Reciprocal Freedom: Private Law and Public Right (Clarendon Law Lectures) 0198754183, 9780198754183

Citation preview

Reciprocal Freedom

c l are nd on l aw series War ANDREW CLAPHAM Land Law (3rd edition) ELIZABETH COOKE Competition Law and Antitrust DAVID J. GERBER Introduction to Company Law (3rd edition) PAUL DAVIES

Natural Law and Natural Rights (2nd edition) JOHN FINNIS Employment Law (2nd edition) HUGH COLLINS International Law VAUGHAN LOWE Civil Liberties CONOR GEARTY

The Conflict of Laws (4th edition) ADRIAN BRIGGS

Intellectual Property MICHAEL SPENCE

Bentham and the Common Law Tradition (2nd edition) GERALD POSTEMA

Policies and Perceptions of Insurance Law in the Twenty-​First Century (2nd edition) MALCOLM CLARKE

Law and Values in the European Union STEPHEN WEATHERILL

Philosophy of Private Law WILLIAM LUCY

Personal Property Law (4th edition) MICHAEL BRIDGE

Law in Modern Society DENIS GALLIGAN

The Idea of Arbitration JAN PAULSSON

An Introduction to Tort Law (2nd edition) TONY WEIR

The Anthropology of Law FERNANDA PIRIE Law and Gender JOANNE CONAGHAN The Concept of Law (3rd edition) H.L.A. HART With a Postscript edited by Penelope A. Bulloch and Joseph Raz With an Introduction and Notes by Leslie Green

Equity (2nd edition) SARAH WORTHINGTON Atiyah’s Introduction to the Law of Contract (6th edition) STEPHEN A. SMITH AND P. S. ATIYAH Unjust Enrichment (2nd edition) PETER BIRKS

Land Law (2nd Edition) ELIZABETH COOKE

An Introduction to Family Law (2nd edition) GILLIAN DOUGLAS

Administrative Law (5th edition) PETER CANE

Criminal Justice LUCIA ZEDNER

Discrimination Law (2nd edition) SANDRA FREDMAN

Contract Theory STEPHEN A. SMITH

An Introduction to the Law of Trusts (3rd edition) SIMON GARDNER

Reciprocal Freedom Private Law and Public Right E R N E ST J. W E I N R I B University Professor Emeritus, Faculty of Law, University of Toronto

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 © Ernest Weinrib 2022 The moral rights of the author have been asserted First Edition published in 2022 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Public sector information reproduced under Open Government Licence v3.0 (http://​www.natio​nala​rchi​ves.gov.uk/​doc/​open-​gov​ernm​ent-​lice​nce/​open-​gov​ernm​ent-​lice​nce.htm) 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: 2022940210 ISBN 978–​0–​19–​875418–​3 DOI: 10.1093/​oso/​9780198754183.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

The juris scientia, objectively understood, is in fact a science or systematic knowledge of the law; since law must be derived from reason, and thus from principles, this science has, indeed, not only a perception of the general outline of the principles of law as a whole, but also a knowledge of the individual parts (differing therein from the subjective knowledge of a vasta congeries of rules, which hang together only by way of their affinity) . . . N.B. The outline of the whole is needed, in the same way as was said of the painter: nulla dies sine linea, i.e., no day must go by without making a sketch of the entire human body. One must therefore be at pains to study the interconnection of all law, i.e., the possibility of the ground to its consequences. I Kant, Lectures on Ethics 27: 535 in Immanuel Kant, Lectures on Ethics (Cambridge University Press 1997) 295. Reproduced with permission of the Licensor through PLSclear.

Preface This book continues a decades-​long exploration of the theory of private law. The lynchpin of this theory has been the idea of corrective justice, the pedigree of which goes back to Aristotle’s account of justice. For Aristotle, corrective justice and distributive justice signified two different structures for one’s relationships with others. Distributive justice divides a benefit or burden among any number of persons in accordance with some criterion of distribution. Corrective justice, in contrast, corrects injustices within the bipolar transactions that make up what we now think of as private law. The first stage of this exploration on my part culminated in The Idea of Private Law.1 That book developed Aristotle’s comments on the bipolarity of corrective justice into an account of the distinctive notions of coherence and intelligibility that are in play in private law’s determinations of liability. Reflecting the truism that the liability of a particular defendant is always a liability to a particular plaintiff, corrective justice takes the relationship between the parties to be the central and pervasive feature of liability. Accordingly, the explication of this feature is the fundamental task of a theory of private law. Within this relationship between plaintiff and defendant, the position of one party is always conditioned by the position of the other. Liability being a bipolar phenomenon, the theoretical analysis of it can therefore not come to rest at either pole. Accordingly, it is a mistake to explain liability by reference to considerations separately relevant to either of the parties (or even to both of them). Rather, if private law is to be coherent, the relational character of the reasons supporting liability has to match the relational structure of liability itself. The Idea of Private Law was devoted to elucidating and connecting the components of the corrective justice approach. It articulated three such components: the legal formalism exhibited by the distinction between corrective and distributive justice; the correlativity of the parties’ positions within the bipolar structure of corrective justice; and the robust Kantian conception of legal rights that provides corrective justice with its content. Using illustrative material drawn principally from the law of torts, the book melded these into an integrated framework from which emerged private law’s distinctive conception of normative coherence. The next stage in the exploration of this theory of private law was to extend the analysis beyond tort law to other branches of private law, as well as to other contexts in which the corrective justice approach might be illuminating. This was the

1

EJ Weinrib, The Idea of Private Law (Harvard UP 1995, OUP 2012).

viii Preface task of Corrective Justice.2 That work presented a corrective justice analysis not only of the duty of care in negligence law, but also of remedies, gain-​based damages, contracts, unjust enrichment, and property. Apart from addressing these familiar substantive legal topics, the book also deployed the corrective justice approach in other settings, such as the Jewish law of unjust enrichment and the activity of legal education within the modern university. The goal was to exhibit in concrete detail and in a variety of contexts private law’s distinctive relational structure, its relational concepts, and its relational mode of reasoning. Whereas these two books largely focused on the internal structure and operation of private law within the corrective justice approach, the present book proceeds in a different direction. Although it starts with the structure of the private law relationship, it moves gradually outwards to consider the situation of private law within the legal order as a whole. The previous books were about private law in itself; this one, as it unfolds, places private law in the larger world of law. Accordingly, the last half of the book deals with the state’s role in forwarding distributive justice, with the horizontal application to private law of constitutional rights and values, and with the rule of law as an idea that governs every variety of state action, including both public law and private law. Anticipating this destination, the first half of the book introduces ideas that, while fully respecting the integrity of private law, prepare the ground for this outward movement. Thus, although it does not exhaust the theme, the book brings the theory of private law up to the border of public law, creating a vantage point from which to discern how private law fits within the broader legal landscape. In exploring these issues, the book explicitly takes Kant’s legal philosophy as its guide. The book’s starting point in corrective justice postulates a content that matches the correlativity of the parties’ situations in that form of justice. That content is found in Kant’s conception of rights and their correlative obligations. Of course, an emphasis on rights is not unique to Kant. But what makes Kant’s account of law particularly apposite to the relational character of corrective justice is that, perhaps more than any legal philosopher, he treats law as unremittingly relational. For Kant, this attention to the relational does not emerge merely as the consequence of rights that might have non-​relational grounds; rather, it informs the very grounding of the rights themselves. Just as Aristotle characterized justice as directed ‘towards another’3 and specified corrective and distributive justice as different ways in which this other-​directedness can be structured, so Kant takes legal philosophy to be concerned not with the normative status of a person’s conduct considered on its own, but with the relationship of one person’s action to another person’s freedom. Moreover, Kant’s work presents a comprehensive picture of the different kinds of legal relationship, whether between persons and other persons,

2

EJ Weinrib, Corrective Justice (OUP 2012).

3 Aristotle, Nicomachean Ethics V, 1129b27, 1130a4, 1130a13.

Preface  ix or between persons and their own states, or between one state and other states, or between persons of one state and a different state. These he connects to one another and to the normative ideas that underlie (and in his view must underlie) them. His account thus provides a normative map that describes the location of different legal configurations relative to one another and that explicates the single set of interconnected normative ideas that nourishes them all. He thereby both preserves the distinction between what we term private law and public law, and yet outlines the conceptual moves that connect the former to the latter. The central insight of Kant’s account is that law is necessary for the actualization of reciprocal freedom, that is, for the existence of a totality of social relations in which each person’s freedom co-​exists with everyone else’s. The rights to personal integrity, property, and contractual performance—​which Kant groups under the term ‘private right’ as legal categories conceivable in a state of nature—​as well as the institutional apparatus of public and coercively enforced norms in a civil condition—​Kant’s ‘public right’—​are integral to this freedom. The notion of public right is crucial both to the existence of a system of private law and to the transition from private law to what lies beyond. The present book retraces this movement in the context of contemporary legal and jurisprudential issues. The argument proceeds in a sequence of stages, each of which is complemented by, and presupposed in, the one that follows. After beginning with the correlative structure of the private law relationship (Chapter 1), the argument moves to the Kantian notion of rights and their correlative obligations. As markers of reciprocal freedom, these rights and obligations make up the content of private law’s correlatively structured relationships. The Kantian conception of a right normatively unifies, with respect to the object of the right, the different categories of jural entitlement subsequently articulated by Hohfeld (Chapter 2). Illustrating this connection between Kant and Hohfeld is the right to ownership, the Kantian justification for which provides the ground for bringing together the Hohfeldian liberty to use and the Hohfeldian claim-​right to exclusivity. When conceived in Kantian terms, ownership in turn necessitates the systematicity and the public institutions that are present in a condition of public right (Chapter 3). Public right then, in its turn, affects what reciprocal freedom requires. It elicits adjustments in the operation of the concepts of private right to reflect their public character in a functioning system of law (Chapter 4). It insists on the legislative maintenance of the independence of persons through distributive justice (Chapter 5). It authorizes the consideration of constitutional values in private law controversies (Chapters 6 and 7). And finally, through the application of the rule of law, it secures the constitutive aspects of a polity committed to the actualization of reciprocal freedom (Chapter 8).4 4 The topics of these last four chapters are salient in contemporary discussions of law, but they do not, of course, even come close to forming a complete catalogue of the circumstances in which the theoretical connection between private law and public law might be profitably explored.

x Preface Throughout these explorations of the theory of private law, from the first book to the present one, four ideas have consistently been in play. Of these, the first pair deals with the structure and content of the private law relationship, and the second pair with the nature and limits of private law theory. The first idea is that fair and coherent reasons for liability are correlative in structure in that they treat each party’s position as the mirror image of the other’s. Correlatively structured reasons focus not on either party separately from the other but on the relationship between them as doer and sufferer of the same injustice. Such reasons are fair to both, because they treat the parties as equals within the relationship; considerations relevant to only one of them do not determine the legal consequences for both. Such reasons are also coherent because they reflect the parties’ relationship as such, rather than referring to a hodge-​podge of factors (such as the defendant’s deep pocket or the plaintiff ’s need) that apply to each party separately. Consequently, arguments that seek to have private law achieve goals external to the parties’ relationship—​whether utilitarian, distributive, or economic—​are all structurally inconsistent with fair and coherent determinations of liability. In contrast to such goal-​oriented arguments, correlatively structured reasons treat the parties as participants in a legal relationship organized by the principle of its own internal fairness and coherence. The second idea is that rights and their correlative obligations provide the content for private law’s correlatively structured reasoning. By their very nature right and obligation are correlative concepts. Every private law right implies that others are under an obligation not to infringe it; similarly, in private law, no obligation stands free of its corresponding right. Presupposed in the rights and obligations of private law is the conception of the person as a free being who has the capacity to set his or her own purposes. In light of this conception of the person, rights and their correlative obligations function as the juridical markers of the equal and reciprocal freedom of the parties in relation to each other. The third idea is that the activity of theorizing about private law involves not the construction of a utopia but the understanding of an ongoing normative practice. In the most highly developed versions of this practice, those entrusted with authority over its elaboration have striven, of course not always with success and never without dispute, to work out the fair and coherent terms on which persons ought to interact with each other. The theory of private law takes this material as its starting point and enquires into its structure, its presuppositions, and the internal connections among its most pervasive features. The aim is to identify the most abstract unifying conceptions implicit in the law’s doctrinal and institutional arrangements and to enquire into the rationality that inheres in the law’s processes. In this effort, the contemporary theorist need not start from scratch. One may avail oneself of the history of philosophical reflection, whose leading figures provide exemplars for one’s own efforts, as Kant put it, to ‘exercise the talent of reason’.5

5

I Kant, Critique of Pure Reason (P Guyer and A Wood trs, CUP 1998) B866.

Preface  xi These figures may point the contemporary theorist of private law in the direction of certain ideas whose structure they have presented with extraordinary clarity and whose implications they have explored with extraordinary profundity. For example, the first two ideas that I mentioned above—​the significance of correlativity as a structural feature of reasoning about liability and the role of rights in providing the content of correlatively structured reasoning—​are drawn from Aristotle and Kant, respectively. Aristotle attached the term ‘corrective justice’ to the operations of law that are structured by the correlativity of the parties’ positions as doer and sufferer of the same injustice. Kant was perhaps the greatest expositor of the systemic significance of rights as expressions of human freedom. My own work has been devoted to the fairly modest objective of demonstrating the significance of these previously ignored Aristotelian and Kantian ideas for understanding the structure and content of private law. Although I think that it is important to be explicit about the relationship between one’s own work in legal theory and the great figures of our philosophical tradition, my purpose in invoking Aristotle and (in the present work) especially Kant as extensively as I do has been quite circumscribed. I have not aimed to reconstruct the place of law within an Aristotelian conception of ethics or within a Kantian metaphysics of practical reason. The point is not, as has sometimes been mistakenly supposed, to present private law or any branch of it as giving effect to a comprehensive moral philosophy. Rather, the task of legal theory, as I see it, is to bring to the surface the most pervasive ideas latent in law as a normative practice. The greatest thinkers are relevant to this conception of legal theory only because, and to the extent that, they provide insights helpful to the understanding of law in its own terms. In this book, I continue along these lines. Its principal difference from previous books is the attention it pays to the notion of public right that Kant deployed to illuminate the relationship between legal norms and legal institutions. My intention is to draw out the implications of this notion both for the theory of private law and for the connection between private law and other aspects of the legal order. This brings me to the fourth idea. The account that I offer is subject to the inevitable limitations on the scope of any theoretical account of legal norms. The theorist is not a philosopher-​king in academic robes who can work theoretical abstractions into a complete, definitive, and determinate code of law. Rather, a theory of private law is concerned with the conceptual structure and the normative presuppositions of the phenomenon of liability. Its function is to orient us in the conceptual space of the possible reasons for liability by identifying the kinds of reasons that are properly available and by showing how reasons of those kinds can come together in a fair and coherent system of liability.6 Theoretical reflection, however, cannot

6 This is an adaptation of Rawls’ formulation of the role of orientation in political philosophy; see J Rawls, Justice as Fairness: A Restatement (Harvard UP 2001) 3.

xii Preface supplant the activity of lawyers in specifying the full range of legal norms or in applying them to particular cases. Different legal systems organize themselves differently and have different histories and different mechanisms of decision. The diversity of their legal materials expresses the diverse ways in which the different legal systems strive for fairness and coherence. Accordingly, every sophisticated legal culture has a body of legal knowledge that is specific to it, as well as its specific techniques for applying and developing the law. It also has lawyers who are versed in this knowledge and skilled in these techniques. In carrying out these activities, lawyers are not theorists. Nor do whatever theoretical insights theorists have qualify them to act as lawyers. The conceptual space within which theory orients us cannot, itself, be expected to supply all the specific norms required to fill that space. Indeed, this book’s emphasis on the relation between the abstract representations of norms and their determinations by the positive law shows why that must be so. Haunting the preface to every book is the question asked by the ancient historian Livy in the dactylic opening of the preface to his great history of Rome: ‘Facturusne operae pretium sim . . .’7 Is this work going to be worthwhile? Livy’s own answer was, ‘I don’t really know, and if I knew I wouldn’t dare say’. In this Preface I have tried briefly to indicate the book’s background, scope, theme, and animating ideas. The aspiration of the book is to exhibit reciprocal freedom as the normative idea pertinent to the legal order as a public and authoritative system in which private law occupies a distinctive place. To this end, the book presents an abstract, sequenced, and doctrinally informed argument along Kantian lines for understanding law as necessary to our co-​existence as free beings. With reference to this book, the answer to Livy’s question ultimately rests on whether the book’s execution is worthy of this august aspiration. Of course, to make that determination is a matter for the judgment of each person who decides to read it.



