, full members of the moral community. As sketched so far, the claim is that those beings above the threshold of RAC capacity and within a wide range all share full personhood and basic equality status, but there could be beings that have far greater RAC capacity than the human range and have a higher status and special moral prerogatives. Since we humans lack any acquaintance of any such beings, apart from in science fiction, this issue does not force itself on our attention as a practical matter.
2. RACʼs Appeal The appeal of RAC is that it organizes and explains a huge number of judgements we intuitively want to make about the moral status of a wide range of beings. First, the RAC account helps us understand what makes humans, so far as we know, special and unique (at least in our small corner of the universe). So far as we are now aware, only humans become persons. RAC also has a simple explanation of why nonhuman animals are morally considerable. A morally considerable being has interests that impose on us some duty to help bring about their fulfilment. The explanation is that nonhuman animals possess some RAC, although below the threshold level of personhood. Moreover, the RAC account implies a plausible view of the variable moral status of various nonhuman animals—the greater your developed rational agency capacities, the higher your moral status, up to the personhood threshold. Normal whales and gorillas, being smarter, count for more than cows. Elephants have affective capacities that spiders seem to lack. Killing spiders for fun might be morally acceptable even if not highly virtuous; killing elephants for fun is vicious and immoral. The fact that RAC is guiding judgement manifests itself when our judgement is tentative and uncertain. If there is a lot we do not know about the cognitive abilities and the mental life of whales, there is a lot we do not know about the moral status and moral considerability of these animals. The RAC judgement is made case by case on an individual basis, not broadly by types of species membership. This seems right. A superchimp who is genetically freakish and turns out to have normal human rational agency capacities would qualify as a person despite his belonging to a species whose members generally lack rational agency capacity at that level. In a similar way, as we read fictional accounts of a spider named Charlotte who gives
312
Basic Equality, Rational Agency Capacity, and Potentiality
sage advice to pig acquaintances and has qualities you would be glad to find in a friend, we have no difficulty regarding this individual spider as a person with the dignity that personhood status confers. RAC also avoids speciesism. Speciesism is a form of arbitrary moral discrimination, on par with racism. It is wrong to treat persons differentially simply on the basis that some are members of a particular race or ethnicity or on account of having one or another skin colour. Mere biological species membership does not in itself determine what we owe to different individual beings. An extra-terrestrial, possessing each of cognitive, affective, and volitional abilities at the threshold level, qualifies as a person. Not being human is no disqualification from personhood status. The moral treatment of humans and other animals is regulated by impartial moral principles that pay no heed to species membership as such. The speciesism worry: consider this trio of claims: (1) Each individual human is morally considerable, and has the specially high status of full personhood, independently of the quality of any of the traits possessed by that very individual. (2) Whether an individual animal is morally considerable, and to what degree, depends entirely on the quality of some traits possessed by that very individual. (3) Humans are animals (specifically, mammals). Since 3 is clearly true, and 1–3 are together inconsistent, either 1 or 2 must be rejected, and the better candidate for rejection is 1. Rejecting speciesism says nothing about what level of moral consideration is due any particular animal; it simply says our treatment of all animals must be even-handed, principle-based, and not morally arbitrary (for contrary views, see, for example, Nozick 1997; Nussbaum 2006; Waldron 2017; and for discussion, Jaworska and Tannenbaum 2021)
3. A Controversial but Defensible Implication Besides having some unassailable, or at least highly plausible implications, the RAC account of moral considerability and of personhood status also has implications that are more contestable. On this view, personhood is a stage in the life of a human individual. If one is lucky, one starts life as a nonperson human foetus, at some point in time acquires sufficient developed rational agency capacities and becomes
Richard Arneson
313
a person, and remains a person until one’s death at a ripe old age. Another possibility is that injury or disease inflicts losses of RAC on one, so that one loses nonpersonhood status for a time, but without losing all possibility of restoring RAC capacity. A yet more unfortunate scenario involves permanent loss of RAC capacities, as occurs, for example, when someone is afflicted with Alzheimer’s disease and gradually becomes severely mentally impaired, to the point that while remaining morally considerable, a being to whom we have some duties, one drops below the RAC threshold and ceases to be a person as here defined. Various forms of mental illness involving combined losses of cognitive, affective, and volitional capacities can reduce one to a similar plight. What by way of moral consideration is owed to the severely mentally impaired former person is a vexed moral question. Even the bare-bones characterization given so far offends against many thoughtful people’s views as to the proper respect and consideration the severely mentally impaired deserve. In the laws of most countries and in a common-sense morality that aligns with the law, someone uncontroversially mentally disabled below personhood status as conceived on the RAC account nonetheless remains a person with the basic rights of persons (so far as it is possible to accord them to the individual). If I develop Alzheimer’s that has progressed to a severe stage, my wife is moved by pity for my incapacitation, confused distress, and the grim future that awaits me, and painlessly kills me, this is legally a murder, and in many people’s eyes murder from a proper moral standpoint as well. Here the plain implications of the RAC account are jarringly out of step with most people’s views, but, I would submit, plausible nonetheless in their broad outline. Ex hypothesi I lack the mental capacities of our family cat, and no one would think to condemn my wife for taking a very aged, infirm, and ailing cat to the veterinarian to be painlessly killed. Following standard reflective equilibrium methods for arriving at plausible moral views, we seek a view that coheres with ordinary common-sense judgements that withstand reflection, along with the revised judgements that emerge and attract considered allegiance as less plausible views are discarded. The view that at the present time, as best we can determine, is overall most coherent and plausible might turn out to be jarringly out of step with most people’s views on some topic. If our ordinary convictions are latently inconsistent, this will have to be true for some entrenched judgements. Of course one possibility is to discard the basic equality claim altogether. This chapter proceeds on the assumption that doing that would push one out of the frying pan and into the fire of greater epistemic implausibility.
314
Basic Equality, Rational Agency Capacity, and Potentiality
Of course, the killing of me described above fits well within the category of a mercy killing, one that is to the benefit of the individual who is killed, so does not pose in any vivid way the controversial implications of the RAC view of basic equality applied to the situation of the severely mentally impaired. If the badness of death is that it cuts off a future of life worth living (such that a rational person would be glad not sad to remain alive, either for the sake of the prudential goods that would accrue from longer life or agency goods deriving from doing good for others), a mercy killing does not cut off an all-thingsconsidered life worth living for the person who dies, but rather is a benefit to that person, by cutting off future life that would be worse than no further life at all. A mercy killing so construed might take place against the informed will of the person who is killed, and arguably be morally objectionable on that basis. But the individual suffering from advanced Alzheimer’s is incapable of forming an informed will. Hence, the consent issue will presumably not arise in the decision problem I am imagining my wife facing, unless I had back when I was still mentally competent set my mind against being killed in the circumstances that now obtain. Another possibility is that once there is no possibility of competent will formation, due to lack of mental capacity, an incompetent will set against being killed suffices to erect a personal autonomy moral bar against being killed (but for discussion, elaborating further possibilities, see McKerlie 2013; McMahan 2002). The RAC account on the face of it tends to support more controversial verdicts. Common-sense moral judgements affirmed by many would support killing some animals (cats, for example) just for convenience. If with sufficiently advanced Alzheimer’s I have mental and decisional capacities roughly comparable to a cat, or less than that, then killing me painlessly just for convenience, not as a mercy killing, would seem to be rightly judged permissible on the same footing. Beings who are roughly on a par in their RAC capacities should be entitled to roughly the same level of protection by deontic constraints. This way of thinking is surely counterposed to common-sense moral views. The RAC account, as Jeff McMahan (2002; 2007; 2008) reminds us, does not by itself determine whether killing for the convenience of the killer is morally permissible or forbidden, but it insists on treating like cases alike, with similarity fixed by individuals’ degrees of possession of RAC (or, possibly, further considerations beyond RAC). Common-sense morality demurs on just this claim, in effect claiming that humans are morally special (for a defence of speciesism, see Kittay, this volume). But this looks to be speciesism, prescribing discriminating in treatment between beings whose morally relevant trait possession is the same, on the morally arbitrary basis of species membership. So perhaps the RAC
Richard Arneson
315
account advocate should not be moved by the common-sense criticism, and we should not be troubled by RAC’s revisionary implications for choice of policy and conduct in this instance. I can think of two reasons for being sceptical of this diagnosis. One is that the example of the individual suffering from advanced Alzheimer’s differs from an individual animal of comparable mental capacity in that the former has the status of having once been a person and the latter has no such status. This might rule out killing former persons now severely cognitively impaired merely for convenience. Another possibility is that the common-sense views can be rationalized by assimilating them to other instances of permissible partiality in non-consequentialist moral outlooks. Morality might permit favouring members of our species just as it permits favouring members of our own family over non-kin, our friends over strangers, and, according to some, our fellow countrymen over citizens of other lands. The question would then become why favouring members of our own species resembles the good form of partiality (for example, to friends) and not the bad form of partiality (for example, to white men). One might hold that one of the necessary conditions for acceptable partiality is that it is extended only to individuals to whom one has close personal acquaintance, and hence partiality extended across large anonymous groups such as nation states, supposed races, ethnic groups, those who are biologically male rather than female or the reverse, and also biological species is morally inadmissible. Without diving deeper into these issues, I take it that the RAC account so far has a host of plausible implications and, at least, the jury is still out on the moral status of humans who are from the start of their lives or later become severely deficient in RAC terms.
4. A Devastating Objection The simple RAC view runs into another problem that emerged into view in philosophical discussions of abortion and infanticide in the 1970s (Tooley 1972; Warren 1973; 1984). The RAC account yields the verdict, which many find plausible, that the human foetus lacks developed rational agency capacity anywhere near (any plausible specification of ) the threshold level that confers personhood. However, even after becoming sentient sometime around twenty-two weeks after conception, the late-stage foetus (and the newborn human as well) will lack personhood status on the same basis. The plausible RAC account of the moral permissibility of abortion seems also bound to insist on the moral permissibility of infanticide.