7

Titus Livius, Ab Urbe Condita, praefatio.

Acknowledgements This book began its life as the Clarendon Law Lectures delivered at Oxford University in November 2014 under the title ‘The Jurisprudence of Corrective Justice’. The first two chapters are expansions of the first two of those lectures. The third lecture, which is not represented in this book, was the basis of the article ‘Causal Uncertainty’ (2016) 35 Oxford J of Legal Studies 135. I am grateful to Oxford’s Faculty of Law, to the then dean Timothy Endicott, and to the Faculty’s members for their hospitality and their critical attention during my stay in Oxford. Although everything in this book was written with the book in mind, earlier versions of three segments of it have appeared elsewhere: (i) ‘The Normative Structuralism of Corrective Justice’ in Hanoch Dagan and Benjamin C Zipursky (eds), Research Handbook on Private Law Theory (Edward Elgar Publishing 2020) 484–​98, reproduced with permission of the Licensor through PLSclear; ‘Ownership, Use, and Exclusivity: The Kantian Approach’ (2018) 31 Ratio Juris 123 (Blackwell Publishing); ‘Private Law and Public Right’ (2011) 61 U Toronto LJ 191, used with permission of the University of Toronto Press (https://​utpj​ourn​als. press), DOI: 10.3138/​utlj.61.2.191-​211. © University of Toronto Press 2011. While writing this book I presented parts of it in lectures, workshops, and seminars at Oxford University, the University of Toronto, Fordham University, and especially at Tel Aviv University, to which for a number of years I annually made happy and productive visits in the course of a valued university appointment. I very much appreciate the comments I received from scholars at those institutions and others, as well as the friendship, collegiality, and intellectual rigour to which these comments attested. I also am grateful to Gerard (Kees) Westland for his research assistance on German legal material. As I explain in the Preface, this book makes a protracted foray from the domain of private law into that of public law. This would not have been possible without the learning, encouragement, and support of two public law scholars who were close at hand, and willing to share their ideas and to engage with mine repeatedly and at a moment’s notice: Lorraine Weinrib, professor of constitutional law at the University of Toronto, and Jacob Weinrib, professor of constitutional and administrative law at Queen’s University. Although they should not be taken to agree with everything in this book, their influence on it has been enormous, although miniscule in comparison to their other contributions to my life.

Contents 1. Structure 

1

2. Rights 

26

3. Ownership 

47

4. Public Right 

67

5. Distributive Justice 

94

6. Horizontality: Presuppositions and Functions 

117

7. Horizontality: Scope and Operation 

149

8. The Rule of Law 

181

Index 

211

1

Structure 1.  Corrective Justice as a Structural Idea Over the last century or so, the prevailing orthodoxy in the common law world, once one went beyond the sheer analysis of doctrine, has featured two assumptions. The first is instrumentalist: legal doctrine is justifiable to the extent that it promotes desirable goals (such as deterrence, compensation, loss-​spreading, economic efficiency, and commercial utility) that can be specified independently of the law. Under this first assumption, law is simply the tool through which these independently specifiable goals are, or at least ought to be, realized. The second assumption is pluralist: legal doctrine reflects a number of different and even irreconcilable goals, the costs and benefits of which are implicitly or explicitly traded off within the reasoning of a sagacious judge. Many of the large theoretical questions that arose during this period, at least in North America, took this orthodoxy as its starting point but inquired into sources of tension within it. One question, of central concern in the legal process approach,1 dealt with the proper role of the judiciary and with the differentiation of the judicial and legislative function: If, in the final analysis, judges selected and balanced the goals that are to be inscribed into the law, how could their activity legitimately be distinguished from that of the legislators? A second question was whether the instrumentalism could be presented under a single goal so uniform and comprehensive as to oust or restrict the pluralism. This was the project pursued by the brilliant progenitors of the law and economics movement, who attempted in their different ways to establish efficiency as the dominant goal, thereby eliminating or marginalizing other goals.2 A third question was whether the pluralism was so unbounded and omnipresent, and legal doctrine consequently so indeterminate, that the standard conception of legal ordering had to be rethought. This was the direction taken by proponents of Critical Legal Studies.3 In contrast to these explorations within the prevailing orthodoxy, the corrective justice approach to private law challenges the orthodoxy itself. By private law, I roughly mean the ensemble of normative ideas (rights, obligations, powers, and 1 HM Hart and AM Sacks, The Legal Process: Basic Problems in the Making and Application of Law (The Foundation Press 1994). 2 G Calabresi, The Cost of Accidents (Yale UP 1970); R Posner, Economic Analysis of Law (Harvard UP 1972, now in its ninth edition). 3 D Kairys (ed), The Politics of Law: A Progressive Critique (Basic Books 1982).

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0001

2  Ernest J. Weinrib so on) and institutional arrangements (authoritative procedures and remedies) that legally link two persons—​at the litigation stage, the ‘plaintiff ’ and the ‘defendant’—​ to each other. On the corrective justice view the focus of corrective justice is on the inner structure of this relationship between plaintiff and defendant. The normative considerations relevant to this relationship emerge through consideration of the structure and presuppositions of the relationship itself. Corrective justice therefore has no room for, and indeed denies the relevance of, the independently desirable goals postulated by instrumentalist analysis. Moreover, corrective justice looks on the relationship as an internally unified whole, in which the parties occupy correlative positions as doer and sufferer of the same injustice. The unity of their relationship is evident in the litigation that links them as plaintiff and defendant, in the doctrines that relate the parties directly to each other, and in the remedy that applies simultaneously to both of them. The idea that the parties’ relationship is an internally integrated normative phenomenon is inconsistent with the pluralism of the prevailing orthodoxy. Corrective justice thus rejects both of the orthodoxy’s assumptions. The basic idea of corrective justice is that the reasons for holding the defendant liable to the plaintiff should match the normative structure implicit in the very notion of liability. In private law, liability is a correlatively structured phenomenon: the liability of a particular defendant is always a liability to a particular plaintiff. Accordingly, liability so relates the parties that their legal positions mirror each other’s. Corrective justice views the correlativity of the parties’ positions not merely as a coincidental outcome of a particular legal process, but as the structural characteristic that informs, or at least should inform, the reasoning that leads to that outcome. For if the reasoning did not embrace both parties as correlatively situated, the correlative operation of liability, that is, the holding of this particular defendant liable to this particular plaintiff, would itself be problematic. Private law could then hardly maintain its categorical insistence that the damages or the specific relief to which the plaintiff claims to be entitled should come from the defendant. Correlativity, then, is the most pervasive and general structural feature of private law. It is present in the legal reasoning of private law, in the concepts that inform that reasoning, in the institutional framework of litigation, in the liability that links plaintiff to defendant, and in the determination of the remedy. Corrective justice is the theory of this correlativity: of its normative presuppositions and implications, of its relation to the adjudicative context, and of the terms on which it figures in the various grounds of liability. What renders this form of justice ‘corrective’ is that it first ascribes to liability the function of correcting an injustice and then takes the correlativity of the parties’ positions within this correction to be the structure implicit in the reasons for the correction. The theoretical ambition of corrective justice is to exhibit the unity that underlies private law. This unity embraces the legal reasoning, the conceptual framework, the institutional operation, and the remedial consequences of liability. The

Structure  3 unity is also the locus of a distinct kind of normativity that reflects not the desirability of independently justifiable goals but normative considerations internal to the very idea of correlatively structured legal relationships. Hence the intrinsically normative unity of corrective justice stands in contrast to the instrumental pluralism of the prevailing orthodoxy. Corrective justice has been much discussed in contemporary legal scholarship, often in different senses. The view of corrective justice that I have just sketched differs from the conceptions of corrective justice put forward by others. One such conception, advanced by John Gardner, holds that corrective justice has remedial significance only: it refers solely to the operation that corrects by allocating goods back from the defendant to the plaintiff.4 In contrast, the corrective justice to which I have been referring sees the correlative operation of the remedy as an indication of the correlative structure of the reasoning that properly animates determinations of liability. What renders this form of justice ‘corrective’ is the connection between the function of liability in correcting an injustice between the parties, and the role of correlativity in structuring the legal reasoning appropriate to this function. Accordingly, corrective justice as I present it is not about the bipolar correction alone, but about the reasons for liability that make such corrections normatively intelligible. This reasoning necessarily has reference to the normative character of the relationship between the parties antecedent to the litigation between them. Similarly, consider Jules Coleman’s suggestion that the corrective justice is the principle under which an individual has a duty to repair the wrongful losses that his conduct causes.5 Coleman regards corrective justice not as the idea that structures the entire phenomenon of liability but as a substantive principle that imposes a particular duty, that of repairing wrongful losses. Other occasions for liability, where losses are not wrongful (such as cases of strict liability in tort law) or are not at issue in the plaintiff ’s claim (as with unjust enrichment) fall beyond the principle’s reach. This view of corrective justice as a substantive principle about the repair of wrongful losses has little in common with the view that corrective justice deals with the correlatively structured character of liability as such. Despite the currency of these (and other) diverse conceptions of corrective justice, the question of which is the true conception (‘what is corrective justice really?’) is not important. What matters is the distinctive complex of ideas to which the each of these conceptions refers. These ideas should be assessed on their merits, that is, on the degree to which they illuminate the phenomenon of liability (and the interpersonal relationship that liability presupposes), rather than on their conformity to what one imagines the words ‘corrective justice’ mean. In any case, it is certainly a mistake to suppose that those who gather under the banners of corrective justice march to the same beat or in the same direction. For not only do

4 5

J Gardner, ‘What is Tort Law For? Part I: The Place of Corrective Justice’ (2011) 30 L and Phil 1. J Coleman, Risks and Wrongs (CUP 1992) 325.

4  Ernest J. Weinrib ‘corrective justice theorists’ widely diverge on fundamental issues, but also some of the most vigorous critics of ‘corrective justice theory’ themselves hold views that are barely distinguishable, if distinguishable at all, from particular versions of corrective justice.6 In comparison with other views, the distinctiveness of corrective justice as I understand it (to which I will henceforth refer simply as ‘corrective justice’) lies in this: Corrective justice addresses the structural character of liability. It does not confine itself to particular features of liability (as does Gardner by equating corrective justice with the law’s remedial operation) or to particular instances of liability (as does Coleman by restricting corrective justice to the repair of wrongful losses). Accordingly, my approach to corrective justice proceeds on a broader front than do other conceptions. It holds out the promise of a comprehensive unifying theory for private law that orients us within the conceptual space of all the possible justifications for holding someone liable. That space, structured as it is by the correlative normative positions of the two parties, allows us to identify the kind of justifications appropriate to determinations of liability and to explore the extent to which such justifications animate the private law of particular jurisdictions.

2.  Structure as Abstraction Structure signifies the most abstract representation of what is distinctive about the way that private law normatively situates the parties in relation to each other. Constitutive of the parties’ relationship are doctrinal ideas and institutional arrangements—​factual findings specific to the case, rules, standards, concepts, bases of liability, adjudicative procedures, a finding of liability, and so on—​that have a determinate content. For the doctrinal ideas, the content consists in the actions that the ideas permit, require or forbid, and in the particular legal categories governing those actions; for the institutional arrangements, the content consists in the legal operations that they authorize in the relevant legal bodies. One arrives at the structure of the private law relationship by abstracting from this normative and institutional content to the form of the relationship, that is, to the relationship’s being made up of two parties who are correlatively situated with respect to the justness or otherwise of their interaction with each other. Because it abstracts from all the doctrinal ideas and institutional arrangements of private law, correlativity is the barest and most comprehensive abstraction that is distinctive of the private law relationship. In this sense correlativity exhibits the structure pervasively embodied

6 E Weinrib, ‘Civil Recourse and Corrective Justice’ (2011) 39 Fl St ULR 273. In JCP Goldberg and BC Zipursky, Recognizing Wrongs (Belknap Press 2020) 358, the progenitors of civil recourse theory entertain the possibility that there may be ‘little space’ between their view and mine, and that civil recourse may merely be a better label for the phenomena we are all dealing with.

Structure  5 in the content of private law; the various doctrinal and institutional repositories of this content are the specific determinations of this abstracted structure. This process abstracts from the legal relationship’s content, not from its normative character. The relationship’s normative character remains as present to the structure as it does to the doctrinal ideas and institutional arrangements from which it is abstracted. Indeed, the point of so abstracting is to provide a perspicuous representation of the completely relational normativity that informs private law, thereby highlighting the difference between private law and other normative enterprises. The doctrinal ideas and institutional arrangements of private law are normative in that they specify (or provide the mechanisms for specifying) the terms on which persons must interact with each other. As the abstraction from that content, the structure is normative in that displays the relational framework to which the content gives specific expression. The same normativity is operative along the whole spectrum from specific to abstract, holding the parties to corrective justice, that is, to the form of justice applicable to their being correlatively situated in relation to each other. In abstracting to correlativity as the structural idea underlying the private law relationship, corrective justice merely extends the tendency to abstraction that marks the activity of thinking within private law. Although the events that give rise to a private law relationship are particular—​John Doe did such and such to Richard Roe—​lawyers understand these events in terms (such as tort law’s notions of cause, duty of care, and fault) that abstract from the particularity of the occurrence. These abstractions, in turn, are embedded in further and further abstractions, such as the cause of action generally and the components of the broader taxonomy of the various kinds of liability. Private law is thus replete with abstractions (and their corresponding determinations) that operate at different levels of generality and particularity. The determination of a more comprehensive abstraction may itself be an abstraction that has its own more particular determinations. For example, each basis of liability has a theme that summarily describes its specific subject matter (consensual agreement in the case of contract, wrongful risk-​creation in the case of negligence, and so on) and that differentiates it from other subject matters; and each basis of liability has its specific set of conceptual components that renders the basis of liability more determinate. Thus, the various bases of liability are all determinations that thematize the working of correlativity in the different kinds of private law relationship; yet these bases of liability are themselves categories that abstract from conceptual components that are still more particular. The interplay of abstraction and determination in private law takes two different forms, one of subordination and the other of coordination. The first form is that a determination may instantiate the abstraction under which it falls and to which it is accordingly subordinate. The law is familiar with this in contexts where the issue is whether a set of particular events does or does not qualify as an instance of a legal category. Thus, for example, the fact that some act of John Doe created

6  Ernest J. Weinrib an unreasonable risk to Richard Roe is a circumstance specific to their particular relationship. The legal significance of that act derives from the requirement of negligence law that one must act with reasonable care, a requirement that abstracts from any particular act that instantiates it and has a place of its own within negligence law. The second form of interplay is that the determination is one of a set, the members of which are coordinated within the larger unity that they together comprise. This is the case with determinations that are themselves abstractions within a more inclusive abstraction. One context for this is in the aligning of the different conceptual components of a single cause of action. To continue with our example of the negligent John Doe: the question of whether he exposed Richard Roe to unreasonable risk is part of a more general enquiry into liability for negligence that includes reference to the concepts of duty, proximate cause and factual cause. The totality formed by all these concepts defines the basis of negligence liability in a general way. Thus, while we can isolate the particular issue of reasonable care for purposes of an ordered legal analysis, the treatment of this issue has to be conjoined as well with the other negligence concepts. Negligence as a basis of liability is an abstraction that is determined by the interaction of the less inclusive abstractions of which it is composed. The same form of interplay arises at an even more comprehensive level whenever the issue of concurrent or overlapping liability arises, for example, between tort law and contract law, or between contract law and unjust enrichment. Then particular bases of liability are related to the general doctrinal totality of private law. A still more comprehensive interplay of this sort arises when doctrine is linked to legal institutions, especially to the institution of adjudication. This interplay aligns legal doctrine with judicial competence and with the limitations inherent in the judicial process, so as to yield a system of liability that operates in an institutionally appropriate manner. In this interplay of abstraction and determination, the determinations specify the abstractions, and the abstractions regulate the determinations. On the one hand, the determination that John Doe’s act created an unreasonable risk to Richard Roe specifies what the requirement of reasonable care means for the particular relationship between Doe and Roe. On the other hand, the idea of reasonable care is regulative for determining whether Doe’s act instantiates it, for in making that determination one must consider what the law regards as the standard and the components of reasonable care. Thus, determining whether Doe’s act was an unreasonable risk requires an act of judgment that fuses the particularity of Doe’s conduct and with generality of the legal category of reasonable care. Similarly, the requirement of reasonable care is itself a determination of one of the conceptual components of negligence as a basis of liability. That basis of liability is an abstraction that, through the entire ensemble of requirements, regulates the conditions under which unreasonable care leads to liability. Consequently, the only unreasonable risks that figure in the law of negligence are those that materialize

Structure  7 in injury (thereby satisfying the requirement of factual causation) and those that create the danger of the very kind of injuries to the very class of persons that leads us to regard the risk as unreasonable to begin with (thereby satisfying the requirements of proximate cause and duty). The totality of all the concepts that form the basis of liability in negligence is, accordingly, regulative of the scope of each of the concepts taken on its own. That totality in turn is ultimately regulated by the structural idea of correlativity. The relationship between the regulative role of abstraction and the specificatory role of determination is dynamic and interdependent. On the one hand, the abstraction is not operational in the absence of a mechanism for specifying what it means in the quotidian interactions of social life. The positive law authoritatively supplies this mechanism by formulating, or creating the machinery that formulates, the needed determinations. It may decide, for example, that John Doe really did create an unreasonable risk, or that reasonable risk is a function of the probability and gravity of a prospective injury, or that the idea of reasonable care has to be coordinated with the concepts of factual causation, duty and proximate cause. On the other hand, the determination must plausibly be attuned to the regulative character of the abstraction under which it falls (otherwise it would not specify the meaning of that particular abstraction), and can be faulted if it fails to do so. Thus, the judgment that John Doe created an unreasonable risk can be more or less felicitous; and the coordination of concepts within a given legal system can be more or less adequately achieved. Correlativity, then, is the abstraction distinctive to private law that lies at the apex of a system of nested abstractions and determinations. One can imagine this system as a juridical version of Jacob’s ladder, ‘set earthward with its head reaching into the heavens, and the angels of the Lord ascend and descend on it’.7 Each rung of this ladder both abstracts from the determinations below it and provides the determination for the rung above it. The ascent up the ladder is from the more particular to the more inclusive. The most inclusive abstraction of all is correlativity, because it abstracts from the content of all the other abstractions of private law. And like the angels who first ascend and then descend, corrective justice works its way up from the particular normative ideas and institutional arrangements of private law to the abstract representation of its structure, and then works back down to hold the particulars true to that structure. Now although the focus of corrective justice on correlativity extends the law’s own tendency towards abstraction, correlativity differs from the abstractions of private law in one significant respect. Those abstractions comprise the content of private law at different levels of generality. The focus on correlativity, in contrast, involves an abstraction from all of the content of private law to the form of the private 7 Genesis 28.12; M Maimonides, The Guide of the Perplexed, vol 1 1, 15 (S Pines tr, University of Chicago Press 1963) 41.

8  Ernest J. Weinrib law relationship. Whereas reasonable care or negligence liability or tort law, for example, are abstractions that explicitly figure within the law, the correlativity that preoccupies corrective justice is a theoretical idea that refers to the structure underlying the legal relationships marked by those abstractions. What counts as the determination of correlativity is the private law itself, which specifies for the legal relationships of a given jurisdiction what the correlativity of the parties’ normative positions means and how it works. This determination is subject to correlativity’s regulative function. This means that the content of the private law relationship is subject to the requirements implicit in the relationship’s structure. Whether, and with what degree of adequacy, a jurisdiction’s private law in fact conforms to those requirements is, of course, another matter. Corrective justice, then, is nothing but correlativity considered as the supreme regulative principle for private law. This process of abstraction is juridical, not ontological. The point of abstracting from the content of private law is not to postulate an independent reality, like the ‘brooding omnipresence in the sky’ that Oliver Wendell Holmes famously disparaged.8 Correlativity is no more an independent reality than is proximate cause or the duty of care or unjust enrichment or any of the other abstractions that populate the domain of private law. Nor does it make sense to criticize corrective justice on the grounds that it construes the plaintiff so abstractly that he or she can never have need of compensation.9 Rather, corrective justice aims to exhibit the most general standpoint from which to consider the normative character of the reasoning appropriate to private law. It operates on the parties here and now by viewing their controversy as subject to the set of abstractions that normatively connects the parties to each other. The process of abstraction works back from the content of private law to the structure that is most comprehensively implicit within it, in order then to hold the law’s content true to that structure. The value of invoking this abstraction is that its very sparseness affords an uncluttered view of the normative framework into which private law reasoning fits or should fit. In particular, this structure suggests certain notions of fairness and coherence. To these I now turn.

3.  Fairness and Coherence For corrective justice, then, the correlative structure of liability indicates the correlative structure of the reasoning that supports a finding of liability. Conformity to corrective justice requires this reasoning to reflect the unity of the parties’ relationship as doer and sufferer of the same injustice. The injustice that liability corrects is understood not as a brute event but as a normative ascription that arises through a process of reasoning. The components of that process of reasoning should apply

8 9

Southern Pacific v Jensen 244 US 205, 222 (1916). W Lucy, Philosophy of Private Law (OUP 2007) 298: ‘Do abstract agents need real compensation?’

Structure  9 equally both to the injustice of what the defendant has done and to the injustice of what the plaintiff has suffered, because that injustice is the same from both sides. Two implications of this are worth noting. The first implication is that corrective justice embodies a conception of fairness as between the parties. By holding the reasons for liability to the structure of correlativity, corrective justice always focusses on what is normatively relevant to the relationship as such, rather than to either of the parties taken individually. Accordingly, no consideration is admissible that is unilaterally applicable, for good or ill, to one of the parties. Thus, neither party has a preferential or disadvantageous position in setting the terms of their interaction. Within their relationship they are normative equals; the reasoning applicable to their controversy refers only to considerations that are as relevant to the defendant as the doer of the supposed injustice as they are to the plaintiff as its sufferer. Neither of them can, therefore, be subjected to reasoning that makes liability an arrangement that sacrifices either of them to the interests of the other. The fairness of corrective justice, accordingly, consists in the law’s treating the parties in accordance with reasoning that reflects their transactional equality. This conception of fairness arises out of structural nature of corrective justice. It involves no independent substantive judgment about whether a given determination of liability is fair; rather, the fair terms of the parties’ interaction emerge out of reasoning that treats the parties as correlatively situated. Misplaced are intuitive judgments about fairness or judgments about fairness in the overall distribution of resources in society, the former because they reflect no context of reasoning, the latter because they reflect the wrong context. The fairness indigenous to private law is that which pertains to the relationship between the parties. Because that relationship is structured by the correlativity of the parties’ normative positions, the appropriate conception of fairness is one that reflects reasoning that accords with that correlativity. Such reasoning features, in its own distinctive way, the equality rightly associated with the notion of fairness. The second implication is that corrective justice presents a conception of coherence in legal argument. Because of its focus on correlativity, corrective justice treats the relationship as a unity in which each party’s normative position is reciprocally intertwined with the other’s. The legal reasoning that applies to the relationship must partake of and reflect this unity. Formulated negatively, this means that liability cannot be based on a multiplicity of mutually independent considerations—​ hence the rejection of pluralism that I noted at the outset. Formulated positively, the reasoning that supports liability should have a justificatory force that occupies and is coterminous with the entirety of the relationship’s normative space. For example, tort liability cannot be justified by the combination of deterrence and compensation; given their independence of each other (neither deterrence nor compensation is a normative idea whose justificatory force entails the other), their combination would not reflect the unified character of the relationship. Moreover,

10  Ernest J. Weinrib deterrence and compensation, each taken on its own, do not cover the entirety of the relationship, but rather are unilaterally relevant to one but not the other of the parties. And taken together, their justificatory force is not coterminous with the relationship’s normative space but extends to all whose conduct should be deterred and whose injuries need to be compensated, thereby providing no reason to link any particular defendant to any particular plaintiff. When the justificatory force of reasoning about liability is coterminous with the parties’ bipolar relationship, the reasoning truly functions as reasoning. Reasoning should have authority with respect to the material to which it applies; the point of engaging in reasoning is to allow that authority to govern whatever falls within its scope. This occurs under the corrective justice conception of coherence, because there the reasoning about liability occupies the entire space to which its justificatory force entitles it. Coherence is thus the internally regulative principle for reasoning that takes the justification of liability seriously. Conversely, when liability is the product of mutually independent considerations (as, for instance, when tort liability is seen as the combined product of deterrence and compensation), the scope of each is artificially truncated by the others. The result is that they do not truly function as the justification for linking this defendant to this plaintiff. Now one might wonder how this coherence is possible. The apparent difficulty is that, although coherence signifies a unity of the applicable reasoning, this unity cannot be achieved by a single monolithic reason. This is because the unity in question is not that of a single monolithic entity but of a sequence that is internally differentiated into what one of the parties has done and what the other has suffered. These may well be separate events, as the example of negligence liability indicates. Because the negligent act and the negligently suffered damage are not identical, how can the reasoning applicable to them be identical? What overcomes this difficulty is the conceptual structure of the various grounds of liability. Any particular ground of liability is fashioned out of a series of components. A coherently structured ground of liability coordinates these components into an ensemble of interlocking and reciprocally referential elements. Components that are coordinated in this way constitute an integrated set that, taken as a whole, assembles what the defendant has done and what the plaintiff has suffered into the single injustice that is the same for both parties. The normative positions of the plaintiff and the defendant are thereby melded into a unified normative unit that situates the parties correlatively. Take negligence law as an example. The components of the negligence enquiry translate into normative terms the progression from the defendant’s creation of an unreasonable risk to the materialization of that risk in injury to the plaintiff. One segment of the reasoning focusses on the act by the defendant that creates an unreasonable risk, and another segment on the damage that this act causes

Structure  11 the plaintiff. The normative space between these two segments is bridged by the concepts of duty and proximate cause, so that the unreasonableness of the risk created by the defendant lies in its potential to cause to persons in the plaintiff ’s situation the very sort of damage that occurred. The result is that the risk that materializes into injury to the plaintiff is the same as the risk that the defendant unreasonably created, and the wrong done in its creation is the same as the wrong suffered in its materialization. Rather than being independent, every component of this sequence conditions and is conditioned by every other. The justification for holding the defendant liable lies in the satisfaction of an integrated series of requirements that singles out and links the defendant’s risk creation to the plaintiff ’s injury. Thus, the constituents of liability interlock in an articulated unity that renders the injustice of what the defendant does the same as the injustice of what the plaintiff suffers. Coherence, then, lies in the correspondence of legal reasoning to the structure of liability. The justificatory force of such reasoning occupies and is coterminous with the relationship’s normative space, connecting the parties through a set of interlocking components that discloses the single injustice that the defendant did, and that the plaintiff suffered. Fairness and coherence are normative conceptions internal to corrective justice, bringing out what it means for the parties to be juridically related in terms of the correlativity of their normative positions. Indeed, fairness and coherence merely highlight different aspects of the normativity inherent in corrective justice as a structural idea. Fairness points to the equal standing of the two parties within the correlatively structured reasoning that sets the terms of their interaction; coherence is the condition under which the process of reasoning about liability can be adequate to the parties’ correlatively structured relationship. Consequently, an inconsistency with one of these normative conceptions is also an inconsistency with the other: any unilateral consideration is incoherent because it fragments the unity of the relationship, and any incoherence unfairly favours one of the parties by failing to view them as correlatively situated. These normative conceptions are internal to corrective justice, in the sense that they are not brought to bear on the relationship from a standpoint outside it. Nor are they the products of a general moral theory of fairness or coherence pertinent to the full range of normative phenomena and then applied to the specific case of private law. Rather, the structure of the parties’ relationship itself imposes limits on the sort of reasoning relevant within the relationship and determines the normative conceptions that hold good from within its own standpoint. Through its normative conceptions of fairness and coherence, corrective justice provides a proper role for adjudication. One may well ask: on what terms does the court, as a politically unaccountable public institution, appropriately render coercively enforceable judgments? The answer given by corrective justice is that

12  Ernest J. Weinrib courts have the normative jurisdiction to resolve disputes on the terms in which they present themselves, that is, as alleged injustices that the plaintiff suffers at the defendant’s hands. In doing so a court elucidates and applies the legal categories that pertain to the parties’ interaction. The coherent exercise of this adjudicative power takes the process of justification seriously by supporting its decisions through reasoning that is congruent with the contours of the parties’ relationship. Adjudication so conceived is fair to both parties because it treats the parties as equals, entitled to reasons for liability that reflect the correlative nature of the supposed injustice between them.