316
Basic Equality, Rational Agency Capacity, and Potentiality
Perhaps the RAC account of what makes a human individual a person, with special moral status, could accommodate the implication that a newborn human baby or a young infant fails to qualify as a person. We treat human newborns and infants as if they were full persons, and morality requires us to do this, for several obvious reasons. Our psychological dispositions are such that involvement with newborns, especially in parental roles, triggers attachment, love along with desires to protect, care, and nurture the helpless young human. Regardless of the moral status of the newborn, it would be a great harm to its parents if others treated it or displayed attitudes to it that showed anything less than the attitudes of respect and concern we accord to full members of the moral community. Moreover, to give the person the newborn human will become good life prospects, the newborn needs to get enormous love, care, protection, and nurturance now. Also, once out of the womb, the human baby starts interacting with other humans, making eye contact, experiencing, learning, in impressive ways. The baby’s potential for rational agency capacities is gradually becoming transformed into actual developed capacities. We could interpret the situation in these terms. There is implicitly a convention, a moral rule we adopt in our society, requiring us to treat human babies as if they were full members of the moral community. A convention is not a rule that registers at the level of fundamental moral principles, but is one we adopt to facilitate their fulfilment. Think of the rule that we all should pay taxes. We could fund government in other ways, but uniform taxes seem a good way, and once it is in place, it binds us. However, this proposal for reconciling the RAC account of personhood and moral status with deeply rooted convictions about our moral duties to newborns infants and children looks increasingly implausible as applied to young children as they develop. The difficulty is that it seems that the grounds for holding that each of us has moral duties to show special care and concern for a human child we encounter, more than whatever duties we have towards a nonhuman animal, are duties owed to this child, based on its valuable traits that command moral respect. At the limit, it would surely be wrong to kill a young human child just for convenience, whereas it would be morally permissible, many of us think, to kill a morally considerable nonhuman animal for convenience. (For example, imagine a family is moving to a new city, where for various reasons it would be very hard to continue to care for a pet dog Fido. They impose painless death on Fido. This is not a murder.) It does not seem possible to explain this view of our duties to young children, supposing we accept it, on a RAC account, in terms of the developed
Richard Arneson
317
actual RAC of the human child compared to that of the nonhuman animal that lacks a serious right to life. True, a rabbit child will have less developed RAC than the human toddler. But compare a young chimpanzee child and a human child at or just before two years of age, before the human child starts to develop capacity for language use. By any sensible cross-species intelligence or cognitive ability tests we can devise, the human child and the young chimp will score about the same. Nor is there reason to attribute greater volitional capacity to the human child than to the chimp. But I suppose most will agree that the prelinguistic human child has a serious right to life, of the sort that belongs to persons, and that is much stronger than our moral duty not to kill a young chimp except for good moral reasons. The RAC account comes up short here. Call this the ‘young childicide’ problem. A complication here is that the prelinguistic human does have some developed affective capacity beyond what the chimp has, and we are supposing that affective capacity is a component of RAC. A very young human child, placed so it sees someone stumble crossing a doorway, will reflexively reach out its hand to help, whereas a chimp that sees the same situation will show no such behavioural sign of being affected in feelings by what is seen and of caring a bit that bad things not befall the stranger such as taking a tumble. But this does not seem a sufficient basis for our view of the moral status of the young human child versus that of the young chimp. (Suppose human children did not start exhibiting any difference in affective capacity compared to young chimps until after developing capacities for language use.) To repeat: the simple RAC view appears committed to the in principle moral permissibility of ‘young childicide’. Here ‘in principle permissibility’ registers the thought that no traits of the young child bar killing him for mere convenience, though extraneous moral considerations might do so. Of course, in a very short time, the human child will start learning to talk and thereby think and reason competently. Thereafter the human child, though still immature, greatly outscores the chimp in cognitive abilities and overall RAC. But the difficulty for the RAC account lies just there. The future developed capacities of the prelinguistic human toddler do not register as boosting its present developed capacities. What intuitively distinguishes the young chimp and the young human child (prior to acquiring language capacity) is that the normal human has great RAC potential that the chimp lacks. But claiming that possession of potential for RAC at the threshold level of personhood confers personhood status would be endowing not only the newborn human and the young human child with personhood status, but also the human foetus just after conception (if not the separated sperm and egg prior to conception). The RAC account looks to be unable to accommodate
318
Basic Equality, Rational Agency Capacity, and Potentiality
strongly entrenched moral convictions, some of which would appear likely to withstand extended critical deliberation and persist in reflective equilibrium. One can understand why some writers on abortion have been resistant to the idea that having potential for RAC makes a being a person or contributes significantly to its having personhood status. After all, it would be fallacious to infer from the fact that an entity is a potential X to the claim that the entity has the properties and prerogatives of an X. But one can incorporate potentiality in a version of a RAC account without committing this mistake.
5. A Modified RAC Account Combines Potential and Developed Capacity This talk of potential for RAC needs explication. One has a developed capacity, for example, to speak French just in case one is able to converse in French if alert and awake and provided with appropriate stimulus. One has a potential capacity to speak French just in case one could undergo a learning process that would bring about this developed capacity. In general, one has a potential capacity of a sort if in reasonably favourable circumstances one develops the capacity and important causes of the development of the capacity (if that happens) are internal to the individual who is claimed to have the potential of that sort (but cf. Vallentyne 2007, doubting that the idea of potential can bear the argumentative weight I seek to place on it). So understood, having a potential capacity varies by degrees along two dimensions. A potential for a capacity is more robust, the less favourable the circumstances need to be for development of the capacity to occur, and a potential is more intrinsic, the more it is the case that important causes of the development of the capacity are internal to its bearer. The more robust and the more intrinsic an individual’s potential for a capacity, the greater the degree to which the individual has the potential. One might have a potential at one time and not at another. On the account of potential for capacity just sketched, Arneson has only a small capacity for learning quantum physics, because circumstances would have to be very favourable indeed, involving years of study under dedicated and patient teachers, for this capacity to develop in him. A human foetus just after conception, initiating a normal pregnancy, has a good potential to develop the capacities that would render it a person. If a ruthlessly efficient and entrenched dictatorship kills all newborn Jewish children born on the territory it controls, these newborns have potential for personhood, because under readily imaginable and generally available favourable circumstances, they would undergo this development. This can be so even if there
Richard Arneson
319
is no close possible world in which the dictatorship collapses or mends its ways so that a Jewish newborn has a good chance of surviving past infancy and beyond.¹ A modified version of the RAC account of personhood can incorporate potential capacity in a way that accommodates the judgement that the young child by virtue of possession of RAC traits has a serious moral right to life greater than that of the chimp while the human foetus lacks such a serious right to life. The modification is to maintain that having ‘enough’ developed RAC is necessary for personhood, and at any level of developed RAC having greater potential for RAC up to the personhood threshold amplifies one’s moral considerability. The scare quotes around the word ‘enough’ in the previous sentence indicate that a satisfactory elaboration of the view would have to provide a non-arbitrary specification of the sufficient level. I suppose one should work to specify this level by considering a range of examples involving beings with a range of developed RAC along its various dimensions and judging in which examples there is and is not full personhood. The threshold reached need not involve any discontinuous jump at any particular magnitude, just a point along a continuum. In other words, what needs to be acknowledged is that both potential RAC capacity and developed RAC capacity matter for moral considerability status and that the function that takes us from a being’s potential at a time and developed capacity at a time to its moral considerability status is nonadditive. Very roughly, think of the function as multiplicative, so that if either a being’s potential or developed RAC at a time is zero, their product is zero. Having zero developed RAC, the human foetus is not morally considerable, even if its potential RAC is deemed virtually identical to the potential RAC of the young child or for that matter an ordinary adult. Attaining sentience, the human foetus has a tiny bit of developed RAC, and thus a total RAC score that is higher than that of an identically sentient young rabbit. The newborn human has roughly the same developed RAC as the late-stage foetus, but after birth acquires developed RAC rapidly. Adding to the RAC account the idea that potentiality has importance enables it to deliver more nuanced and plausible verdicts about moral status than declining to acknowledge potentiality’s significance. Applied to the puzzle of the status of the young chimp and the prelinguistic human, the hybrid view delivers the verdict that by virtue of having greater potential RAC along with comparable developed RAC, the prelinguistic child is morally more ¹ I thank Reuven Brandt for calling this example to my attention and helping me see that an account of potential in an earlier draft of this chapter yielded an unsatisfactory judgement about it.
320
Basic Equality, Rational Agency Capacity, and Potentiality
considerable, considerably more person-like than the chimp, so has a much more serious right to life. Depending on weights and numbers assigned, the modified RAC account might even rule out infanticide as impermissible. It readily rules out ‘youngchildicide’. More would need to be said about how potential and developed RAC interact to determine a being’s personhood moral status, before we would be in position to assess the proposed modification of the RAC account. The loose sketch offered so far just provides a framework within which specific candidate views might be developed. One might hold that at some levels of developed and potential RAC, a being with greater potential RAC and lesser developed RAC can be all together morally more considerable, closer to personhood status, than another being with lesser potential RAC and greater developed RAC. One might also hold that a being with potential for personhood-level RAC, as its developed RAC increases, at some point of increased developed capacity becomes very close to being as morally considerable as a being with developed RAC at the personhood threshold. In this circumstance, further RAC gains attained for this being increase its moral considerability only a little. Notice, as loosely sketched, the modified RAC view does not entirely remove the counterintuitive shadow that we have seen falling on the RAC account. That is to say, neither the modified RAC nor the simple RAC account yields the verdicts that the newborn baby, the infant, or even the young toddler attain full personhood status, given their developed RAC remains sub-threshold.² But this may not matter much. Notice that on the modified RAC account, (1) a baby that has the potential for personhood-level RAC and has enough developed RAC, and even more definitely (2) the young child with even greater developed RAC and that same potential for personhood-level RAC, are so close to the full personhood status attained by the older child that has developed linguistic competence and full personhood that for all practical purposes both (1) and (2) ought to be treated as though having attained personhood status. (Here a near miss is almost as good as hitting the threshold). But if you insist that an acceptable doctrine of basic equality must hold that a prelinguistic human toddler of normal range developed and potential RAC has exactly the same equal moral status as those who have gained threshold level RAC and are fully persons, the RAC account sketched here falls short. ² I thank Giacomo Floris for pressing this issue on my attention, and for his further insightful comments on a draft of this chapter.