4.  Substantive Coherence and Structural Coherence Over the last few decades, the significance of legal coherence has been much discussed.10 I want now briefly to relate the corrective justice conception to this literature. The standard treatment of coherence takes as its subject the existing law—​its rules, principles and standards—​and then asks such questions as: Is the entirety of the law subject to a single coherent scheme (‘global coherence’) or is coherence a characteristic of specific fields (‘local coherence’)?11 Is the coherence of legal doctrine valuable independently of the doctrine’s moral soundness? If so, how strong is its limiting effect on the operation of on other normative considerations? Is the appeal of coherence vitiated by the conservatism of requiring subsequent doctrinal development to conform to already existing doctrine? And, of course, what does coherence mean in addition to the absence of inconsistency among the cohering elements? The crucial difference between the current literature on coherence and corrective justice is this: The current literature proceeds from an initial focus on the law’s substantive norms, which serve as the material that is or might be coherent. Some of the most prominent expositors of coherence then juxtapose this multiplicity of norms against the unity that coherence is thought to demand,12 perhaps a

10 See especially A Amaya, The Tapestry of Reason: An Enquiry into the Nature of Coherence and Its Role in Legal Argument (Hart Publishing 2013); A Amaya, ‘Legal Justification by Optimal Coherence’ (2011) 24 Ratio Juris 304, 306–​13; J Dickson, ‘Interpretation and Coherence in Legal Reasoning’ in E Zalta (ed), Stanford Encyclopedia of Philosophy (Winter 2016) ; K Kress, ‘Coherence’ in D Patterson (ed), A Companion to Philosophy of Law and Legal Theory (2nd edn, Blackwell Publishing Ltd 2010). 11 On this distinction see J Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics (OUP 1995) 298. 12 Amaya, The Tapestry of Reason (n 10) 70–​71 observes that most coherence theories do not define coherence in terms of unity but rather seek to provide a method for decision-​making in the face of the pluralism of irreducibly independent principles. I leave these views aside, because then the root contrast with corrective justice revolves around pluralism rather than coherence.

Structure  13 shared higher order principle or value,13 or a common justificatory premise.14 For corrective justice, in contrast, the point of departure for thinking about coherence is structure, not substance. The unity presupposed in coherence is simply the correlative structure of liability. Coherence lies in reasoning about liability that corresponds to that structure. For corrective justice, the questions preoccupying the current literature on coherence receive responses that are anchored in the stability of the relationship’s structure. Thus, so far as corrective justice is concerned, the coherence that it postulates is not global. Far from applying to the legal order as a whole, it is restricted to private law as the domain of correlatively structured relationships. Moreover, the questions of whether coherence is valuable independently of moral soundness and of how strongly it limits other normative considerations do not arise. For corrective justice, structure shapes substance. By providing the structure into which all the normative considerations relevant to private law have to fit if the law is to be fair to both parties, corrective justice allows no tension or tradeoffs between coherence and other normative considerations. Nor does corrective justice stand exposed to the reproach of conservatism in having doctrinal development conform to already existing doctrine. The corrective justice conception of coherence, being a reflection of structure, is indifferent to the order in which doctrine develops; it holds not that later doctrine should conform to earlier doctrine, but that the reasoning supporting any doctrine, whether earlier or later, should conform to the coherence implicit in the structure of the private law relationship. This may well require the transformation of earlier doctrine that is the outgrowth of incoherent reasoning. Of these two approaches to coherence, corrective justice comes in at a more fundamental level. For suppose that one identified the unity sought by the literature on coherence with some general principle discernible in different private law contexts. Unless this principle was itself coherent, its application to different legal situations would hardly contribute to the unity of the legal norms that fell under it. A principle that, for instance, juxtaposed two independent values would impart its segmented character to every situation to which it was applied. From the standpoint of coherence, what would be gained by the coherent application of an incoherent principle, even if this were possible? Corrective justice supplies, at least with regard to private law, the sense of coherence without which the effort of the standard literature to assess the nature and significance of coherence would be a vain enterprise. Because no substantive conception of coherence in private law could be adequate unless it conformed to the correlative structure of liability, corrective justice, one might say, supplies the condition for the very possibility of coherence for the substantive norms that determine liability. 13 N MacCormick, ‘Coherence in Legal Justification’ in A Peczenik, L Lindahl, and B van Roermund (eds), Theory of Legal Science (Reidel 1984) 235, 238. 14 R Alexy and A Peczenik, ‘The Concept of Coherence and Its Significance for Discursive Rationality’ (1990) 3 Ratio Juris 130, 132.

14  Ernest J. Weinrib

5.  Internal Understanding So far, I have been outlining the structural significance of corrective justice. I now want to address the connection between corrective justice as a theoretical idea about structure and private law as a familiar legal practice. To enquire into this connection is to ask what it is about private law that corrective justice illuminates. My answer is that corrective justice offers an internal understanding of the justificatory character of private law. In this and the following section, I want to enlarge upon, first, the internal dimension of this understanding and, then, its treatment of private law’s justificatory character. In private law, the gist of the plaintiff ’s claim is that he or she has suffered an injustice at the hands of the defendant. In dealing with such claims every sophisticated system of private law aspires to work out the juridical ideas through which they can be coherently and fairly evaluated. Because the claim is relational, the juridical ideas that address it are also relational. In the common law this relational character is evident in the procedure of the plaintiff suing the defendant, in the remedy that moves from defendant to plaintiff, and in many of the principal organizing concepts for the various bases of liability (for example, causation, contractual privity, the defendant’s duty of care to the plaintiff in negligence law, unjust enrichment at the expense of the plaintiff, the duty owed by fiduciary to beneficiary, and so on). This is not to say that private law, as it exists at any particular moment, does not have unfair or incoherent doctrines. An aspiration is not a guarantee. The degree of the law’s success in achieving coherence is a contingent matter. Nonetheless, the aspiration to fairness and coherence and the manifest presence of relational ideas within private law allow the initial working hypothesis that the law’s labours have not been completely in vain, and that the concepts elaborated in the law can be regarded, at least provisionally, as the products of good faith efforts to live up to that aspiration. In taking this attitude corrective justice considers these concepts to be objects worth understanding in their own terms, with a view to examining whether legal reasoning along their lines is truly expressive of fairness and coherence. The inquiry is internal to private law in every possible sense: it starts with the concepts as they are understood within the law; it examines them from the standpoint of the law’s own aspiration to fairness and coherence; it invokes a conception of fairness and coherence that is internal to the structure of the parties’ relationship; and the concept of structure itself highlights the internal composition of the nexus between plaintiff and defendant. This internal approach can be contrasted to external approaches, that is, to approaches that focus not on the internal structure of the private law relationship but on the substantive considerations thought to be normatively attractive independently of private law. Treating these considerations as the goals that private law does or should realize, an external approach enquires into the extent to which the norms of private law forward them, into the adjustments that would bring the law into

Structure  15 closer conformity with them, and into the tradeoffs necessary to accommodate all of them. Four points of contrast are salient. The first deals with the treatment of legal concepts, the second with the function of legal norms, the third with the role of law, and the fourth with the source of indeterminacy. First, the treatment of legal concepts. As already indicated, because the relational structure of corrective justice readily matches the relational concepts that organize private law, corrective justice can take them seriously in their own terms. This is not the case for the external approaches. The goals through which external approaches strive to elucidate legal arrangements bring with them their own distinct principles of organization. These principles typically focus on the advantages that accrue from particular ways of regulating the position of one or the other of the parties, or even of both seriatim, but not of each in relation to the other. For example, the most prominent of the contemporary external approaches, the economic analysis of law, postulates that private law does or should promote economic efficiency by using the imposition or withholding of liability as incentives for the parties to act in a cost-​justified manner. These incentives apply to each of the parties separately and for separate reasons; they do not link the parties to each other. Accordingly, economic analysis can neither provide a theory for the nexus between the parties nor explain the legal concepts through which the law expresses that nexus. The various external principles that contemporary scholars have invoked differ in content, but they all share this one-​sided orientation. Indeed, their one-​ sided orientation is what makes them external principles, because it alienates them from the relational character of private law. Consequently, the correlativity that structures private law from within has no more than a derivative importance, if it has any importance at all. As a result, external approaches either struggle to assimilate the relational features that figure so prominently within private law, or they dismiss standard legal concepts as otiose,15 as mere rituals,16 as veils to be pierced by clear-​headed analysis,17 or even as salutary obfuscations that themselves have instrumental value.18 Secondly, the function of legal norms.19 Under external approaches, the norms are means for forwarding and bringing about the postulated goal. It makes sense to think of them as means because an independent substantive goal exists to which they can be instrumentally related. Economic analysis, for instance, posits economic efficiency as the goal, and then considers how the rules for liability might 15 WA Landes and RA Posner, The Economic Structure of Tort Law (Harvard UP 1987) 227. 16 Jerome Frank, ‘What Courts Do in Fact’ (1931) 26 Ill LR 653. 17 Felix S Cohen, ‘Transcendental Nonsense and the Functionalist Approach’ (1935) 35 Columbia LR 809. 18 Guido Calabresi, ‘Concerning Cause and the Law of Torts: An Essay for Harry Kalven, Jr’ (1975) 43 U Chicago LR 107. 19 On this and the following paragraph see the elegant and suggestive observations of M Stone, ‘Planning Positivism and Planning Natural Law’ (2012) 25 CJLJ 219, 228–​33.

16  Ernest J. Weinrib promote this goal under various circumstances. In contrast, corrective justice reflects no independent substantive goal. To the extent that legal norms accord with corrective justice, they are not about means but about meaning. As I have been contending, corrective justice refers to the structure of correlativity implicit in the normative ideas and institutional arrangements that mark the parties’ relationship. It would hardly be intelligible to think of the norms as the means towards the promotion of their own implicit structure. Rather, the norms are specifications of the meaning that that structure has in the concrete legal relationships in which it figures. The norms are thus constitutive of the relationship, not instruments for bringing it about. In this respect, correlativity as the most inclusive abstraction of private law stands no differently with respect to its determinations than do the law’s less inclusive abstractions. Lord Atkin’s discussion of the role of foreseeability in the duty of care, for example, is an exposition not of the means through which a duty of care can be brought into being, but of the meaning of the duty of care once one abstracts from the particular duties of negligence law to the general conception that all such duties share.20 Similarly, Lord Atkin’s articulation of this general duty of care is not an instrument for achieving the substantive goal known as corrective justice—​corrective justice is not a substantive goal—​but rather a specification of one aspect of what corrective justice means in the relationship between negligent actors and the victims of their negligence. Thirdly, the role of law. Because external approaches postulate substantive goals that are independent of private law, the law for them is merely a contingent way of bringing the goals about. It may well be the case that the goal in question can be more perfectly attained outside private law; the independence of the goal at least allows one to entertain this possibility. Economic analysis once again provides a convenient illustration. Nothing about economic efficiency as an idea requires it to be entrenched within private law. Indeed, from the standpoint of economic efficiency private law has the disadvantage of encasing it within the confines of the relationship between plaintiff and defendant instead of allowing it to float free in the pursuit of its own distinctive end—​a disadvantage evident in the difficulties that economic analysis has with the relational concepts of private law. The point is implicitly recognized in economic analysis that ‘decouples’ the position of the defendant from that of the plaintiff.21 In contrast, corrective justice is not independent of private law. The correlativity that corrective justice highlights is the most inclusive abstraction from private law’s normative ideas and institutional arrangements. That abstraction would remain empty in the absence of its subordinate determinations of greater and lesser generality. These determinations, in turn, can come only within a posited system of private law. Of course the plaintiff, 20 Donoghue v Stevenson [1932] AC 562 (HL). 21 AM Polinsky and Y-​K Che, ‘Decoupling Liability: Optimal Incentives for Care and Litigation’ (1991) 22 Rand J Econ 562.

Structure  17 the defendant, and even third parties observing the controversy between them (moral philosophers, for instance) may have their own notions of how these determinations should go. But whoever specifies these determinations thereby sets the boundary between what the defendant can do and what the plaintiff must suffer. This boundary cannot be set unilaterally by the private will of either party or of any third party observer, but must rather be the product of impartial and disinterested deliberation by a legal authority. Accordingly, the determinations of corrective justice have to be specified by a body of law that validly and authoritatively governs the relationship between the interacting parties within its jurisdiction, that is, by a posited system of private law. Thus, law has a different role for corrective justice than it has for external approaches. The independent goals suggested by external approaches may possibly be realized without private law; for corrective justice this possibility does not exist. Unlike the independent goals, corrective justice finds a home only within private law. Fourthly, the source of indeterminacy. A goal postulated under an external approach and developed with suitable sophistication (utilitarianism and economic analysis are examples) may in principle allow for completely determinate prescriptions. For such a goal, the possible sources of uncertainty lie beyond it. One of these sources is factual uncertainty about the world or about the future. Another is uncertainty about how, if multiple goals are in play, the demands of one goal should be reconciled with the demands of other goals. However, in a world of perfect information where only one goal is relevant, it is conceivable that this goal might operate in a perfectly determinate way. The person charged with forwarding the goal under these circumstances would need knowledge but not judgment. For external approaches, then, the presence of indeterminacy is as contingent as is the need for law. And for the same reason: for if one thinks that a function of law is to specify what is otherwise indeterminate, the absence of indeterminacy renders superfluous the law’s performance of that function. In contrast, corrective justice is necessarily indeterminate. This is not a weakness but a strength, because it reflects what every sophisticated legal system acknowledges: the dynamism and necessity of judgment in the application of legal concepts. Corrective justice supposes an interplay of abstraction and determination at various levels of generality and particularity, with the most general and inclusive abstraction of all, the correlative structure of the parties’ relationship, acting as the highest regulative principle. In this interplay the more general levels do not generate a unique content for the particulars that fall under them; if they did, the difference between general and particular would evaporate. Rather, every abstraction is the principle of ordering for its determination, in that the most plausible judgment—​that is, the judgment optimally attentive to the normative significance of the abstraction—​has to be made about how that abstraction is to be specified. The making of that judgment is the responsibility of the relevant legal authority. These authoritative exercises of judgment can be more or less felicitous in how they

18  Ernest J. Weinrib specify the relevant abstraction. They also can (and inevitably do) vary with social change and from jurisdiction to jurisdiction, as different expositors of the positive law living at different times and under different social conditions may have different appreciations of how fairness and coherence are to be achieved in particular cases or groups of cases. The diversity and mutability of legal doctrine are consequential on the necessity for corrective justice to become determinate through judgment.22 Thus, a further feature of the internal aspect of corrective justice is that it treats determinacy not as something produced by goals independent of the law, but as the consequence of judgments authoritatively rendered within the legal system in accordance with the law’s internal dynamic for assigning a determinate meaning to its abstractions. In this way, corrective justice reflects the fact that law operates through institutions that provide determinacy for the abstractions relevant to private law relationships.

6.  Interpretive and Normative Dimensions I now move to the question of how this internal understanding illuminates the justificatory character of private law. How, in other words, does corrective justice’s understanding of that character work? This understanding has two dimensions, one interpretive and the other normative. Of course, in contemporary legal scholarship interpretation is a much discussed and contested idea. Without passing judgement on how others use the term or how it figures in their work, I want merely to indicate what I mean by it in the context of corrective justice’s understanding of private law. From the standpoint of corrective justice, to interpret is to understand the normative significance of some aspect of private law in relation to a larger whole of which it forms a part. Corrective justice supplies the largest notion of a whole that is relevant to private law: the structure of the parties’ relationship as the most inclusive unity with reference to which reasoning about liability can be understood to be coherent. As already explained, this whole is an abstraction made up of the less inclusive abstractions—​smaller wholes—​that operate at their respective levels of generality. The process of interpretation involves an appeal up the ladder of abstractions, from less inclusive ones to more inclusive ones, in the effort to understand whether or to what extent the matter being interpreted realizes the possibility of coherence implicit in the structure of the private law relationship.

22 J Stapleton, Three Essays on Torts (OUP 2021) 21–​24 presents the mistaken view that a non-​ pluralistic theory cannot account for doctrinal change and diversity. For more extensive reflections on this issue see EJ Weinrib, The Idea of Private Law (Harvard UP 1995, OUP 2012) 214–​29; EJ Weinrib, Corrective Justice (OUP 2012) 230–​34, 260–​61.