Richard Arneson
321
Notice though that your being of slightly lower moral status than full persons does not rule out such judgements as that you rather than a full person have a greater claim to be rescued if only one of the two can be saved. Other moral considerations can tilt the balance of reasons. For example, the toddler can expect to gain many more years of valuable life if she survives the present peril than the older person who could instead be saved.
6. Discussion of Other Views Don Marquis (1989) argues that almost any abortion of a human foetus cuts short a future life like ours (of great value) and is almost always morally wrong for that reason. In contrast, the modified RAC account holds that a necessary condition for a being’s having personhood status is that it has some level of developed RAC and that this, plus having potential for threshold RAC, nearly suffices for personhood. David Boonin (2003) amends Marquis by saying that the combination of having a future life like ours plus some developed capacity for mental activity suffices for personhood. This view is a close cousin of the position I am adumbrating and deserves further discussion. Gina Schouten (2017) suggests that IF the foetus is a person from the moment of conception, then the human community, not uniquely or especially the pregnant woman, owes it care and nurturing (so perhaps abortion would be impermissible but the pregnant woman should then be compensated for bringing her pregnancy to term). Boonin (2019) disagrees and holds that abortion would be permissible even if the antecedent of the conditional is true. Modified RAC denies the antecedent of this conditional. Agnieszka Jaworska and Julie Tannenbaum (2014; 2021) ingeniously propose that being able to participate with success as rearee in a rearing relationship that is rationally guided by the aim of full development of RAC in the rearee is necessary and sufficient for personhood status. They add that by this criterion, someone with a duty to care for a young human, even one with tragically limited RAC potential, is rationally guided by the aim of full development, and so the rearee in this rearing relationship qualifies as a full person. In contrast, someone with a duty to care for an animal, such as a dog, would not be rationally guided in rearing activities by any such aim, since a dog can fully flourish without developed RAC at this threshold. My concern about this proposal is that it will either turn out to be over-inclusive (the dog qualifies as full person) or veers into speciesism (the human with no more
322
Basic Equality, Rational Agency Capacity, and Potentiality
RAC potential than, for example, a dog qualifies as a person but the dog is arbitrarily excluded from this status). The problem is that it is plausible to hold that there is one invariant ideal of living a flourishing life, the same for all beings that have any capacity at all for good life. It would be wonderful, for example, if dogs could learn physics, without offsetting losses. So the individual who has a duty of caring for a young dog is rationally guided by the aim of developing full RAC (as much RAC as possible) in the dog rearee, just as much as would be the case if the rearee were a young human with tragically low RAC potential. Shelly Kagan (2016) argues in effect that a living human with no potential for acquiring personhood status and no more potential for developed rational capacity or developed rational agency capacity than a nonhuman animal might yet have a higher moral status than that animal by virtue of being what he calls a ‘modal person’. A modal person is a being that metaphysically could have been a person, even if it is causally impossible that the being could ever become a person. To illustrate the idea, he distinguishes a human infant born with anencephaly, which prevents the upper brain from developing, even though the infant’s genetic instructions themselves don’t block development, and a conceivable infant born with a similar condition stemming from its genetic endowment, so this infant could not ever have been a person. Depending on the values that attach to modal personhood compared to other traits that confer greater moral considerability, Kagan’s view (stated at 14–15) allows the possibility that an anencephalic human infant might have greater moral status than an infant gorilla with greater developed rational agency capacities and greater potential for rational agency capacity. We should grant that Kagan has identified a coherent position, but I doubt that we should share his judgements about modal personhood enhancing moral status. In other domains metaphysical possibility seems inert in this regard. The sheer fact that Arneson metaphysically could have been a great violinist or an excellent football player, even though he utterly lacks developed capacity and potential in these respects, does not enhance his human perfection achievement capacity status. Another worry is that by parity of reasoning, Kagan’s strategy for defending a view that is uncomfortably close to speciesism would serve to rationalize views uncomfortably close to racism. It is metaphysically possible that all individuals of European ancestry could have belonged to a white race of superior qualities. On a more ecumenical note, I conclude, these issues regarding the boundaries of personhood, besides being politically divisive in our time, are also intellectually hard to resolve, and merit further study.
Richard Arneson
323
References Arneson, R. 1999. ‘What, if Anything, Renders All Humans Morally Equal?’. In Peter Singer and His Critics, edited by D. Jamieson, 103–128. Oxford: Blackwell. Arneson, R. 2015. ‘Basic Equality: Neither Rejectable Nor Acceptable’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by U. Steinhoff, 30–52. Oxford: Oxford University Press. Boonin, D. 2003. In Defense of Abortion. Cambridge: Cambridge University Press. Boonin, D. 2019. Beyond Roe: Why Abortion Should Be Legal—Even if the Fetus Is a Person. Oxford: Oxford University Press. Carter, I. 2011. ‘Respect and the Basis of Equality’. Ethics, 121: 538–571. Cohen, G. A. 2013. ‘Notes on Regarding People as Equals’. In Finding Oneself in the Other, edited by M. Otsuka, 193–200. Princeton and Oxford: Princeton University Press. Jaworska, A., and Tannenbaum, J. 2014. ‘Person-Rearing Relationships as a Key to Higher Moral Status’. Ethics, 124: 242–271. Jaworska, A., and Tannenbaum, J. 2021. ‘The Grounds of Moral Status’. Stanford Encyclopedia of Philosophy, rev. https://leibniz.stanford.edu/. Kagan, S. 2016. ‘What’s Wrong with Speciesism? (Society for Applied Philosophy Annual Lecture 2015)’. Journal of Applied Philosophy, 33: 1–21. Marquis, D. 1989. ‘Why Abortion Is Immoral’. Journal of Philosophy, 86: 183–202. McKerlie, D. 2013. Justice between the Young and the Old. Oxford: Oxford University Press. McMahan, J. 2002. The Ethics of Killing. Oxford: Oxford University Press. McMahan, J. 2007. ‘Infanticide’. Utilitas, 19: 131–159. McMahan, J. 2008. ‘Challenges to Human Equality’. Journal of Ethics, 12: 81–104. Nozick, R. 1974. Anarchy, State, and Utopia. New York: Basic Books. Nozick, R. 1997. ‘Do Animals Have Rights?’ In Socratic Puzzles, 305–310. Cambridge, MA: Harvard University Press. Nussbaum, M. 2006. Frontiers of Justice: Disability, Nationality, Species Membership. Cambridge, MA: Harvard University Press. Rawls, J. 1999. A Theory of Justice, rev. ed. Cambridge, MA: Harvard University Press. Schouten, G. 2017. ‘Fetuses, Organs, and a Famous Violinist: On the Ethics and Politics of Abortion’. Social Theory and Practice, 43: 637–665. Sher, G. 2015. ‘Why We Are Moral Equals’. In Do All Persons Have Equal Moral Worth?: On ‘Basic Equality’ and Equal Respect and Concern, edited by Uwe Steinhoff, 17–29. Oxford: Oxford University Press. Singer, P. 2011. Practical Ethics, 3rd ed. Cambridge: Cambridge University Press. Tooley, M. 1972. ‘Abortion and Infanticide’. Philosophy and Public Affairs, 2: 37–65.
324
Basic Equality, Rational Agency Capacity, and Potentiality
Vallentyne, P. 2007. ‘Of Mice and Men: Equality and Animals’. In Egalitarianism: New Essays on the Nature and Value of Equality, edited by N. Holtug and K. LippertRasmussen, 211–237. Oxford: Oxford University Press. Waldron, J. 2017. One Another’s Equals: The Basis of Human Equality. Cambridge, MA: Harvard University Press. Warren, M. 1984. ‘On the Moral and Legal Status of Abortion’. The Monist, 57, 1973, reprinted with ‘Postscript on Infanticide’. In The Problem of Abortion, edited by J. Feinberg, 102–119. Belmont: Wadsworth Publishing.