Structure  19 This conception of interpretation is the juridical analogue of the grammatical notion of construing. To construe is to set out the syntactical connection of the words within a sentence. For an inflected language this exercise involves identifying the part of speech of each word; the number, gender, and case of each noun, pronoun, and adjective; the mood, tense, voice, and person of each verb; and so on. The point is to see how the various parts of the sentence fit together as a unit of meaning. Construal thus exhibits the sentence’s construction, that is, the way its components form a structure. On the corrective justice side of this analogy, structure sets the terms on which the components of private law can coherently be linked. To interpret is to examine the connection between the reasoning that supports one element of a legal relationship and the reasoning that supports the other elements. Just as construal lays bare the syntactical connection among the components of a sentence, so interpretation explicates the normative connection among the components of a private law relationship. And just as construal situates specific words within a semantically meaningful set of words, so interpretation situates specific legal doctrines within a normatively coherent set of doctrines. Interpretation is thus an exercise in the exploration of normative connectedness, under the presupposition that the law does not live in fragments. The antithesis of interpretation is the treatment of private law as a collection of unrelated dooms, what Tennyson called ‘that wilderness of single instances’.23 In contrast, corrective justice attempts to understand the extent to which private law can be seen not as a wilderness but as a harmonious garden, where different groupings of flowers and foliage come together in larger and more inclusive clusters of mutually reinforcing beauty. In the case of private law, of course, the point of the exercise is not syntactical or aesthetic, but normative: can one regard particular doctrines of law as participating in broader legal ideas that express a coherent justificatory ordering? The great cases of the common law are often examples, sometimes halting and always provisional, of interpretation so conceived. This is, perhaps, what makes them great cases. The signal achievement of Lord Mansfield in Moses v MacFerlan,24 for instance, was to suggest that the seemingly disparate circumstances under which the law required the refund of money (mistake, failure of consideration, duress, undue advantage, etc.) were all examples of the same larger principle. And even though he did not articulate that principle very clearly (‘an obligation from ties of natural justice to refund’), his judgment laid the basis for the later extrapolation of the principle of unjust enrichment—​an achievement reached through the same process of recognizing the ‘distinct unity’ of ‘a group of situations . . . that had never been dealt with as a unit’.25 Similarly, Lord Atkin’s great judgment in Donoghue v Stevenson was premised on the interpretive assumption that the law cannot treat

23

Alfred, Lord Tennyson, Aylmer’s Field (1793). (1760) 97 ER 676, 2 Burr 1005 (KB). 25 WA Seavey and AW Scott, ‘Restitution’ (1938) 54 LQR 29. 24

20  Ernest J. Weinrib the particular instances of the duty of care as a chaotic miscellany of independent norms. As he put it, ‘there must be, and is, some general conception of relations giving rise to a duty of care, of which the particular cases found in the books are but instances’.26 In adopting this interpretive attitude, corrective justice constantly presses from the particular to the general, and thence to the more general. The point of interpretation is not to match corrective justice to any particular doctrine or even to any number of particular doctrines in accordance with a criterion of ‘fit’.27 Nor is it a criticism of corrective justice that corrective justice does not ‘fit’ this or that doctrine. Invoking ‘fit’ in this way would treat the doctrine in question as fixed, static, and self-​standing. Corrective justice, in contrast, views doctrine from the standpoint of the dynamism of coherent reasoning within a system. The interpretive importance of a doctrine increases the higher the level of generality at which the doctrine operates. Abstractions are thus interpretively more significant than the determinations they regulate, and the more inclusive abstractions are interpretively more significant than the less inclusive ones. Accordingly, one interprets a doctrine by interpreting the reasoning that supports the doctrine; that interpretation, in turn, reflects the place of this reasoning within the broader ensemble of abstractions pertinent to the particular relationship between plaintiff and defendant; and that ensemble of abstractions, in its turn, is ultimately understood in terms of the correlative structure of private law relationships in general, because conformity to that structure renders private law a coherent justificatory practice. In linking interpretation to coherence, corrective justice does not assert that the private law of any given legal system actually is coherent. Whether, to what extent, in what respects, and through what doctrinal mechanisms private law is coherent is for the interpretive project itself to disclose. In adopting its interpretive stance, corrective justice postulates, first, that coherence is indeed possible (thereby rejecting the facile dismissal of coherence by sceptics and pluralists), and second, that because incoherent justifications do not justify, coherence is necessary if private law is to be a truly justificatory practice. Interpretation fills the modal space between the possible and the necessary, by inquiring into the actuality of coherence in the justificatory moves made within a specific system of private law. What is responsive to the obvious contingency of coherence in the actual working of private law is the normative dimension of corrective justice. This dimension requires that private law bring itself into conformity with corrective justice. The normative dimension reflects the regulative status of the idea that reasoning about liability should correspond to the correlative structure of liability.

26 [1932] AC 562, 580 (HL). 27 The notion of fit was introduced to contemporary scholarship by R Dworkin, Law’s Empire (Belknap Press 1986) 67. I am not suggesting that Dworkin himself uses fit in the way mentioned in the text.

Structure  21 Because abstractions are regulative of their determinations, the normativity of structure, which is the most inclusive abstraction, cascades down the ladder of abstractions from the more general to the more particular. Animating this normative dimension is the conception of relational fairness implicit in corrective justice. As mentioned in section 3, corrective justice is fair as between the parties because, in looking solely to the parties’ legal relationship and thus to their correlative positions, it excludes considerations that are unilaterally applicable, for good or ill, to one or the other of them. Corrective justice thus treats the parties as normative equals. Its conception of fairness places private law and its institutions under an imperative to make this normative equality decisive for the elaboration of legal doctrine and for the disposition of disputes. Accordingly, corrective justice is both interpretive and normative. The interpretive dimension originates in corrective justice’s conception of coherence and makes that conception the basis of its interpretation of private law insofar as reasoning about liability reflects the structure of liability. The normative dimension is rooted in corrective justice’s conception of fairness and draws from that conception the requirement to bring private law into conformity with corrective justice. It will be recalled, however, that the two conceptions are not separate, because from the standpoint of corrective justice no incoherence can be fair and no fairness can be incoherent. From this one can conclude that because coherence and fairness are intertwined, the interpretive dimension and the normative dimension are also intertwined. Corrective justice, in other words, integrates interpretation and normativity. The idea that interpretation is intrinsically normative is a position associated with the work of Ronald Dworkin. As is well-​known, Dworkin contended that the interpretation of a law involves offering a justification that shows the law to be the best it can be from the standpoint of political morality.28 Corrective justice shares with Dworkin’s project the commitment to understand law without positing an external Archimedean point.29 However, corrective justice, with its Aristotelian origin, draws on a much older tradition of philosophical inquiry, one that goes back to the Greek preoccupation with a thing’s logos, that is, with providing an account of a thing’s rational intelligibility. For many Greek thinkers, the thing’s form or structure was crucial to the explication of its logos. In this spirit, one might say that I am claiming that corrective justice is the logos of private law. Given, however, that this this kind of formulation grates on contemporary ears, it might be at least be worth outlining how the idea of structure differentiates corrective justice

28 R Dworkin, ‘Law’s Ambitions for Itself ’ (1985) 71 Va LR 173, 178. The point is thematic for Dworkin (n 27). 29 On this aspect of Dworkin see A Ripstein, ‘Introduction: Anti-​Archimideanism’ in A Ripstein (ed), Ronald Dworkin (CUP 2007) 1.

22  Ernest J. Weinrib from Dworkin’s treatment of interpretation. The fact that Dworkin’s position is deservedly celebrated and familiar allows me to be very brief. First, Dworkin’s argument moves from interpretation to normativity. The basic step is to establish that law is an interpretive practice. The argument then is that the very exercise of interpretation has the justificatory dimension of showing the object of interpretation in its best light. The account of law that Dworkin provides is normative by virtue of its being interpretive. In comparison, corrective justice claims to be both normative and interpretive by virtue of being corrective justice. The normative dimension is not an aspect of the law’s interpretive activity. Rather, both the interpretive and the normative are aspects of a justificatory practice that is structured so as to have its own implicit conceptions of fairness and coherence. This structure both ordains the kind of interpretation that is in conformity with it and prescribes the kind of normativity that actualizes it. The movement of thought is from normative ideas and institutional arrangements of private law back to the structure of the private law relationship and then forward to the interpretive and normative dimensions. Secondly, Dworkin and corrective justice differ in their proposed justificatory standpoints: Dworkin’s lies in political morality; corrective justice’s lies in the structure of the private law relationship. Thus, Dworkin’s eye is fixed on a political ideal that, while compatible with our legal and moral practices, is certified as best by ‘abstract moral theory’,30 that is, as a matter of moral argument at large. Corrective justice, in contrast, looks more narrowly to the normativity in the structure of the parties’ legal relationship. That normativity is not political, but juridical; that is, it is specific to the kind of legal relationship to which it applies.

7.  Two Aspects of Legality Having mentioned the juridical, I want in this the final section of the chapter to make some observations about its jurisprudential significance. As a legal phenomenon, private law has two aspects internal to it: the positive and the juridical. Legality qua positive reflects the concepts of validity and authority that have been standard to general jurisprudence since the rise of legal positivism. These concepts respond to the basic theoretical question raised by legality qua positive: what does it mean for law to exist? Validity constitutes law’s specific mode of existence, and authority refers to the specific mode of bringing law into existence. Validity and authority are formal matters that are indifferent to whether the law is good or bad. We are all familiar with John Austin’s famous pronouncement that ‘the existence of law is one thing, its merit or demerit is another’.31

30 31

Dworkin (n 27) 301. J Austin, The Province of Jurisprudence Determined (WE Rumble ed, CUP 1995) 28.

Structure  23 In the case of private law, legality qua juridical, however, refers the conceptions of fairness and coherence implicit in the correlative structure of the private law relationship.32 Just as validity is law’s specific mode of existence, so correlativity structures private law’s specific mode of justification. And just as validity is the internal characteristic of law that exists, so fairness and coherence are the internal characteristics of adequacy in the elaboration of private law doctrine. In these respects legality qua positive and legality qua juridical are parallel. In other respects they are not. Although both are formal, legality qua positive is indifferent to the law’s content, whereas legality qua juridical aims at grasping the content that conforms to private law’s internal criteria of justification. In other words, legality qua positive is form that can have any content, whereas legality qua juridical is form that requires an adequate content. Consequently, law qua positive is either existent or non-​existent—​existence has no gradations—​whereas law qua juridical can be realized to varying degrees of adequacy. These two aspects of legality operate in private law in different ways. Legality qua juridical has a regulative function through the abstractions of varying generality that guide the process of reasoning. Legality qua positive has a specificatory function that determines the meaning of the abstraction in particular circumstances. Because a determination at one level of generality can be an abstraction at another, it is usually the case that both aspects of legality are simultaneously present in the same doctrine. The neighbour principle formulated by Lord Atkin in Donoghue v Stevenson,33 for instance, was a determination of the concept of duty that effectively supplanted the common law’s previous determinations. It was also an abstraction that invited and subsequently received further determinations both subordinate (did the neighbour principle apply to dangerous conditions that others could have inspected and discovered?)34 and co-​ordinate (did the new general conception of duty require a revision in the conception of proximate cause?).35 The overlapping of the positive and the juridical aspects of legality in a single doctrine that both specifies the law that exists and regulates the law that should exist may make Austin’s dichotomy between existence and merit more complex than at first sight, but it does not fundamentally undermine it. Corrective justice presents the same overlap of the juridical and the positive not with respect to particular doctrines but at the highest level of generality. The correlative structure of the parties’ relationship is the abstraction for which the entire system of private law is the determination.

32 Similarly, beyond private law, the juridical (as I use the term in this book) refers to what is normatively implicit in other kinds of legal relationship or in law as such. In ch 8 below, for instance, I will treat the rule of law as a juridical idea relevant to both private law and public law. 33 See n 26 above. 34 Clay v AJ Crump & Sons Ltd [1964] 1 QB 533 (CA). 35 Overseas Tankship (UK) v Morts Dock & Engineering (The Wagon Mound No 1) [1961] AC 388 (PC).

24  Ernest J. Weinrib For corrective justice the difficulty with Austin’s dichotomy lies in his mention of merit or demerit. Stating the dichotomy in this way opposes to legality qua positive not another aspect of legality but the entire range of possibly meritorious considerations that are not legal, in the sense that they have not—​or not yet—​been brought into existence as valid norms of the positive law. What I have been calling ‘legality qua juridical’ is thereby subsumed under a general notion of morality. This general notion includes moral ideas that figure in external understandings of law, that is, ideas that are identifiable and thought to be morally justified independently of the law. The law can then be conceived in instrumental terms as the means for promoting them. So far as its content is concerned, law becomes simply morality plus the accoutrements of validity plus the consequence of coercive enforceability. Austin’s own utilitarianism is a paradigmatic example of the relation so conceived between law and morality. In a wonderfully suggestive article Martin Stone has argued that, since no reputable natural law theorist had denied the positivity of law, the point of Austin’s legal positivism was to transform not our conception of law but our conception of morality.36 Be that as it may, by insisting that the juridical is an aspect of legality, corrective justice is true to its natural law roots. Corrective justice is the special kind of morality that is distinctive of private law, that must be expressed through private law, and that is reflected in the characteristic content of private law. An inconsistency in private law with this special kind of morality is not merely bad law but bad as law, without its validity being thereby affected. Within the natural law tradition this special kind of morality is known as jus, the concept directed towards the rational intelligibility of law as a distinctive normative order. This chapter has merely emphasized the role of structure in the jus (the juridical understanding) of private law. Should a legal positivist object to this? That depends on what the core of the positivist project is taken to be. If the core is that the existence of laws or legal systems is a matter of their validity only, there is no conflict with corrective justice, given its recognition of law qua positive. If the core is that one understands law as such in terms of its existence only and not of its rational intelligibility, then corrective justice will seem misconceived.37 36 M Stone, ‘Legal Positivism as an Idea about Morality’ (2011) 61 UTLJ 313. 37 What a legal positivist will certainly object to, however, is the possibility that a juridical consideration may play a constitutive rather than merely regulative role for any legal relationship. This happens when a juridical consideration is thought to be so indispensable to the project of law that its violation is necessarily beyond a lawmaker’s authority. This was the gist of Julius Ebbinghaus’ analysis of the Nazi regime; J Ebbinghaus, ‘Positivismus: Recht der Menschheit—​Naturrecht—​Staatsbürgerrecht’ in J Ebbinghaus, Gesammelte Schriften, Band 1, Sittlichkeit und Recht, Praktische Philosophie 1929-​1954 (Bouvier Verlag 1986). Although this possibility qualifies what it means for law to exist, it does not equate law with its moral merit; Ebbinghaus expressly disavows the idea that a law’s injustice affects its validity. In his view the problem with Nazi orders was not that they were unjust but that they violated the law of humanity (Recht der Menschheit) that is required for the very idea of law. Ebbinghaus’ position is more fully (and approvingly) described below in ch 8, s 4. Contrast Ebbinghaus’ characterization of Nazi orders as pseudo-​law, with HLA Hart’s more famous positivist analysis of Nazi orders as valid

Structure  25 In the growing elaboration of legal positivism many have wondered what exactly is the issue between the positivists and their supposed opponents—​a perplexity all the more baffling because few of those opponents deny law’s positivity. The idea that so far as private law is concerned, corrective justice is legality qua juridical may be a way of seeing what, if anything, is at stake.

but morally odious law; HLA Hart, ‘Positivism and the Separation of Law and Morals’ (1958) 71 Harv LR 593.