Index For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. A ableism, 16, 275 abstract equality, 65, 75 Anderson, E., 8–9, 77 acceptable vs unacceptable social hierarchies, 139–140 moral equality, 88 relational egalitarianism, 81, 233–234 social (in)equality, 81, 92–94, 97–99, 234–235 wrongness, expressivist theory of, 92–93 animals (nonhuman animals) (1e) Basic Equality (of Living Entities), 23–24, 26–27 animal rights, 24–25, 150, 262 animal welfare, 262, 265–266 autonomy, 165, 302 children’s moral superiority over, 28 comparisons between children and, 316–322 comparisons between cognitively disabled humans and animals, 280, 313, 315, 322 comparisons between cognitively disabled humans and animals (moral individualism), 25–26, 264–270, 272–276, 279 concepts of time, selfhood, and reasons, 291–292 consequentialism, 27, 107–109, 296–297, 299–302 contractualism, 27, 300–302 as each other’s equals, 23–24, 308 equality-grounding properties of, 228–229 higher-primates, 24–25, 263, 311–312, 316–317 human and animal interests, 110–112, 292, 298–300
human equality renders nonhumans as moral inferiors, 262 human moral superiority over, 24, 26–27, 73, 199, 210, 217, 262, 306 humans/animals moral equality, 27, 287, 290–292, 294–296, 298, 301–303 humans’ duties towards, 27, 292, 293–296, 301, 302–303, 311 humans’ greater mental abilities, 27, 295, 299–302 ill-treatment of, 266–270, 290 Kant, E. on, 301–302 killing of, 266–267, 293–294, 299–300, 313, 314, 316 Korsgaard, C. on, 295, 301–302 moral and epistemic access to, 282–283 moral equality and animal ethics, 10–11, 103, 110–115, 119, 121 moral inequality between human beings and other animals, 27–28, 178–179 moral standing, 290–291 moral status, 10–11, 26–27, 200, 263, 287, 290–291 moral status hierarchy and animal ethics, 113–115 moral status and subjective perspective/consciousness, 26–27, 287, 290–291, 306 RAC account and, 311–312 rational capacity, enhancing of, 161–162, 171 right to life, 150 same moral standing as humans, 287, 291, 292, 295, 301 sentient/non-sentient animals, 23–24 Singer, P. on, 110–112, 121, 265–268, 270, 292–293, 299–300, 308 social status hierarchy and animal ethics, 111–112
326
Index
animals (nonhuman animals) (Continued) social status hierarchy between animals and humans, 120–121 speciesist account of moral equality and, 265–266 as ‘subjects-of-a-life’, 23–24 transgenic comparisons of goodness, 275–276 treating humans and animals differently, 27, 295 Variation Problem, 311 Aquinas, T., 50 Arendt, H., 42–43, 47 political equality, 225, 232–234 Aristotle, 40–43, 55, 188–189 Metaphysics, 39–40 ‘natural’ slaves, 221–222 Nicomachean Ethics, 40–41 Politics, 40–41 Arneson, R. basic human equality as modern idea, 36 consequentialist/non-consequentialist divide, 5–6, 85–86 Continuity View, 211 moral status, 84–86 opacity respect, 134–135 ‘pressure of reason’, 15, 175, 177–179 RAC, 27–28, 224–225 artificial intelligence, 24–25, 264 autonomy animals, 165, 302 conscious beings, 150 living entities, 150, 165 Moral Status #2, 20, 201–202 moral status hierarchy and, 150, 158, 165–167 B Ball, J., 50–51 bare property, 19–20, 179, 291–292 basic equality classical social contract theory, 6 complacency about, 4, 12, 232–233, 235 ‘equal concern and respect’, 4, 72–73, 202 justice theories, 1, 3 key theses, 2 moral theory, 6 normative theory, 1, 3–4, 6, 10–12, 35–36 as ‘one of the most profound problems of moral philosophy’, 3
opacity respect and, 127, 129, 130–131, 135, 145 problem of, 177–180 virtue theory, 6 basic equality: multiple meanings, 4, 14 abstraction amidst multiplicity, 6–7 polysemy and disunity in moral, political, and legal theory, 6–7 Basic Equality Thesis (1), 2, 4, 6, 176 (1∗ ): No Basic Equality Thesis, 14–16 (1a): being equals in a (particular) morally significant sense, 4 (1b): being equals in some (any) morally significant sense, 5 (1b), rejection of, 14 (1b) and relational equality, 10 (1c): Basic Equality (of Human Species), 22–23, 25–26 (1d): Basic Equality (of Persons), 23–24, 28 (1d), rejection of, 23–24 (1e): Basic Equality (of Living Entities), 23–24 basic human equality: historical emergence, 35, 221–222 (4b) historical thesis, rejection of, 7–8 BHE as historical achievement, 37–38 BHE as modern idea, 36 BHE and normative egalitarianism, 35–37, 54–55 BHE as parity relation and gradability objection, 37, 55–56 BHE as sameness, 35–37, 54–56 blind spots in contemporary moral and political philosophy, 37–38, 56–57 coinage metaphor, 54 ‘egalitarian’: as late nineteenth-century coinage, 38–39 egalitarian consequences, lack of, 38–39 equality as relational property: balance, indifference, parity, 54 failure of historicism, 38–39 high-road histories, 37 human equality as normatively hierarchical principle, 37, 47–50 modern BHE as combination of ‘equality-as-parity’ and ‘equality-as-indifference’, 54 nature/facts of nature, 222–223 political equality, 54–55
Index see also Christianity; Greece; Levellers; parity; premodern concepts of equality; Rome Basic Inequality Thesis (9), 16, 177–178 discrimination, 16 unequal moral worth as ground for, 16 basic and relational equality, 9, 67 (5d) relational egalitarianism thesis, 9 (5e) relational grounding egalitarianism thesis, 9 basic equality, non-relational implications, 9, 77 basic equality, normative implications, 72 basic equality as ground for relational equality, 77–78 basic equality and human dignity, 73–74 basic equality and moral status, 74 basic equality as normative position, 72–75 basic/relational equality distinctions, 65–67, 71–72, 76, 78 basic/relational equality similarities, 70–71, 75 concept/conception distinction, 75 moral realism and, 68 relational equality: normative implications of, 72–73 relational equality as normative position, 75 supervenience and basic equality, 69, 71–72, 74–75, 78 supervenience and relational equality, 70 supervenience and shapelessness, 71 as two distinct normative commitments, 9, 67–68 see also relational equality Bejan, T., 7–8, 221 Bellarmine, R. (Cardinal), 51–52 Bennett, C., 133–134 Bentham, J., 110–111, 272–273 Berlin, I., 13–14 BHE (basic human equality), see basic human equality: historical emergence Bible Acts of the Apostles, 48 Book of Job, 49 Genesis, 38, 48 Gospel of Matthew, 48–49 see also Christianity
327
Boonin, D., 321 Burns, R., 54 Buss, E., 252–253 C Carter, I., 206, 212 basic equality: ‘equality of certain entitlement-grounding properties’, 6, 12 gradability objection, 56 opacity respect, 18, 25, 241–244 ‘Respect and the Basis of Equality’, 17–18 caste, 221–222, 305 CEDAW (Convention on the Elimination of All Forms of Discrimination Against Women), 105 Chang, R., 55–56 children’s moral equality, 241, 257 anencephalic babies, 207, 209, 322 attitude-based account of the basis of moral equality, 243–244 babies’ moral status and equality, 199, 207–210, 217 basic equality, 24–25 capacity to flourish, 245–247, 254 care-rearing relationship as basis for moral equality, 321–322 choosing between a child and an older person’s life, 320–321 civil/political status and rights, 168–169 comparisons between children and animals, 316–322 as constitutive requirement of what is owed to children qua moral status-holders, 25, 246, 249–250, 255–258 equal moral status, 151–152, 168–169 friendship, 250 friendship (integrated), 25, 242, 253–254, 257–258 inferiorizing consideration and treatment, 242, 248–250, 254–256 killing of impaired infants, 264–265 moral inferiority and, 242, 246, 250, 255 moral personality and, 25, 241–242, 245–247, 250–254, 257–258 moral status, 151–152, 168–169 moral superiority and, 242, 250, 251–254 moral superiority over nonhuman animals, 28
328
Index
children’s moral equality, (Continued) newborns, 316–320 opacity respect and, 25, 241–246 parent/child relationship, 264–265, 276, 277–278, 316 positive respect and, 25, 241–242, 245–247, 249–257 RAC account: abortion and ‘young childicide’ problem, 28, 315–318, 321 racism/racial segregation, 248 rational agency, potential for, 254, 316 relational resources, 250–251 right to health care, 246–247, 249–250 self-respect, 25, 242, 246, 250, 254–258 sense of justice and conception of the good, 245, 250–253 status-conferring property, 241–242, 245–247, 249–250, 254, 256–257 superior consideration and treatment, 252–254 Variation Problem and, 25, 242–243, 246, 247, 249–250, 254, 257 with respect to one another, 241, 247, 256 see also positive respect Christianity (5b) Egalitarian Thesis, rejection of, 7–8 Aquinas, 50 early Christianity, 38–39, 46, 48–50, 221 ‘equality-as-balance’, 7–8, 49 ‘equality-as-indifference’, 7–8, 48, 49, 51–52, 221 ‘equality-as-parity’, 7–8, 48–50 Gregory the Great (Pope), 49 human equality as normatively hierarchical principle, 49–50 Lactantius, 48–50 Mary Magdalene’s case, 221 medieval theology/thought, 49–52 men and women created in the ‘image of God’, 38, 48–49 natural law theories, 36, 51–52 Nietzsche: ‘death of God’, 189–190 Original Sin doctrine, 38, 49, 50 political equality, 51–52 popular sovereignty, 51–52 Roman influence on, 46, 48–50 see also basic human equality: historical emergence; Bible Christiano, T., 18–19, 212–213 Cicero, 44–47, 50
De Republica, 44 De Senectute (‘Of Old Age’), 46 classism as ‘blind spot’, 37–38 class contempt, 232–234 cognitive capacities cognitively disabled human beings, 23 infants, 23 persons as each other’s equals, 23 persons as superior to nonhuman animals and other living entities, 23 under-inclusiveness objection, 23–24 see also disability/impairment; RAC; rationality Cohen, G. A., 54, 306 Coke, E., 51, 56 consciousness/subjectivity animals: consciousness and moral status, 26–27, 287, 290–291, 306 autonomy of conscious beings, 150 humans as moral equals and, 26–27, 291, 306 humans’ moral status and, 97 moral standing and, 288–292, 297–298 moral status hierarchy and, 151 Variation Problem and, 242–243 consequentialism, 5–6 act-consequentialism, 27, 296–297, 299–302 interests: consequentialists/nonconsequentialists divide, 5–6, 85, 103, 110, 296–297, 301 moral standing and equality of interests, 5, 85 rule-consequentialism, 296–297, 301, 302 welfare consequentialism, 103–105, 107–110 contractualism, 6, 27, 300–302 Crary, A., 272–273 Cromwell, O., 52–53 D Darwall, S., 132, 230, 231 De Las Casas, B., 221–222 democracy and equality, 99, 131–132, 144 equal public status, 151–152 Greece (classical), 41–43 liberal democracies, 232–233 right to vote, 232–233 De Waal, F., 275–276