2

Rights 1.  From Structure to Rights As suggested in the previous chapter, the fundamental idea of corrective justice is that reasoning about liability should match liability’s correlative structure. The point of liability is to correct an injustice. Because liability is a correlatively structured phenomenon, that is, because a particular defendant is always held liable to a particular plaintiff, the injustice that liability corrects is itself correlatively structured. And because injustice is not a brute event but a normative ascription on the basis of reasons, the reasons for considering something an injustice should also be correlatively structured. Only on this basis can private law treat the parties fairly with respect to each other and link them through coherent reasoning. Thus, the correlative structure of the liability allows us to identify the kind of reasons appropriate to determinations of liability. Moving from structure to substance, in this chapter I explore the significance of rights within corrective justice. Attention to rights follows naturally from connection between reasons and the correlative structure of liability. The reasons for liability can be correlatively structured only if they employ concepts that are correlatively structured. A right is such a concept, as it immediately implies a correlative obligation. Liability arises when an act or condition that is imputable to the defendant is inconsistent with the plaintiff ’s right; the defendant, accordingly, is obligated to avoid creating the inconsistency and to rectify it once it occurs. In determining that the defendant is liable to the plaintiff, a court affirms the existence of the plaintiff ’s right and its correlative obligation, sets out the ground for imputing to the defendant an inconsistency with the plaintiff ’s right in the circumstances of the particular case, and specifies how the defendant is to remedy this inconsistency. The correlativity of right and obligation thereby links the normative positions of the parties in accordance with the correlative structure of their relationship. The question to be addressed now is this: what is the conception of rights that is most fully expressive of the correlative structure highlighted by corrective justice? In this connection, the following objection might arise. Corrective justice, of course, has no monopoly on the commonplace notion that rights operate correlatively to their attendant obligations. Accordingly, one might be tempted to say that the very triteness of the idea that rights have correlative obligations shows the emptiness of corrective justice. Whatever the positive law lays down as a right on whatever grounds would necessarily entail the correlativity of the corresponding Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0002

Rights  27 obligation. It may well be the case that the correlative structure of liability is an important and indeed characteristic feature of private law, but this has no substantive implication for legal reasoning. So long as the positive law treated the reasoning about liability as specifying interests that are to be accorded the status of rights within the bipolar litigational framework of private law, then correlativity would be fully in play.1 In other words, the corrective justice move from structure to substance leads nowhere. This objection, however, rests on a misapprehension. Corrective justice is concerned not merely with the correlative operation of posited rights and obligations, but with the juridical adequacy of rights as markers of correlatively structured relationships. As mentioned at the end of Chapter 1, juridical adequacy refers to reasoning that conforms to the conceptions of fairness and coherence implicit in the structure of the parties’ relationship. Accordingly, corrective justice requires that the considerations that undergird a right be inherently correlative, not merely that they become correlative through the operation of the positive law. So far as corrective justice is concerned, the correlativity of right and obligation is intrinsic to the reasoning that endows them with their normative character, so that the content of the rights and obligations fits the correlative structure that they instantiate. Correlativity does not start from the rights of positive law but goes all the way down into the composition of the rights themselves. This desideratum affects the kind of consideration relevant to the normative character of a right for purposes of corrective justice. Corrective justice is an idea that focuses not on one or the other of the parties to the relationship (or even on both of them), but on the relationship itself. Accordingly, the rights and corresponding obligations of corrective justice arise not out of factors that pertain only to one or the other of the parties, but out of the interaction, actual or prospective, between the two of them. The crucial issue is the normative character of this relationship as a relationship rather than as a juxtaposition of parties for whom a consideration separately relevant to one of them is enshrined as a right that somehow imposes an obligation on the other. Consequently, in accounting for rights, corrective justice excludes considerations, such as the welfare or the interests of the plaintiff, that pertain only to one of the parties. Considerations of that sort make no correlative reference to anyone else. If rights were conceived merely as shorthand allusions to components of the plaintiff ’s welfare or to subsets of the plaintiff ’s interests, then rights would ultimately be as unexpressive of correlativity as welfare and interests themselves. From the standpoint of corrective justice, the advance achieved by introducing the notion of correlativity would be undone by the non-​correlativity of the content of those rights. 1 Examples of this move are P Cane, The Anatomy of Tort Law (Hart Publishing 1997) 10–​15; H Dagan, The Law and Ethics of Restitution (CUP 2004) 224–​28.

28  Ernest J. Weinrib To avoid one-​sided considerations such as welfare or interests, corrective justice views the parties purely as interactors: the correlativity of their situations consists in their being normatively related to each other as, so to speak, interactor and interactee. The law is interested in whether the defendant, in exercising his or her purposiveness in a way that affects the plaintiff, acted rightfully. Under the conception of fairness implicit in corrective justice,2 the normative premise of this interaction is that the parties rank as equals, so that neither is subordinated to the purposes of the other. Thus, rights are components of a regime of equal freedom in which one can act for one’s purposes, whatever they are, so long as one’s actions are consistent with the equal freedom of others to act for their purposes. Private law rights so understood are the legal categories that, without regard to any particular purpose, provide the framework for the exercise of the capacity for purposiveness in relation to another. The point of these rights, and of the limits they set on the conduct of others, is adverbial: they govern how persons are to go about pursuing their purposes, not what their purposes have to be. This adverbial character manifests itself on both sides of the relationship. On the right-​holder’s side the rights set out spheres of freedom within which the right-​holder’s actions are consistent with the freedom of others. Rights are thus the legally guaranteed facilities for the pursuit of one’s self-​chosen ends. On the obligor’s side, rights provide a set of negative markers, allowing the obligor in turn to pursue his or her self-​chosen ends in any way that does not transgress the right-​holder’s guaranteed sphere of freedom. Private law, accordingly, is not an exercise in the moral perfectionism of mandating the pursuit of particular individual or societal purposes, no matter how laudable such purposes might seem. The mandating of particular purposes is incompatible with the correlativity of corrective justice, because a person’s particular purposes pertain only to the person whose purposes they are, and have no correlative significance for anyone else. And society’s enforcement of particular purposes through private law would be inconsistent with the equal freedom of all to set and pursue each their own purposes. Accordingly, the various kinds of private law rights serve as the juridical conduits through which persons interact on terms of their equal reciprocal freedom. This is evident in all the kinds of rights fundamental to private law: the right to bodily integrity, the right to property, the right to contractual performance, and the right to fiduciary loyalty. First, one’s body is the organ of one’s purposive activity, so that interferences with the bodily integrity of others are necessarily inconsistent with their capacity to determine their purposes for themselves. Secondly, a system of property binds everyone to the rights connected to everyone else’s exercises of purposiveness in the acquisition, use and alienation of objects external to oneself. Thirdly, contract rights instantiate the equal freedom of the contracting parties



2

See ch 1, s 3.

Rights  29 to create rights through expressions of mutual consent. Fourth, in fiduciary relationships, where one person’s interests are subject to the purposiveness of another, the law postulates a right to the fiduciary’s loyalty, thereby precluding fiduciaries from exercising this purposiveness for their own benefit. None of these kinds of rights renders any particular purpose mandatory. Rather, they constitute the legal categories under which the exercise of purposiveness by one person, to whatever ends that person chooses, can be consistent with the purposiveness of others.

2.  Kant, the Concept of Right, and Subjective Rights Kant’s legal philosophy provides the classic articulation of this conception of rights. For Kant, the different kinds of rights demarcate different kinds of juridical relations that are compatible with the freedom of everyone. The point of the system of rights is not to promote particular ends, but to provide the conditions for the rightful interplay of whatever ends persons might choose to have. These conditions secure the independence of each person from the constraint of another’s will. In Kant’s opaque but concise formulation, ‘[I]‌n this reciprocal relation of choice, no account is taken of the matter of choice . . . All that is in question is the form of the relation of choice on the part of both, insofar as the choice is regarded merely as free, and whether the action of one can be united with the freedom of the other in accordance with a universal law.’3 Rights are the legal categories through which private law gives expression to the freedom to which Kant refers in this statement. Mention of a philosopher as forbidding and complex as Kant may cause lawyers to shudder. Some may be tempted to dismiss his ideas as irrelevant, on the grounds that those who developed the common law were not influenced by him4 or were motivated by different concerns.5 This is a self-​impoverishing attitude. Whatever else their motivations, the common law judges, especially the greatest of them, sought to elaborate a reasoned and coherent body of law. To be sure, Kant’s aim, to establish an a priori metaphysics of right, is of no interest to lawyers. Nonetheless, his pursuit of that aim required him systematically to elucidate the distinctive coherence and rationality of juridical relations. He thus supplies a repository of insights, formulated with unparalleled rigour, into the conceptual presuppositions of the very activity that engages the jurists of any sophisticated system of law. These insights are especially pertinent to the nature of rights, for Kant was arguably the greatest expositor of what it means to have a system of rights. Kant traces the theoretical implications of a system of rights from its origin in the equal reciprocal freedom of all to its actualization in the institutions of a legal order. His

3

I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 27 [6:230]. R Stevens, Torts and Rights (OUP 2007) 329. 5 S Hedley, ‘Is Private Law Meaningless?’ (2011) 64 CLP 89, 99. 4

30  Ernest J. Weinrib account suggests how features of rights (the special status of the right to personal integrity, the difference between in rem and in personam rights, the role of acts of will in the acquisition of rights, the function of courts in determining rights, and so on) might be understood as fitting together in a coherently integrated ordering. His unfamiliar language deals with a familiar legal world. The point of my invoking Kant in this context is juridical not metaphysical. As already mentioned, attention to the internal fairness and coherence of private law’s specific mode of justification is the mark of the juridical so far as private law is concerned. My goal in the remarks that follow is not to vindicate Kant’s larger philosophical ambitions, but to deploy his ideas in exhibiting the interior conceptual organization of private law. Corrective justice seizes on the truism that liability links the plaintiff and the defendant, in order to explore the kinds of reasons for liability that match liability’s correlative structure. Only such reasons, so the argument goes, can be fair and coherent as between the parties. Rights and their correlative obligations are the concepts that figure in the correlatively structured reasons for liability. But how are the rights themselves to be understood? The significance of Kant is that his conception of rights, grounded as it is in the co-​existence of the defendant’s action and the plaintiff ’s freedom, is uniquely attuned to the requirement of corrective justice that reasons for liability be correlatively structured. Whether this contention about Kant is right or wrong, there is nothing metaphysical about it. With respect to private law, one can discern three levels at which rights operate in Kant’s account: at the level of the legal order as a whole, of relationships between persons, and of specific entitlements. As with the abstractions and determinations of correlativity discussed in Chapter 1, these three levels of rights form a set of more and less comprehensive abstractions, with the less comprehensive ones serving as the determinations of the more comprehensive ones. In this way, the mode of enquiring into rights runs parallel to the mode of enquiring into correlativity. I want now to consider these three levels in turn. The present section deals with the first two of these and the connection between them. In the next section, I will focus on the relationship between the second and the third. First, at the most general level, the entire system of rights is informed by the concept of Right (das Recht) itself. Right is the totality of conditions under which the action of one person can co-​exist with the freedom of another. Right thereby signifies the general character of the legal system as a systematic domain of equal reciprocal freedom. This general character informs every aspect of law: the various bases of liability; the institutional arrangements that specify and secure what each person rightfully has; and the requirements, associated with the rule of law, that legal norms be publicly knowable, systematically organized, and applied within the boundaries of lawful authority. Right so understood is not the subjective holding of any particular person, but the objective condition that qualifies the legal system as a whole when that system is conceived in terms of rights. To borrow a phrase from

Rights  31 German constitutional jurisprudence, Right refers to ‘the objective order of value’6 into which the various subjective rights and their correlative obligations fit. Because it focuses on the co-​existence of one person’s action with another’s freedom, the concept of Right asserts, at the most general normative level, the law’s relational character. Every right under this concept is infused with the concept’s relationality. Therefore, the rights that a person has do not owe their legal status to the possibility that they encapsulate interests that are important to the right-​holder on their own, as if this importance is first conceived independently of the law’s preoccupation with relationships and then imbued with legal form. Rather, under the concept of Right, a right held by a particular person is conceivable only in and through a legal relationship with another person, upon whom a corresponding obligation is immediately incumbent. A right to physical integrity, for example, does not derive its legal status from the importance of bodily integrity considered on its own, but from the idea that, because the body is the organ of one’s freedom, action cannot co-​exist with the freedom of a person whose physical integrity the action injures. This brings us to the second level for the operation of rights, the level of the particular kinds of legal relationships between persons. At this level, the subjective rights (that is, the rights held by particular persons) are the normative markers of how the freedom of one person is connected to the action of another. The subjective rights are thus constitutive of Right’s objective order of value.7 Moreover, each of the subjective rights has a correlative obligation, thereby instantiating the correlativity that defines corrective justice. The various subjective rights indicate the different ways in which persons can be related through their reciprocal freedom. Among the subjective rights the central organizing distinction is that between right as innate and right as acquired. The innate right is the right that one has by virtue of one’s very existence, so that one does not have to do anything to acquire it.8 Defined as independence from being constrained by another’s will, the innate right is merely the concept of Right expressed in terms of the subjective right immediately implicit in it. One’s physical embodiment is a manifestation of this right. It is innate because one does not have a right to one’s body by virtue of performing an act of acquisition; and it is implicit in the concept of Right, because without our having a right to our physical integrity, equal reciprocal freedom would be inconceivable. In contrast to the innate right, acquired rights are rights to objects external to the person, which become one’s own through an appropriate act of acquisition. Whereas the innate right refers to what is rightfully every person’s by virtue of his 6 BVerfGE 7, 198, B, II, 1 (Lüth 1958). 7 Compare Windscheid’s dictum that ‘[d]‌ie Rechtsordnung ist die Ordnung der Rechte’ cited in H Dedek, ‘From Norms to Facts: The Realization of Rights in Common and Civil Private Law’ (2010) 56 McGill LJ 77, 100. 8 Kant (n 3) 33 [6:237].

32  Ernest J. Weinrib or her existence, a person’s acquired rights are contingent on the particular history of each right-​holder. Because the objects of the acquired rights are distinct from the person and are acquired through an act of the will, Kant calls them ‘external objects of choice9—​a compendious term that includes not only the objects of property rights but also actions in fulfilment of contracts and conduct appropriate to fiduciary relationships. The rights to the various external objects of choice comprise legal categories through which persons associate in terms of reciprocal freedom, such as the reciprocal freedom of persons bound to respect one another’s property holdings, or of parties who are consensually bound to each other in contract, or of participants in a fiduciary relationship in which the right to act on behalf of another is exercised solely in the other’s interests. Acquisition connects the right-​ holder to an external object of choice in a way that places others under an obligation with respect to that object of choice. Right’s objective order of value and the kinds of various subjective rights comprising that order form an integrated but articulated movement of thought. On the one hand, the characterization of the objective order of value indicates that the subjective rights are not devices for achieving welfare at a higher level of generality; the concept of Right is the highest level of generality, and it merely reiterates more comprehensively the idea of equal reciprocal freedom instantiated in the subjective rights. On the other hand, the concept of Right needs to be specified in terms of the subjective rights, for only those rights allow the concept of Right to be actualized in legal relations. In the common law, for instance, a person’s innate right is protected in a number of ways, including the law of negligence and the intentional torts that safeguard the body’s integrity and its freedom of movement. In this way the bases of liability are the modalities through which equal reciprocal freedom becomes effective in the positive law. This last point is sometimes overlooked. In his searching essay on rights, for instance, Nigel Simmonds remarks on the emptiness of the Kantian notion of equal freedom. After referring to the co-​existence of everyone’s freedom under Kant’s concept of Right, he observes that ‘[i]‌n every conflict of will upon which we must adjudicate there will be gains and losses of freedom whatever decision we give’.10 His assumption in making this comment seems to be that the concept of Right applies directly to the adjudication of conflicts; because adjudication will produce a winner and a loser, the result will always be a restriction of the freedom of one of the parties and an enhancement of the freedom of the other. This comment ignores the significance of the subjective rights in mediating between the abstract notion of equal freedom and the outcomes of particular conflicts. A system of rights does not organize itself around the gains and losses from adjudication relative to the 9 Kant (n 3) 42 [6:247]. 10 NE Simmonds, ‘Rights at the Cutting Edge’ in MH Kramer, NE Simmonds, and H Steiner, A Debate Over Rights (OUP 2000) 190.