Index dignity, see human dignity Dillon, R., 248 disability/impairment (people with), 25–26, 261 Alzheimer’s disease, 279–280, 283–284, 312–315 anencephalic babies, 207, 209, 322 being human as sufficiency condition for moral equality, 25–26, 262, 263, 282, 284 care-rearing relationship as basis for moral equality, 279–280, 283–284 children with severe disabilities, 264–265 cognitively disabled human beings, 23, 25–26 coma and vegetative state, 279–280 comparisons between cognitively disabled humans and animals, 280, 313, 315, 322 comparisons between cognitively disabled humans and animals (moral individualism), 25–26, 264–270, 272–276, 279 congenital disability/impairment, 264– 265, 272–275, 277–278, 284 Convergent Assimilation, 274 genetic conditions, 275, 279–280 humans with severe mental disabilities and basic equality, 24–25, 241 ill-treatment of people with disabilities/impairment, 268–270, 281 killing of impaired/disabled human beings, 264–265, 267, 269, 272, 273–274 McMahan, J. on, 26, 272, 273–274, 276–279, 281–284 mercy killing, 313–315 modal personhood, 322 moral and epistemic access, 282–283 moral equality and extreme disability argument, 19, 162–163, 168 moral responsibility/credit/worth and exclusion of humans without agency, 194 moral status and relational properties, 26 ‘mother’s child’ (relational property), 26, 278, 284
329
Nazi killing of disabled people, 276, 280–281 nonhuman animals as intellectually superior/on par with people with severe cognitive disabilities, 276 personhood and cognitive impairments, 263 RAC account and, 312 relational properties as basis for equality, 26 severely cognitively impaired humans: moral status and equality, 199, 207–210, 217 Singer, P. on, 25–26, 264–270, 272–273, 281 special relations as ground for moral equality, 26, 264–265, 276–278 speciesist account of moral equality and, 270 what about ‘us’ is morally relevant?, 279 who is the ‘us’?, 281 see also moral individualism; speciesist account of moral equality Disciplinary Thesis (2), 2 (2a): being equals in a (particular) morally significant sense, 5 (2a), rejection of, 6–7, 12 (2b): being equals in some (any) morally significant sense, 5–7, 12 (2b), rejection of, 11–12 Dworkin, R. and, 5 rejection of, 4, 10–11 discrimination, 141 Basic Inequality Thesis (9), 16 social status hierarchy and, 110, 117, 119 speciesism and, 22–23, 315 see also inequality; racism; sexism; speciesism distributive equality, 65 basic equality and, 66–68, 127 relational equality and, 8–9, 65–66, 76, 127, 233–234 Douglass, F., 226–228 Dworkin, R., 37–38, 73 concept of equality, 75 Disciplinary Thesis and, 5 ‘equal concern and respect’, 202 ‘life performance’, 183–184 people’s right to equality as ‘axiomatic’, 35 we are all one another’s equals’, 1
330
Index
E Ecumenical Thesis (7), 2–3 rejection of, 11 ‘egalitarian plateau’, 1, 3–4, 6–7, 37–38 Egalitarian Thesis (5), 2, 7 (5a): being equal in a (particular) morally significant sense, 7–8 (5b): being equal in some (any) morally significant sense, 7 (5b), rejection of, 7–8 (5c): Distributional Egalitarian Thesis, 8–9 (5d): Relational Egalitarianism Thesis, 9 (5d): Relational Egalitarianism Thesis, rejection of, 10 (5e): Relational Grounding Egalitarianism Thesis, 9 (5e): Relational Grounding Egalitarianism, rejection of, 10–11 egalitarian justice, 8–9 see also basic and relational equality; relational equality; substantive egalitarianism egoism doctrine, 305 England early modern England and basic human equality, 36–37, 54–55 ‘equality-as-parity’, 37, 50–52, 54–55 natural equality, 51–52 parity as synonymous with aristocracy, 51 Peasant’s Revolt, 50–51 see also basic human equality: historical emergence; Levellers environment (natural environment), 261, 277 ethicists, 23–24, 85–86 Euclid, 35, 39–40, 65 eugenics, 81, 281 evil, 16, 24–25, 262–263, 273 Explanatory Thesis (6), 2, 11 (1∗ ) No Basic Equality Thesis, 14–16 explanation of basic equality: problem of, 14 ‘grounding’ of basic equality, 14–15 rejection of, 21 Variation Problem, 15–16 expressivism basis for relational egalitarianism, 94 wrongness, expressivist theory of, 92–94
F feminism, 248, 253, 310 feudalism, 117–118 Fleischacker, S., 38 Floris, G., 25 Frankfurt, H., 182 G Geneva Conventions, 74 genocide, 280–282 good, conception of, 177–178 children, 251–253 opacity respect and, 243–244 Rawls, J., 17–18, 88–89, 127–128 Goodin, R., 277 Grandin, T., 295 Greece (classical) adult male citizens as isoi, 41–43, 47 agora/marketplace, 7–8, 39, 40–43 Aristotle, 39–43 Athenian citizens as isoi by birth, 41–42 basic human inequality, 41–42 coinage, 41–43, 48–49 democracy, 41–43 dokimasia, 42 ‘equality-as-balance’, 7–8, 39, 40–43 ‘equality-as-indifference’, 40 equality and justice, 7–8, 40–41, 43 Euclid, 39–40 human equality as normatively hierarchical principle, 50 isonomia and isegoria (democratic institutions), 41–43 isotes (equality), 7–8, 36, 39–41, 50 Isotêta, 7–8 Plato, 40–43 political exclusion of Athenian women, criminals, non-citizens, and slaves, 41–42 slavery, 41–42 to homoion (‘the like’), 39 to ison (‘the equal’), 39 to tauton (‘the same’), 39 see also basic human equality: historical emergence Gregory the Great (Pope), 49 grounding, 12–13 ‘grounding’ of basic equality, 12–14 grounding/justification distinction, 12–14
Index see also moral status and equality: grounding challenges H Hare, R., 91–93, 100 health-related issues children’s right to health care, 246–247, 249–250 COVID-19 pandemic, 270, 281 people with disabilities/impairment, 269–270, 281 see also disability/impairment Hill Jr., T. E., 247–248 Historical Thesis (4), 2, 7 (4a): being equals in a (particular) morally significant sense, 7–8 (4b): being equals in some (any) morally significant sense, 7 (4b), rejection of, 7–8 Hitler, A., 16 Hobbes, T., 153 homophobia, 232–233 Hull, D., 263 human dignity basic equality and, 73–74 critique of dignity as grounding for equality, 226–227 modern discourse: dignity and upward equalization of rank, 229–230, 305 as normative position, 73 pre-egalitarian discourse: dignity and hierarchies of rank and social standing, 229–230, 305 reason as basis for, 262 rights and, 74, 122 as status, 74 Waldron, J. on, 229–232 I inequality Greece: basic human inequality, 41–42 justification for, 223 Levellers: political inequality, 53–54 moral equality and material inequality, 21–22, 229, 232–235 Schaar, J. H.: ‘Nature … shouts ‘inequality’, 3
331
see also Basic Inequality Thesis (9); discrimination; moral inequality; political (in)equality; social (in)equality inferiority/superiority relations children: inferiorizing consideration and treatment, 242, 248–250, 254–256 children: moral inferiority, 242, 246, 250, 255 children: moral superiority, 242, 250, 251–254 children: moral superiority over nonhuman animals, 28 children: superior consideration and treatment, 252–254 colonialism and segregation, 204 human equality renders nonhumans as moral inferiors, 262 human moral superiority over animals, 24, 26–27, 73, 199, 210, 217, 262, 306 inferiority of entire populations, 261, 281 inferiorizing treatment, 226–227 premodern concepts of equality and, 47, 49–50, 52 racism, 248 relational egalitarianism and, 8–9, 81, 82 slavery and, 41–42, 224 social status hierarchy and, 10, 110, 116–118 treating one as inferior as injury to one’s sense of self, 207–208 universal prohibition against treating others as inferiors, 16–17, 178 women, 41–42, 224, 248 wrongfulness of, 110, 117 see also moral inequality; moral and social equality; social (in)equality interests consequentialists/non-consequentialists divide, 5–6, 85, 103, 110, 296–297, 301 human and animal interests, 110–112, 292, 298–300 interest-independent factors, 108, 116–117, 119–121 moral equality and equal consideration of interests, 4, 10–11, 84–86, 93, 116–117, 122, 288, 305 moral equality and social equality, 96, 98
332
Index
interests (Continued) moral standing: ‘having interests that matter’ formulation, 287, 296, 297 moral standing and equality of interests, 5, 85, 272–273 RAC account, 310–311 racism, sexism, and equal consideration of interests, 110–113 rights and equality of interests, 5 social status hierarchy/social inequality and, 114, 117, 122 subjectivity and, 288–290 utilitarianism, 85, 93, 111–113, 200, 211 International Covenant on Civil and Political Rights, 74 J Jaworska, A., 20–21, 24, 279–280, 321–322 Judaism, 38–39 justice/justice theories Aristotle, 40–41 basic equality and, 1, 3 capacity for moral personality and equal justice, 308–309 corrective (commutative) justice, 40–41 distributive justice, 40–41 duties of justice, 90–91, 250–251 egalitarian justice, 8–10 Greece: equality and justice, 7–8, 40–41, 43 Plato, 40–41 relational equality and, 90–91, 118–119 justice, sense of, 177–178 children, 245, 250–253 opacity respect and, 243–244 Rawls, J., 17–18, 55, 88–89, 127–128 justification, 13 justification of basic equality, 13–14 grounding/justification distinction, 12–14 justifying equality, 21–22, 222–224 equality as commitment and claim, 21, 224, 235 equality as conditional, 224–225, 231 equality needs no justification, 21, 225–229 inequality, justification for, 223 justification for gradation, 222 justification and legitimization of exclusion/subjugation, 21–22, 223, 235
logic of justification, 21–22, 224–225, 235 moral equality and material inequality, 21–22, 229, 232–235 moral/material distinction, 21–22, 229, 232 moral status and, 21–22, 231–232, 235 moral worth and, 21–22, 229, 230–232, 235 nature (facts of nature), 222–223 political commitment, 225–226 property-based justification, 224–227, 235 racism and, 21–22, 222, 223, 232–233 relational egalitarianism, 233–234 sexism and, 21–22, 222, 223–224 slavery and, 221–224, 226 Justinian: Digest, 36, 44–45, 51–52 K Kagan, S., 113–117, 119–120, 122, 322 Kant, E., 4, 206, 223, 271, 296–297, 301–302 killing of animals, 266–267, 293–294, 299–300, 313, 314, 316 difference between animals, 265 difference between killing persons and animals, 108, 264–265 of impaired human beings, 264–265, 267, 269, 272, 273–274 indignation and blameworthiness over, 216 McMahan, J.: ethics of killing, 264–265, 273–274 mercy killing (RAC account), 313–315 moral status and, 201–202, 204–205, 213, 216 Nazi killing of disabled people, 276, 280–281 right not to be killed, 204–205, 207, 213 of terrorists, 231 Kirby, N., 6 bare property, 19–20, 179 basic equality and ‘basic moral property’, 6 moral worth, luck and weight, 19–20 Kittay, E. F., 25–26 Korsgaard, C., 113–114, 133–134, 271, 275 animals, 295, 301–302 Kuhse, H., 264–265 Kymlicka, W., 75
Index L Lactantius, 48–50 Lawrence, D. H., 269 Levellers, 52–53 1647 Putney Debates, 37–38, 52 democratization of aristocratic dignity, 52–53 as ‘early egalitarians’, 36, 52–53 early modern England and basic human equality, 36–37 ‘equality-as-balance’, 53–54 equality as birthright, 52 ‘equality-as-parity’, 37, 52–54 Leveller women, 52–53 natural equality, 52 political inequality in Leveller thought, 53–54 see also basic human equality: historical emergence; England liberalism, 138–139 basic equality and, 127–128, 145 moral equality and, 103–104 opacity respect, liberal commitment to, 145 state-based egalitarianism and, 144–145 libertarianism, 203 Lilburne, J., 52–53 Lippert-Rasmussen, K., 5–6, 10–12 basic equality: same basic moral considerations apply to all humans, 6, 85–86 living entities (1e) Basic Equality (of Living Entities), 23–24 autonomy of, 150, 165 being a subject of a life as condition for moral status, 211 human moral superiority over all other living entities, 24 moral status of, 23–24, 149–150 Livy, 44 Locke, J., 38, 262–263 luck-egalitarianism, 65, 69, 73, 75–76 M MacDonald, M.: ‘Natural Rights’, 69–71 McDowell, J., 71 McMahan, J.