Rights  33 situation antecedent to the adjudication; that comparison obviously yields winners and losers. Rather, the benchmark of equal freedom is always the subjective rights themselves and their correlative obligations; these are the categories of equal reciprocal freedom with which everyone’s action must be consistent. Because the various rights are the different ways in which the freedom of the plaintiff and the defendant can be related to each other on terms of equality, adjudication that honours those rights maintains the parties’ equal freedom, regardless of who wins the case. For example, insofar as the right to contractual performance plausibly expresses the equal freedom of promisor and promisee, proper adjudication of a particular controversy between them produces no gain of freedom for the winner and no loss of freedom for the loser. Because the contract itself is the juridical expression of their reciprocal freedom, the parties each leave the courthouse with the same freedom with which they entered: the freedom to which their contract entitled them. As between the parties, the applicable subjective right and its corresponding obligation constitute the juridically decisive tie. This tie operates through the particular object on which the right and the obligation converge. One has a right to something in private law when one is legally so connected to the object of the right that another person is under an obligation not to interfere with it.11 The object of the right is what, in the circumstances of the parties’ relationship, the right vests in the right-​holder—​for example, the right-​holder’s personal integrity or the property she has acquired or the contractual performance of a particular promisor. The various kinds of rights denote the various kinds of objects through which the parties can be juridically related. The object thus signifies both what the right is a right to and what the obligation is directed towards.

3.  Entitlements The right and its corresponding obligation function as the organizing ideas for the specific entitlements of the right-​holder and for the specific acts or omissions incumbent on the obligor. These entitlements, acts, and omissions form yet another level of specificity. On the right-​holder’s side of the relationship are the entitlements authorized by the subjective rights.12 Every right has its particular set of entitlements, for without these the right would not function as a manifestation of freedom in the actual interaction of particular persons. The entitlements specify what it means to 11 Kant (n 3) 37 [6:245]: ‘That is rightfully mine (meum iuris) with which I am so connected that another’s use of it without my consent would wrong me.’. 12 On entitlements see JD van der Vyver, ‘The Doctrine of Private Law Rights’ in SA Strauss (ed), Huldingsbundel vir W. W. Joubert (Butterworths 1988) 201, 212–​14. Kant calls these entitlements ‘authorizations’ (Befügnisse); see eg Kant (n 3) 30 [6:237].

34  Ernest J. Weinrib enjoy the right and what the right-​holder can do by virtue of having the right. They comprise the right’s content. For example, the bundle of rights of which property is often said to be composed—​the right to possess, control, use, enjoy, dispose of, and so on—​refers to the entitlements that specify what the owner of property may do qua owner. Similarly, to ask whether a person can make himself another’s slave, whether the promisee is entitled to the gains realized by the promisor through breach of contract, or whether a fiduciary can use his position to exploit an opportunity not available to the beneficiary is to enquire into the entitlements, respectively, of the rights of personal integrity, contract, and fiduciary loyalty. At the same level on the other side of the relationship are the specific acts or omissions required for conforming to or discharging the obligation. In Kantian terms, these acts and omissions comprise the duties of the obligor. Kant distinguishes significantly between obligation and duty. Obligation is defined as ‘the necessity for a free action under a categorical imperative of reason’;13 duty is defined as ‘that action to which someone is bound’.14 This distinction, obscure as it may seem, merely spells out what is involved in Roman law’s description of an obligation as a vinculum iuris, ‘a legal fetter by which we are constrained by the necessity of performing something according to the laws of our state’.15 Kant’s distinction rehearses the contrast implicit in this description between the necessity for an action and the action to which one is thereby necessitated. Duty, in Kant’s usage, is the ‘matter’ of the obligation. Obligation provides the reason for the necessity of performing certain actions; duty refers to the specific actions that have to be performed in order to fulfil the obligation. In this brief reconstruction of the Kantian picture (Kant does not expressly lay it out the way I just have), right and entitlement, on the one hand, and obligation and duty, on the other, form two sets of parallel ideas. Entitlements are to a right what duties are to an obligation: determinations that specify the content of their more general abstractions. As is always the case within corrective justice and its associated concepts, the more general ideas are regulative of the more specific ones. The entitlements and the duties derive their normative significance from the rights and the obligations that they specify. The rights and the obligations in turn derive their normative significance from the objective order of value inherent in the concept of Right as the most abstract representation of equal reciprocal freedom. A consequence of distinguishing in this way between obligations and duties is that obligation and duty do not have a one-​to-​one correspondence. A single duty can be derived from different obligations. As Kant notes, ‘there can be one and the same duty (as to the action) although we can be bound to it in different ways’.16



13

Kant (n 3) 17 [6:222].

14 ibid.

15 Justinian, Institutes 3.13pr. 16

Kant (n 3) 15 [6:222].

Rights  35 Lawyers recognize this routinely in situations of concurrent liability, when the same act can be required as a matter of both contract law and tort law. The converse is also the case, that a single obligation can give rise to many duties, that is, to a multiplicity of acts or omissions that are incumbent on the obligor. This will be trivially true of a negative obligation (an obligation, say, not to injure someone negligently), for then the acts that the obligor is under a duty not to perform are countless. It is also trivially true when the terms of a single contract impose several duties. Accordingly, under the Kantian approach at least, it is not correct to say that ‘obligations are . . . individuated according to the actions that they make obligatory’.17 Being governed by duties, actions do not individuate obligations. Instead, obligations are individuated by the law’s treatment of the particular legal objects (my physical integrity, the horse that I own, the contract that I have made with you, and so on) towards which the obligations are directed. An obligation can be made up not only of multiple primary duties (the duties to perform or abstain from specific acts), but also of the secondary duty to remedy the breach of a primary duty, for example, by compensating the right-​holder for infringing the right. By failing to perform the specific act that the primary duty requires, the defendant breaches her duty but does not terminate her obligation. The vinculum iuris that ties the defendant’s obligation to the plaintiff ’s right remains intact. If it did not, the law would have no categorical reason to award a remedy that moves from the defendant to the plaintiff. Rather, the right—​and with it, the corresponding obligation—​survives its violation and lives on in the requirement that violator pay damages to the right-​holder. Thus, even if the defendant completely destroys something to which the plaintiff has a right, the defendant does not thereby destroy the right itself. The parties continue to be related through the normative connection that the right-​holder has with the object of the right, and through the consequent obligation on the defendant to act in conformity with that connection. Instead of being embodied in the object that the defendant has destroyed, the right and the corresponding obligation now take the form of an entitlement to have the defendant furnish its value. Despite the change in the specific act required of the defendant, the same obligation is continuously in play from the inception of the primary duty to the discharge of the secondary one. The level of entitlements and duties renders rights more determinate by dealing with the specific ways in which a right can be enjoyed and with the specific acts incumbent on obligors. It is at this level that the Hohfeldian categories of jural relation become relevant. Hohfeld’s four pairs of correlative categories (claim-​right and duty, liberty and no-​right, power and liability, and immunity and disability)18

17 J Gardner, ‘What is Tort Law For? Part 1: The Place of Corrective Justice’ (2011) 30 L and Phil 1, 35 (emphasis added). 18 WN Hohfeld, ‘Some Fundamental Legal Conceptions as Applied to Judicial Reasoning’ (1913) 23 Yale LJ 16, 30. Throughout, I shall use the term ‘liberty’ rather than Hohfeld’s term ‘privilege’.

36  Ernest J. Weinrib provide a typology of the generic legal significance of specific acts. The first branch of this typology (claim-​right and duty, liberty, and no-​right) deals with the status of a specific act as obligatory or permitted. The second branch (power and liability, immunity and disability) deals with the effect of a specific act on the legal position of another. Within private law the Hohfeldian categories exhibit the specific forms of correlativity that arise within the overarching relationship of right and correlative obligation. Thus, correlativity both infuses the parties’ relationship as a whole and characterizes the specific incidents of that relationship.19 Inasmuch as they represent different kinds of entitlements within the private law relationship, the Hohfeldian categories are, as it were, links in the chain of the vinculum iuris. The Hohfeldian and Kantian projects are sometimes considered to be deeply antithetical, so much so that Hohfeld is treated as having forever shattered the possibility of systematizing law on Kantian lines.20 This assertion of a deep incompatibility is overstated. Although in their treatments of rights Kant and Hohfeld diverge in one respect, they converge in another. The respect in which they diverge is that Kant’s account of rights, but not Hohfeld’s, is tied to the specification of the various kinds of object that rights can have. A right for Kant is a right to something, such as to one’s body, to one’s property, or to the contractual performance to which one is entitled. What the right-​ holder is entitled to enjoy by virtue of having the right and what others are under a correlative obligation to do or not do relates to the particular object of the right. Thus, having a property right in a thing imposes on others an obligation to abstain from that thing; and having a contractual right imposes on the other party an obligation to a particular performance under the contract. Such attention to the objects of rights follows from Kant’s project of exploring the conditions under which one person’s action can co-​exist with another’s freedom. Within this project the questions of how freedom is legally constituted by one’s rights, of how those rights limit another’s action, and of how the rights together form a coherent legal order cannot be considered without reference to the kinds of objects that the rights govern. Kantian rights indissolubly link right-​holders and their respective obligors, but they do so through the medium of the various objects of the rights. In contrast, Hohfeld’s treatment of rights makes—​and can make—​no reference to the object of a right. Hohfeld’s exposition is of the different kinds of relations of correlativity that can be instantiated by different activities. ‘Each Hohfeldian relation (e.g., of claim-​ right: duty or of privilege: no-​right) concerns only one activity of one person.’21 19 On the normative significance of correlativity for Hohfeldian powers see C Essert, ‘Legal Powers in Private Law’ (2015) 21 Leg Th 136. 20 Simmonds (n 11) 146–​58. 21 JM Finnis, ‘Some Professorial Fallacies about Rights (1972) 4 Adelaide LR 377, 379. Finnis goes on to explain: There is one complete Hohfeldian relation between X and Y (in Hohfeld’s example) with respect to the topic of going (or not going) on X’s land. There is another complete and logically independent relationship between X and Y with respect to the topic of X’s walking on his own

Rights  37 What is significant is the logical independence of each different kind of relation (and of the activity pertinent to it), not the connection of any relation or any activity with any kind of object. Thus, whereas Kant attempted to articulate the rights that are most comprehensive normative abstractions relevant to law, Hohfeld formulated the simplest and most irreducible components of legal analysis. Where the two converge is that from the Kantian standpoint, Hohfeld usefully differentiates the kinds of entitlements that can figure within a legal relationship in which right and obligation are correlative. The Hohfeldian typology also usefully suggests that the specific acts relevant to a Kantian account of private law figure not only in the conception of duty, which Kant explicitly mentions, but also in the other Hohfeldian pairs.22 This convergence is evident in Kant’s reference to the entitlements that the subjective rights authorize. A conspicuous example of this is the Hohfeldian character of the entitlements that Kant mentions as involved in the innate right.23 Recall that the innate right, that is, the right to be independent of the constraint of others, is the subjective right inherent in Right’s objective order of equal reciprocal freedom. The paradigmatic juridical marker of this independence is the right to physical integrity. Inasmuch as this right imposes the duty that another should not perform any act that infringes it, the right to physical integrity is a Hohfeldian claim-​right. In addition, other entitlements that Kant discerns in the innate right engage other Hohfeldian categories. One of these is the authorization (Kant’s term for entitlement) to communicate one’s thoughts, provided that this does not injure the rights of others—​a Hohfeldian liberty. Another is the entitlement to be considered beyond reproach in the absence of any act affecting the rights of others. This entitlement, which adumbrates the presumption of innocence and allocates the burden of proof to claimants rather than defendants, is a Hohfeldian immunity from unfair legal processes. Yet another, which Kant calls innate equality, is a Hohfeldian immunity from being bound by others more than one can in turn bind them—​an entitlement that precludes legal arrangements that subordinate of one member of a legal relationship to another. Finally, the innate right (like all rights) carries with it the authorization to use coercion against anyone who violates it.24 Within a functioning legal system the authorization to coerce gives the right-​holder a Hohfeldian power to initiate the state’s coercive proceedings for the vindication of one’s rights. Thus, every form of Hohfeldian jural advantage—​claim-​right, liberty, power, and immunity—​and

land. There is yet another and again logically independent relationship between X and Y with respect to the topic of Y’s interfering with X’s walking on his (X’s) own land. And so on. 22 Kant’s references to the familiar ideas of modern jurisprudence are often shrouded in the obscurity of his own vocabulary. For example, he ascribes the power to put others under obligation through acquisition to what he calls a ‘permissive law’ [6:247]. On this complex notion see J Weinrib, ‘Permissive Laws and the Dynamism of Kantian Justice’ (2014) 33 L and Phil 105. 23 Kant (n 3) 33–​35 [6:637–​38]. 24 ibid 27–​28 [6:231–​32].

38  Ernest J. Weinrib consequently every form of correlative disadvantage—​duty, no-​right, liability, and disability—​figures in Kant’s description of the entitlements present in the innate right. This is not surprising. If Hohfeld indeed succeeded in formulating the irreducible elements of legal analysis, then his categories should of course be evident within the operation of the Kantian rights. Kant’s mention of entitlements allows us to locate the Hohfeldian categories within Kant’s exposition. Far from being in tension with the Kantian conception of rights, the Hohfeldian typology sets out the different kinds of entitlements that make those rights more determinate. Within the legal relationship constituted by one person’s right and another person’s corresponding obligation, the Hohfeldian categories specify different aspects of the right.25 A claim-​right is a right considered from the standpoint of the specific act or forbearance required of another. A liberty is a right considered from the standpoint of the specific activity that the right authorizes. A power is a right considered from the standpoint of the legal effect that its exercise has on another. An immunity is a right considered from the standpoint of the absence of legal effect from the activity of someone else. For example, because it sets the terms of interaction between persons, the ownership of property exemplifies a relationship constituted by one person’s right and another’s corresponding obligation. Ownership in turn gives rise to entitlements of every Hohfeldian kind that deal with specific aspects of it.26 My ownership of a particular horse, for instance, gives me a claim-​right that subjects another to a duty not to perform acts inconsistent with my right to the horse, say, by using the horse without my consent. I also have a liberty to use the horse as I wish; correlatively, no-​ one has the right to prevent my use. Moreover, I have a power to alienate (or rather, to participate in the process of alienating) the horse, thereby extinguishing one set of jural relations and creating a new and similar set in another. I also have an immunity against at least some of the legal effects of another person’s dealing with the horse; the unauthorized sale of my horse will not, for example, except under limited circumstances such as market overt, divest me of my ownership. And if someone does indeed infringe my claim-​right, I have a power to initiate a court process aimed at compelling that person to repair the injury to my right as an owner. If I am successful in this process, the outcome puts the defendant under a duty to restore the horse to me or to pay me a liquidated amount in damages. The same nexus of right and obligation, the vinculum iuris between the parties, is continuous through all these instantiations of the Hohfeldian categories, and is reflected in the correlative application of whatever category pertains to any aspect of the parties’ relationship. Conversely, by focusing on the different ways in which I am related to

25 In sketching the role that the Hohfeldian categories have within this relationship, I am not claiming that they have no role outside it. 26 WN Hohfeld, ‘Fundamental Legal Conceptions as Applied in Judicial Reasoning’ (1917) 26 Yale LJ 710, 746–​47.