333
2008 Challenge of Cognitive Disability to Moral Theory conference, New York, 276–279 Convergent Assimilation, 274 critique of, 26, 272, 273–274, 276–278, 281–284 ethics of killing, 273–274 Ethics of Killing, The, 264–265 RAC account, 314 speciesism, 276 McMahon, D., 47 Marquis, D., 321 Methodological Thesis (3), 2 rejection of, 6–7, 10–11 Mill, J. S., 154–157, 223, 275 Miller, D., 82 modernity, 8 basic human equality as modern idea, 36 dignity and upward equalization of rank, 229–230, 305 early modern England and basic human equality, 36–37, 54–55 modern basic human equality as combination of ‘equality-as-parity’ and ‘equality-as-indifference’, 54 moral equality: role in major developments of modern era, 110 Montaigne, M. d., 155 moral equality, 10–11, 103–104, 175 animal ethics and, 10–11, 103, 110–115, 119, 121 basic moral equality, 81–82, 85–86 basic moral equality, popularity of, 10 ‘basic moral property’, 6, 175, 176–180, 185–186, 188–189, 195 binary/range properties and, 13–14, 106–107, 109 equal consideration of interests, 10–11, 110–113, 122 liberalism and, 103–104 moral personality and, 105–106, 308–310 non-consequentialist/consequentialist interpretations of, 103, 110 non-normative properties and, 15, 175, 177–180 rejection of/scepticism about, 10, 103, 109–110, 121–122, 175, 261–262 rights and, 104–105, 122, 150 role in major developments of modern era, 110
334
Index
moral equality (Continued) thinking of moral equality in different terms, 10–11, 109, 121 treating moral equals differently, 295, 299 Variation Problem, 108 virtue ethicists and, 85–86 see also moral and social equality; moral status and equality: grounding challenges; moral worth, luck and weight; rationality and moral equality moral equality, denials of, 95–97, 99–100 (a) type denials, 95–96 (b) type denials, 95–96 (c) elitist denial, 96 (d) non-elitist denial, 96 elitism dimension, 96 extremist denials, 96, 99–100 moderate denials, 96–97, 99 (socially salient) group dimension, 95–96 moral individualism, 263 children/humans with severe disabilities: killing of, 264–265, 267, 269, 272, 273–274 comparisons between cognitively disabled humans and nonhuman animals, 25–26, 264–270, 272–276, 279 critique of, 25–26, 264–265, 268–269, 272–273 empirical knowledge: importance for moral assessments of other humans, 266 intrinsic moral properties as basis for equality, 23, 25–26, 262, 263, 276, 279 intrinsic properties vs relational properties, 276 metaphysical grounding, 273, 283–284 moral individualism vs species membership, 263–266, 271–272 naturalistic conception of human equality, 262 parent/child relationship, 264–265, 276, 277–278 ‘personhood’ (moral) as criterium for moral status, 271–274, 281–282 personist view vs species membership, 272 rational capacity as intrinsic property, 264, 279 ‘rich psychological architecture’ as basis for moral status, 26, 273–275
sentient beings and, 264 see also McMahan, J.; Singer, P. moral inequality, 13–14, 106–107, 188–189 between human beings and other animals, 27–28, 178–179 moral status and, 200, 263 slavery and, 92–93, 100 unequal moral worth as ground for basic inequality, 16 see also inferiority/superiority relations moral personality (normative agency) basic equality and, 17–18 capacity for moral personality and equality, 308–309 children, 25, 241–242, 245–247, 250–254, 257–258 demand to be treated with equal concern and respect, 105–106 friendship and, 251–252 modal personhood, 322 moral equality and, 105–106, 308–310 moral individualism: ‘personhood’ as criterium for moral status, 271–274, 281–282 opacity respect and, 18 personhood and cognitive impairments, 263 RAC account and personhood, 308–313, 319, 320–321 RAC account and personhood boundaries, 321 range property, 18, 89, 106, 109 rationality as basis of personhood, 262–263 Rawls, J., 89, 308–309 rights and, 105–106 self-respect and, 247–248, 250 moral responsibility, 81, 181, 182, 194–195, 278 moral and social equality, 10–11, 81, 100, 117, 122, 305 anti-hierarchical ideal/non-hierarchical social order, 82–83, 87 basic moral equality, 81–86 grounding social equality on instrumental/non-instrumental grounds, 118 incommensurable moral status/incommensurability view, 10, 97–98
Index justificatory relevance claim, 82–83, 99–100 justifying social equality in the absence of moral equality, 95, 100 justifying social equality with moral equality, 89, 100, 118, 122 moral/social equality difference, 88–89 moral/social equality intimate connection, 89, 92, 100 moral status of all humans, 84 moral status hierarchy and social status hierarchy, 113–117, 122 social equality, 86–87 social hierarchies: omnipresence problem, 94–95 social inequality and moral equality, 88 social justice and, 118–119 social status hierarchy and moral equality, 10–11, 103, 110–122 social status (in)equality and moral equality, 10–11, 111–112, 117–118 sufficient moral status/sufficientarian view, 10, 98–99 see also moral equality, denials of moral standing all persons have equal moral standing, 297–298 animals, 290–291 animals: same moral standing as humans, 287, 291, 292, 295, 301 consciousness/subjectivity and, 288–292, 297–298 description, 287 equality of interests and, 5, 85, 272–273 ‘having interests that matter’ formulation, 287, 296, 297 human beings, 290 Sher, G. on, 5, 287, 296 treating moral equals differently, 295 Variation Problem, 288 moral status animal ethics and moral status hierarchy, 113–115 animals, 10–11, 26–27, 200, 263, 287, 290–291 animals: consciousness and moral status, 26–27, 287, 290–291, 306 Arneson, R. on, 84–86 autonomy and moral status hierarchy, 150, 158, 165–167
335
basic equality and, 74 children, 151–152, 168–169 description of, 200 different moral statuses, 149–150 explanatory role of, 149 human beings, 84 justifying equality and, 21–22, 231–232, 235 killing and, 201–202, 204–205, 213, 216 living entities, 23–24, 149–150 ‘moral considerability’, 200 moral status as distinct from civil and political status, 168–169 moral status as distinct from equal public status, 151 moral status hierarchy, 149–150 moral status hierarchy and rationality, 150–152 moral status hierarchy and social status hierarchy, 113–117, 122 ‘person-rearing relationships’ and, 24 psychological unity and, 26 rationality as ground for basic moral status, 18, 122, 149–150, 152–153, 224–225, 262, 264, 279, 322 rejection of the idea of, 84 Scope Problem and, 24–25 self-respect and, 247–248 sentient beings, 149–150 as universal to all humans?, 84 moral status: types, 200, 204 degrees of moral status, 20, 204, 214 grounds of moral status, 206 Moral Status #1 (existence and welfare), 20–21, 201 Moral Status #2 (autonomy), 20, 201–202 Moral Status #3 (opacity toward agential capacities), 20, 202–203 Moral Status #4 (treatment by social systems), 20, 203–204 see also moral status and equality: grounding challenges moral status and equality: grounding challenges, 20–21, 199 consistency condition, 201–203 Continuity View, 21, 199, 210–213 Continuity View: reasons for doubts, 213, 217 Discrete Interval View (Threshold), 199, 210, 211–213, 217
336
Index
moral status and equality: grounding challenges (Continued) grounds of moral status and importance, 205 human babies and severely cognitively impaired humans, 199, 207–210, 217 human moral superiority over nonhuman animals, 199, 210, 217 over-inclusivity, 207–209 under-inclusivity, 206–209 Variation Problem, 21 see also moral status: types moral worth justifying equality and, 21–22, 229, 230–232, 235 relational attributes and, 277–278 rights and, 231 self-respect and, 247–248 unequal moral worth as ground for basic inequality, 16 moral worth, luck and weight, 19–20, 175, 195 bare property, 179 basic equality, problem of, 177–180 ‘basic moral property’, 175–180, 185–186, 188–189, 195 circumstance, 181–182, 184, 185, 190–191 continuous options, 184–185, 192 desert, 188–189 life performance, 183–184, 190–191, 195 moral luck, 176, 185, 192 moral luck, causal, 188, 193 moral luck, circumstantial, 20, 176, 185–188 moral responsibility/choice, 181–182, 184, 194–195 moral weight, 176, 188, 195 moral worth, 20, 175, 176, 180, 185–186, 195 non-normative properties, 175, 177–180 objection: determinism, 193 objection: discontinuous options, 192 objection: exclusion of humans without agency, 194 objection: lack of alternative options, 194 rational agency, 20, 176, 179–182, 184, 185, 195 see also rationality and moral equality Morgan, G., 2, 4, 5–6, 13–14
N Nazism, 186–188 killing of disabled people, 276, 280–281 Negative Thesis (8), 3 rejection of, 10–12 rights and, 10–11 Nietzsche, F., 189–190 non-consequentialists grounding strategy, 109 interests: consequentialists/nonconsequentialists divide, 5–6, 85, 103, 110, 296–297, 301 moral standing and equality of interests, 5, 85 rights-based non-consequentialists, 5–6, 85, 103–105, 107–110 nonhuman animals, see animals normative theory basic equality and, 1, 3–4, 6, 10–12, 35–36 basic equality, normative implications of, 72 basic equality as normative position, 72–75 basic human equality and normative egalitarianism, 35–37, 54–55 foundational crisis in, 3 human dignity as normative position, 73 human equality as normatively hierarchical principle, 37, 47–50 relational equality, normative implications of, 72–73 relational equality as normative conception of human relations, 76–77 relational equality as normative consequence of basic equality, 66 relational equality as normative position, 67, 75 rights as normative positions, 73 see also basic and relational equality Nozick, R., 103, 308–309 O opacity respect, 17–18, 25, 241–244 Arneson, R. on, 134–135 basic equality and, 127, 129, 130–131, 135, 145 basic equality levels, 130 children’s moral equality and, 25, 241–246 compatibilism vs incompatibilism, 137
Index compatibilist account, 137–138 contexts of practical reasoning, 18, 129, 135 critique of, 129, 134–135, 137, 145, 241–242, 244–245 evaluation vs evaluative abstinence, 137, 140–142 as ‘evaluative abstinence’, 128–129, 243–244 as (in)appropriate moral attitude, 18, 128–129, 132, 134–135, 138–145 incompatibilist account, 137, 142 justification for, 132, 145 liberal commitment to, 145 Moral Status #3 (opacity toward agential capacities), 20, 202–203 as most influential attitude-based view of basis of moral equality, 25, 243–244 participatory perspective, 132–134, 244 range property and, 18, 127–129, 131–132, 135, 137, 143 rationalism and, 133 recognition respect, 132–134, 136 scientific/objective perspective, 132–134, 244–245 Waldron, J. and, 137–138 over-inclusiveness, 21, 24, 207–209, 279, 321–322 Overton, R., 52 P parity aequos et pares, 46–49 ‘equality-as-parity’ (Christianity), 7–8, 48–50 ‘equality-as-parity’ (England), 37, 50–52, 54–55 ‘equality-as-parity’ (Levellers), 37, 52–54 ‘equality-as-parity’ (Rome), 7–8, 47–48, 50 in ethics/Changian parity, 55–56 modern BHE as combination of ‘equality-as-parity’ and ‘equality-as-indifference’, 54 pares as ‘equals’, 51, 54–55 pares as ‘peers’, 7–8, 37, 46–50, 52, 53–54, 56 paritas (Rome), 46 parity/equality distinction, 47 primus inter pares (Rome), 46–48
337
paternalism, 110, 114–115, 119, 142–144, 201–203 patriarchy, 38, 86, 253 Phillips, A., 21–22 plants, 14–15, 108 Plato, 40–43 Pojman, L., 15, 55–56, 177–178 political (in)equality, 54–55 Arendt on political equality, 225, 232–234 Christianity: political equality, 51–52 Levellers: political inequality, 53–54 Polybius, 43–44 positive respect, 25, 241–242, 246, 247, 250, 251–253, 257–258 description, 241–242, 245–246, 257–258 question-begging objection, 254 sidestepping objection, 256 Variation Problem and, 246–247, 249–250, 254, 257 see also children’s moral equality premodern concepts of equality, 37 ‘equality-as-balance’ (Christianity), 7–8, 49 ‘equality-as-balance’ (Greece), 7–8, 39, 40–43 ‘equality-as-balance’ (Levellers), 53–54 ‘equality-as-balance’ (Rome), 43–44 ‘equality-as-indifference’ (Christianity), 7–8, 48, 49, 51–52, 221 ‘equality-as-indifference’ (Greece), 40 ‘equality-as-indifference’ (Rome), 7–8, 44–45, 47, 48, 51–52 ‘equality-as-parity’ (Christianity), 7–8, 48–50 ‘equality-as-parity’ (England), 37, 50–52, 54–55 ‘equality-as-parity’ (Levellers), 37, 52–54 ‘equality-as-parity’ (Rome), 7–8, 47–48, 50 see also basic human equality: historical emergence Putnam, H., 270 R RAC (rational agency capacity), 27–28, 224–225, 310 (1d) Basic Equality (of Persons), 28 abortion and ‘young childicide’ problem, 28, 315, 319–321
338
Index
RAC (rational agency capacity) (Continued) animals and, 311–312 appeal: RAC as simple unified explanation, 310–311 cognitive, affective, and volitional capacities, 308–311 controversial implication: severe mental impairment and loss of RAC capacities, 312 McMahan, J. and, 314 mercy killing, 313–315 modified RAC account: potential and developed capacity, 28, 310, 318, 321 personhood (moral personality), 308–313, 319, 320–321 personhood boundaries and, 321 potential to become rational agents and moral equality, 24 range property, 27–28, 308, 310–311 simple RAC account, 308–309 speciesism and, 312, 314–315, 321–322 species membership and, 311–312 Variation Problem, 311 Rachels, J., 113–114, 263, 265–266 racism, 16, 68–69, 114–115 as ‘blind spot’, 37–38 children as victims of, 248 equal consideration of interests, 110–113 justifying equality and, 21–22, 222, 223, 232–233 moral equality and, 21–22, 95–96, 110–111, 117–118 moral status hierarchy and, 113–115 racialization, 222–223 racial segregation, 248 self-respect and, 248 slavery and, 223–224 social status hierarchy and, 122 special relations and, 276–277 speciesism and, 312, 322 Rainborough, T. (Col.), 37–38, 52–53 range property arbitrariness objection, 17–20, 55–56 BHE and gradability objection, 56 as binary, 13–14, 56, 106–107, 109, 127–128, 306–307 context-dependence of, 135 critique of, 17–18, 27 moral equality and, 13–14, 106–107, 109
moral personality, 18, 89, 106, 109 non-scalar range property, 17–18 opacity respect and, 18, 127–129, 131–132, 135, 137, 143 RAC account, 27–28, 308, 310–311 rationality and moral equality, 18, 152, 159 Rawls, J. on, 17–18, 56, 68–69, 127–128, 152 respect for agency, 128–129 supervenience, 127–128 Variation Problem and, 27, 288 Waldron, J. on, 56, 137 see also bare property Rashdall, H., 95–96 rationality (9) Basic Inequality Thesis, 16 as basis for human dignity, 262 as basis of human equality, 15 as basis of humanity, 271 as basis of personhood, 262–263 divine rationality, 153 rights and, 150 Variation Problem, 15–16, 18 rationality and moral equality, 18–19, 149 arguments for equality, 18, 149, 158–162, 167–168, 171–172 autonomy and moral status hierarchy, 150, 158, 165–167 children: potential for rational agency, 254, 316 collective nature of rationality, 18–19, 153, 156, 163, 167, 212–213 continuity argument/Variation Problem, 18, 158–159, 242–243 contribution argument, 164, 169, 171 contribution argument: objections, 164–165 extreme disability argument, 19, 162–163, 168 fairness consideration, 18–20, 159–163, 166–167 greater rationality as ground for greater moral import, 15, 18, 152–153, 159 imperfect rational nature, 153 intuitive argument for equality, 18, 158 moral status hierarchy and rationality, 150–152 moral worth and rational agency, 20, 176, 179–182, 184, 185, 195
Index participation argument, 18–19, 162 participation argument: objections, 19, 163–164 positive argument for equality, 18–19, 149, 162, 170–172 possible world argument, 169–170 range properties, 18, 152, 159 rational capacity, 18, 149, 152, 157, 163–165, 168, 211–213 rational capacity, enhancing of, 155, 160, 161–162, 170, 171, 212–213 rational capacity, potential for, 155, 322 rational capacity variation, 158, 162–164, 166, 168, 171, 178–179, 242–243 rationality as ground for basic moral status, 18, 122, 149–150, 152–153, 224–225, 262, 264, 279, 322 rationality as non-normative property, 15, 175, 177–179 societal variation argument, 19, 165, 168–171 status individualism, 165–167 see also moral status; moral worth, luck and weight Rawls, J., 2, 4 conception of the good and sense of justice, 17–18, 55, 88–89, 127–128 Difference Principle, 203 friendship, 251 moral equality, 81–82, 88 moral personality, 89, 308–309 non-scalar range property, 17–18 range property, 17–18, 56, 68–69, 127–128, 152 self-respect, 247–248 Variation Problem, 17–18 Raz, J., 12 Regan, T., 23–24, 113–114, 211 relational equality, 8–9, 81 (1b) Basic Equality Thesis and, 10 Anderson, E. on, 81, 233–234 care-rearing relationship as basis for moral equality, 279–280, 283–284, 321–322 distributive equality and, 8–9, 65–66, 76, 127, 233–234 equality as relational property: balance, indifference, parity, 54 inferiority/superiority relations and, 8–9, 81, 82
339
justice and, 90–91, 118–119 as normative conception of human relations, 76–77 as normative consequence of basic equality, 66 normative implications of, 72–73 as normative position, 67, 75 supervenience and, 70 see also basic and relational equality; moral and social equality religion and equality, 38–39, 48, 261 see also Christianity; Judaism respect appraisal respect, 230–231 recognition respect, 132–134, 136, 230–231 as regard for an entity’s autonomy, 201–202 see also opacity respect; self-respect rights (8) Negative Thesis and, 10–11 animal rights, 24–25, 150, 262 autonomy, moral status hierarchy and, 151 basic equality and equality of interests, 5 children, 168–169 disability rights, 24–25 equal basic rights, 104–105 Geneva Conventions, 74 human dignity and, 74, 122 International Covenant on Civil and Political Rights, 74 moral equality and, 104–105, 122, 150 moral personality and, 105–106 moral worth and, 231 rationality and, 150 rights-based non-consequentialists, 5–6, 85, 103–105, 107–110 rights as fundamental, 5–6 rights as normative positions, 73 rights of the unborn, 24–25 UN Declaration of Universal Human Rights, 261 rights (specific rights) children’s right to health care, 246–247, 249–250 right not to be killed, 204–205, 207, 213 right to flourish, 282 right to freedom of expression, 150
340
Index
rights (specific rights) (Continued) right to life, 150 right to vote, 231–233 Rome (classical), 36 212 CE Edict of Caracalla, 47 aequalitas (equality) 36, 43, 44–46, 50 aequitas (equity), 43 aequos et pares, 47 Cicero, 44–47, 50 civil law, 7–8, 44–47 coinage, 48–49 ‘equality-as-balance’, 43–44 ‘equality-as-indifference’, 7–8, 44–45, 47, 48, 51–52 ‘equality-as-parity’, 7–8, 47–48, 50 equality/parity distinction, 47 ‘equal liberty’, 44 human equality as normatively hierarchical principle, 49–50 Justinian: Digest, 36, 44–45, 51–52 Livy, 44 natural equality of human beings: consequences, 45–46 natural law, 7–8, 44–47 omnes homines aequales sunt (‘all humans are equal’), 36, 44–45, 47 paritas (parity), 46 primus inter pares, 46–48 Roman constitution, 43–44 Roman Empire, 47–48 Roman jurisprudence and human equality, 36, 44–45 Roman women and foreigners, 47 Sabinus, 44–45 slavery, 44–45, 47 Ulpian, 44–45 see also basic human equality: historical emergence Rousseau, J.-J., 223 S Sabinus, 44–45 Sangiovanni, A., 5–6, 10–12 grounding and justification, 13–14 harm of treating one as inferior, 207–208 Humanity Without Dignity, 226–227 sense of self, 226–227 Scanlon, T., 232–233, 300–301 Schaar, J. H.: ‘Nature … shouts ‘inequality’, 3 Scheffler, S., 8–9, 66, 76–77, 82
Schofield, M., 44 Schouten, G., 321 Schroeder, M., 141–143 Scope Problem, 21–22 (1c) Basic Equality (of Human Species), 22–23 (1d) Basic Equality (of Persons), 23–24 (1e) Basic Equality (of Living Entities), 23–24 moral status and, 24–25 over-inclusiveness, 21, 24 under-inclusiveness, 6, 21, 23–24 see also over-inclusiveness; under-inclusiveness self-respect children, 25, 242, 246, 250, 254–258 definition, 247–248 inferiorizing consideration and treatment, 248–249 moral personality and, 247–248, 250 moral status and, 247–248 Sen, A., 2, 8 sentience/sentient beings basic equality, 308 moral equality and sentient beings, 104–105, 108, 109, 119 moral individualism and, 264 moral status of sentient beings, 149–150 sentience as condition for moral status, 211 sentience as ground for basic equality, 23–24 sentient/non-sentient animals, 23–24 sexism, 16, 68–69, 114–115 as ‘blind spot’, 37–38 equal consideration of interests, 110–112 justifying equality and, 21–22, 222, 223–224 moral equality and, 21–22, 110–111, 117–118 moral status hierarchy and, 113–115 social status hierarchy and, 122 Sher, G. (1e) Basic Equality (of Living Entities), 26–27 equality of interests, 5 moral standing, 5, 287, 296 subjective perspective/consciousness and moral status, 26–27 Singer, P., 23–24, 114–115, 119–120
Index Animal Liberation, 110–111 animals/animal ethics, 110–112, 121, 265–268, 270, 292–293, 299–300, 308 Continuity View, 211 critique of, 25–26, 122, 264–265, 268–269, 272–273, 276, 281 on disability/cognitive impairment, 25–26, 264–270, 272–273, 281 Should the Baby Live?, 264–265 speciesism, 276 slavery, 38 Aristotle: ‘natural’ slaves, 221–222 De Las Casas on indigenous inhabitants as slaves, 221–222 Greece (classical), 41–42 inferiority/superiority relations, 41–42, 224 justifying equality and, 221–224, 226 moral (in)equality and, 92–93, 100 racism and, 223–224 religion and, 38 Rome (classical), 44–45, 47 social inequality and, 91–92 social status hierarchy and, 119–121 voluntary slavery, 93 wrongness of, 93 wrongness, expressivist theory of, 92–94 Smith, A., 234–235 social (in)equality, 86–87 Anderson, E. on, 81, 92–94, 97–99, 234–235 animal ethics and social status hierarchy, 111–112 animals and humans: social status hierarchy between, 120–121 inferiority/superiority relations and social status hierarchy, 10, 110, 116–118 moral equality and social status hierarchy, 10–11, 103, 110–122 moral equality and social status (in)equality, 10–11, 111–112, 117–118 slavery and, 91–92 social status egalitarianism and, 10–11 wrongness of social status inequality, 111–112 see also moral and social equality speciesist account of moral equality, 22–23, 279
341
(1c) Basic Equality (of Human Species), 22–23, 25–26 acceptable/good partiality, 315 animal welfare and, 265–266 being human as sufficiency condition for moral equality, 25–26, 262, 263, 282, 284, 306 beings are ends and not mere means, 271 cognitive ability, speciesism, and moral standing, 262, 270 collective agency, 271 disability/impairment and, 270 moral individualism vs species membership, 263–266, 271–272 moral relevance of, 265, 270–271 personist view vs species membership, 272 RAC account and species membership, 311–312 species: evolutionary definition as based on relational properties, 263 species membership, 262–263, 265 species prejudice/‘the human prejudice’, 262, 273, 276 speciesism as discriminatory, 22–23, 315 McMahan, J. on, 276 RAC account and, 312, 314–315, 321–322 racism and, 312, 322 rejection of, 27–28, 110–111, 279–280, 312, 321–322 Singer, P. on, 276 Spiegelberg, H., 13–14, 20 status-conferring property, 16–17 (1d) Basic Equality (of Persons), 23 children’s moral equality and, 241–242, 245–247, 249–250, 254, 256–257 see also Variation Problem Steinhoff, U., 16, 55–56, 188–189, 230–231 Strawson, P. F., 132–133, 141–142, 244 subjective perspective, see consciousness/subjectivity substantive egalitarianism, 2, 7–8 (5a, 5b) Egalitarian Thesis, 7 (5b), rejection of, 7–8 suicide, 201–202 supervenience basic equality and, 69, 71–72, 74–75, 78 range property, 127–128
342
Index
supervenience (Continued) relational equality and, 70 shapelessness and, 71 T Tannenbaum, J., 20–21, 24, 279–280, 321–322 terrorism, 231 Tolstoy, L., 267–268 torture, 201–202, 204–205, 231 U Ulpian, 44–45 unborn, the abortion, 321 abortion and RAC account, 28, 315, 319, 321 human foetuses’ basic equality, 24–25, 279–280 personhood status, 315–316, 318–319 rights of the unborn, 24–25 UN Declaration of Universal Human Rights, 261 under-inclusiveness, 6, 21, 23–24, 206–209, 279 utilitarianism, 265–266, 308 interests, 85, 93, 111–113, 200, 211 ‘moral considerability’, 200 preference utilitarianism, 92 V Van Wietmarschen, H., 86–90 Variation Problem, 15–16, 242–243 animals, 311 children’s moral equality and, 25, 242–243, 246, 247, 249–250, 254, 257 grounding challenges of moral status and equality: Continuity View, 21, 199, 210–213, 217 ‘indirect’ (dis)solutions, 20–22 moral equality, 108 moral standing, 288 positive respect and, 246–247, 249–250, 254, 257 RAC account, 311 range property and, 27, 288
rationality as basis of human equality and, 15–16, 18 rationality and moral equality: continuity argument, 18, 158–159, 242–243 Rawls, J., 17–18 solutions, 16–20, 22 subjective perspective and, 242–243 Vartija, D., 223 virtue theory, 6, 85–86 Vlastos, G., 56 W Waldron, J., 9–11 basic equality, 6, 55, 56, 202, 305 ‘decision’ to treat others as equals, 227–228 democratization of aristocratic dignity, 52–53 gradability objection, 56 human dignity, 229–232 One Another’s Equals, 65–66, 69–73 opacity respect, 137–138 on racism, 95–96 range properties, 56, 137 religion and equality, 38, 48 Walsingham, T., 50–51 Walzer, M., 82 Williams, B., 2, 56–57, 186, 262, 276 Wolff, J., 8–9 Wolff, V., 269 women CEDAW, 105 Greece (classical), 41–42 inferiority/superiority relations, 41–42, 224, 248 Leveller women, 52–53 men and women created in the ‘image of God’, 38, 48–49 misogyny, 232–233, 307–308 Rome (classical), 47 wrongness, expressivist theory of, 92–94 Wynter, S., 221–222 Z Zimmerman, M., 181, 184