Rights  39 another through my property right in the horse, the various Hohfeldian categories concretize what it means for me to own this horse. The Kantian conception of right and corresponding obligation normatively unifies the various Hohfeldian categories present in a given legal relationship. Relative to the Hohfeldian categories, the Kantian conception thus has two functions, a normative one and a unifying one. The two functions are interconnected because, on a Kantian view, the unity must make normative sense and the normativity must be appropriate to the unity of the parties’ relationship. I briefly consider the two functions in turn. First, the normative function of the Kantian conception can be contrasted to the exclusively analytic significance of the Hohfeldian categories. Hohfeld’s signal contribution was to clarify the correlatives properly applicable to different kinds of legal entitlements. In particular, as he pointed out, a liberty might indiscriminately be referred to as a right, but was not correlated to a duty on the part of anyone else; correlation to duties was restricted to claim-​rights.27 The positive law, to which the Hohfeldian scheme was neutral, determined the category into which any specific advantage fell. The role of the Hohfeldian scheme was merely, but importantly, to prevent confusion about the appropriate correlative. Hohfeld’s purpose, accordingly, was analytic, not normative: he did not provide a reason for the law’s recognition of particular advantages or for its placing any advantage within any particular category of entitlement, but rather he set out the kind of correlativity required by its being the sort of advantage that the positive law made it. An indicator of the analytic nature of his enterprise is his treatment of correlativity as the expression of descriptive equivalence, that is, of the sameness of the act when described in terms of either of the correlative poles.28 For Hohfeld, no category exercises the normative function of grounding its correlative; rather, within every correlative pairing each category refers to the legal significance of same act or activity viewed from the other side of the relationship.29 Thus, when Hohfeld approvingly quotes the statement that ‘[w]‌hen a right is invaded, a duty is violated’,30 he does not mean that the claim-​right was the basis of the duty, but that the claim-​right consisted in the entitlement to have the duty performed, just as the duty consisted in the requirement not to violate the claim-​right. In the Kantian conception, in contrast, right operates normatively. To have a right to something is to be legally so connected, either innately or through acquisition, to that thing, that others are bound not to interfere with it. What determines the nature of the parties’ relationship as a whole is the kind of right that the right-​ holder has, because the right sets the terms on which the right-​holder’s freedom 27 Hohfeld (n 18) 28–​44. 28 ibid 32: ‘If X has a right against Y that he shall stay of the former’s land, the correlative (and equivalent) is that Y is under a duty toward X to stay off the place’ (emphasis added). 29 ibid 34. 30 ibid 32.

40  Ernest J. Weinrib co-​exists with the action of others. The right thus marks the terms of the obligation to which the obligor’s action must conform. Correlativity in this context expresses the logic of grounding and not merely of equivalence: the connection of the right-​ holder to the object of the right puts others under an obligation to the right-​holder with respect to that object.31 And although the entitlements and the duties that flow from a right and its correlative obligation operate analytically as equivalents in the Hohfeldian manner, the entitlements that specify the right are normatively prior to the duties that specify the obligation. The right is thus the reason for the obligation; conversely, the obligation is correlative to it when it exists by virtue of the right, is owed to the holder of the right, is directed to the object of the right, and is limited to the maintenance of the right. Accordingly, the obligation is not merely the equivalent of the right described from the viewpoint of the other party. Rather, it is the normative entailment of the right, which grounds it and demarcates its scope. Next, relative to the Hohfeldian categories, the Kantian conception of a right has a unifying function. Hohfeld’s categories are analytically distinct, and each of them constitutes its own peculiar jural relation. The question then arises: in the light of these separate jural relations, how is the overall relationship between the parties to be understood? Within the analytical context of Hohfeld’s categories, the only possible answer is that the overall relationship between the parties is simply an aggregation of all the applicable jural relations. Thus, the eminent contracts scholar Arthur Corbin, a colleague and follower of Hohfeld, after defining contract in Hohfeldian terms as ‘the legal relations . . . consisting of a right or rights in personam and their corresponding duties accompanied by certain powers, privileges, and immunities’, noted that ‘[t]‌he sum of these legal relations is often called “obligation” ’.32 Corbin’s comment acknowledges the distinction, at least as a matter of common legal parlance, between the specific duties that a contract imposes and the notion of an overall obligation that comprehends but is not restricted to those duties. However, this notion of obligation is unimportant for his purposes; obligation is simply the collection of all the Hohfeldian categories relevant to the contractual relationship of promisor and promisee. An obligation for him is merely the sum of its Hohfeldian parts. In this he is not mistaken from a Hohfeldian standpoint. For if Hohfeld’s categories indeed constitute the irreducible elements of legal analysis, then the overarching notion of an obligation must necessarily refer to an aggregate of those categories. 31 Kant (n 3) 33 [6:237]. 32 AL Corbin (ed), Anson’s Principles of the Law of Contract (3rd American edn, 1919) 13. Similarly, Hohfeld described ownership of land as ‘a complex aggregate of rights (or claims), privileges, powers, and immunities’. Hohfeld (n 26) 746. In a revealing footnote (ibid 710), Hohfeld observed that one of the chief purposes of his analysis of jural relations was ‘to establish a firm foundation for the analysis and discussion of complex jural relations, or aggregates of jural relations’ for which ‘a more adequate treatment must be reserved for another occasion’. Presumably, his premature death prevented him from working out this more adequate treatment.

Rights  41 The Kantian conception of a right goes beyond Hohfeldian analysis to the normative unity of the parties’ relationship. The fact that the Hohfeldian categories are analytically distinct does not mean that they are normatively autonomous. In the example of owning the horse, for instance, the normative connection between the owner’s liberty to use the horse, the immunity of the owner’s right from the legal effects of others’ dealings with the horse, the duty on others to abstain from using the horse, and the owner’s power to sue for violations of that duty is not merely incidental. If any of these aspects existed without the others, the idea of ownership would be unrecognizable. For inasmuch as the exclusivity of the owner’s right to use the horse lies at the root of ownership, the owner’s liberty is immediately intertwined with the owner’s claim-​right. The owner’s liberty entitles the owner to use the horse, but does not in itself exclude anyone; conversely, the owner’s claim-​right signifies that others are under a duty not to use the horse but makes no reference to owner’s own use. Ownership brings the liberty and the claim-​right together in a distinctive normative idea that relates owner and non-​owner through the exclusivity of the owner’s control.33 Similarly, the exclusivity of ownership would be nugatory if the horse could routinely be sold out from under the owner, or if the owner had no power to seek legal recourse for the breach of the duty. As the ownership example illustrates, the Hohfeldian categories that pertain to a given right, when taken together, are ‘structured in a complex whole’.34 The Kantian conception of a right reflects the unity of this whole. Whereas the Hohfeldian categories considered on their own decompose the parties’ relationship into analytic elements each with its own particular mode of correlativity, the Kantian conception synthesizes them into a unified ensemble animated by the parties’ reciprocal freedom.35

4.  The Contrast with the Will Theory of Rights In locating the normative character of rights not in the welfare or interests of the right-​holder but in a Kantian conception of the purposiveness of interaction, the 33 On the Kantian approach to ownership see further ch 3 below. 34 C Wellman, A Theory of Rights (Roman and Allenheld 1985) 59. 35 The operation of the Hohfeldian categories within the framework of the Kantian subjective rights bears on recent discussions of corrective justice. Some have identified corrective justice with the position that on committing a wrong to the plaintiff the defendant has a duty to compensate. The criticism is then offered that in Hohfeldian terms the positive law does not in fact place the wrongdoer under such a duty. Rather, the victim of a wrong has a power to sue, correlative to which the wrongdoer is under a liability, not a duty. BC Zipursky, ‘Civil Recourse, Not Corrective Justice’ (2003) 91 Georgetown LJ 695, 720; NB Oman, Why There is No Duty to Pay Damages (2011) 39 Florida St LR 43, 57. Even on the dubious assumption that this is a serviceable characterization of corrective justice (corrective justice in my view is restricted to neither wrongdoing nor compensation), the criticism is built on sand. Corrective justice posits the correlativity not of a specific Hohfeldian pair of concepts, but of the overall structure of the parties’ relationship as one of right and obligation. Right and obligation are not themselves Hohfeldian concepts. Rather, they supply the normative principle for whatever Hohfeldian concepts are applicable to the private law relationship.

42  Ernest J. Weinrib corrective justice approach echoes the theme of a persistent contemporary controversy. Over the last decades, legal philosophers have discussed the comparative merits of the interest theory of rights and the will theory of rights.36 To simplify a complex and sophisticated debate, one may say that interest theorists hold that a right protects an interest significant enough to place others under a duty with respect to it, whereas will theorists maintain that a right makes the duty correlative to it subject to the will of the right-​holder. For interest theorists the function of rights is to protect the right-​holder’s interest; for will theorists the function of rights is to protect the right-​holder’s domain of choice. Obviously, the Kantian rights of corrective justice are more closely aligned to the will theory than to the interest theory. Conceiving of rights in terms of interests is incompatible with corrective justice, because, as noted at the outset of this chapter, interests refer to one party without correlatively implicating another. In contrast, the emphasis on the right-​holder’s choice in the will theory is more consonant with corrective justice, where on the Kantian approach, rights reflect the relationship of will to will within a regime of equal reciprocal freedom. This similarity between the will theory and corrective justice is not surprising in view of the acknowledged Kantian inspiration for the will theory.37 Nonetheless, the will theory and corrective justice have different theoretical aims. The will theory, like its rival the interest theory, is an analytic enterprise that attempts to articulate the necessary and sufficient conditions of rights.38 The goal of these theories is to provide a normatively neutral definition of rights that is faithful to our usage of the term and yet is serviceable in discussions of the rights we ought to have.39 Corrective justice, in contrast, is a normative theory that exhibits the meaning of fairness and coherence in the relationships of private law. The issue that arises for corrective justice—​how can the normative character of rights be conceived in a manner consistent with the correlativity of those relationships?—​concerns not the necessary and sufficient conditions of a right but rather the operation of rights as the normative markers of reasons for liability. The Kantian gloss on corrective justice addresses this issue by situating the rights of private law within a system of concepts, legal doctrines, and institutional arrangements that treats the action of one person as compatible with the freedom of another. If the will theory is the effort to locate the necessary and sufficient conditions of rights in the right-​holder’s will, then Kant, despite the frequent reference to will and choice in his Doctrine of Right, is not a will theorist. His aim is relentlessly normative, not analytic. He makes no attempt to tell us what a right is independently of the role that rights play in the realization of the equal reciprocal freedom of all. 36 Kramer, Simmonds, and Steiner (n 10). 37 HLA Hart, ‘Are There any Natural Rights?’ (1955) 64 Phil Rev 175. 38 HLA Hart, Essays on Bentham (Clarendon Press 1982) 188. 39 H Stewart, ‘The Definition of a Right’ (2012) 3 Juris 319, 320. Stewart himself has doubts about the usefulness of this project: see 335–​38.

Rights  43 Even the sole seemingly analytic point that Kant makes—​his clumsy characterization of rights as ‘(moral) capacities for putting others under obligations (i.e., as a lawful basis, titulum, for doing so’40—​is not a normatively neutral definition but rather an elaboration of Kant’s theme that rights involve the distinctive kind of obligation for which an external constraint is possible. This difference between the will theory and the Kantian account becomes salient with regard to the status of incompetent persons. An apparent consequence of the will theory is that by making choice the defining feature of rights, it implies that rights are unavailable to those (such as infants and comatose adults) who are incapable of exercising choice.41 This is usually regarded as a difficulty, especially because sophisticated legal systems treat incompetence not as disqualifying incompetent persons from having rights but merely as occasioning special arrangements for managing their rights. In contrast, in Kant’s account a child is born with an innate right (that is, one not established by an act of the child) to the care of its parents. This right carries with it an entitlement not only to be fed and cared for, but also to be educated and developed into an independent person.42 The fact that every child has a period of complete incompetence evidently does not disqualify the child from holding this right throughout childhood. To the contrary, the necessity for going through a process of development from total dependence to mature independence is precisely what justifies the child’s having the right to the care of his parents, who brought him into the world, as Kant puts it, at their initiative and without his consent.43 As is apparent from their divergent treatment of incompetent persons, the will theory and the Kantian approach assign different roles to the operation of the will. Emblematic of the will theory is the notion that the duty correlative to the right is within the control of the right-​holder, who can then exercise his or her choice to enforce or to waive it. Being in a position to exercise this control is what signifies the existence of the right. Conversely, the physical inability to exercise such control or to make such a choice suggests that the incompetent person has no right. For Kant, in contrast, the right-​holder’s control of the correlative obligation reflects the juridical significance of the right as such; it has nothing to do with the physical capacity of the right-​holder. The obligation incumbent on the obligor is correlative to the right, not to the right-​holder’s competence to exercise that right. How a 40 Kant (n 3) 33 [6:237]. 41 For a response to this difficulty by a leading will theorist see H Steiner, ‘Working Rights’ in MH Kramer, NE Simmonds, and H Steiner, A Debate Over Rights (OUP 2000) 259–​61. Steiner acknowledges that infants would indeed have no rights under the will theory, but denies that this has any practical or analytic significance, provided that infants are owed duties of protection, and that the rights correlative to these duties are held by someone who is competent to exercise them. This, however, would seem to imply the significant consequence from the standpoint of private law, that a wrongdoer could breach a duty to an infant without thereby wronging, and therefore being liable to, the infant. 42 Kant (n 3) 70–​71 [6:280–​82]. 43 ibid 70 [6:281].

44  Ernest J. Weinrib legal system is to provide for the exercise of rights on behalf of those who cannot exercise them on their own is a subsequent issue that has no bearing on the rights’ existence. Kant’s concern is with the equal reciprocal freedom of the interacting parties. The questions that he addresses are: what does it mean for one person’s action to be consistent with another person’s freedom, and what legal categories and institutional arrangements are expressive of this consistency of action and freedom? Equal reciprocal freedom may require that the incompetent person have a right even though it cannot be personally exercised. The authority of parents to subject their child to their managerial actions, for example, would not be consistent with the child’s freedom unless children had a right to parental conduct directed towards their nurture and the development of their independence. In the absence of this right, parent and child would not be juridical equals; the dependence and vulnerability of children would expose them to parental absolutism or worse. Thus, whereas for the will theory the infant’s incompetence analytically nullifies the child’s right, for Kant the incompetence normatively necessitates it. Is Kant’s assumption that the incompetent person is not disqualified from being a bearer of rights consistent with the foundations of his thinking? In Kant’s account, at the heart of the concept of Right—​that is, at the heart of the most general characterization of law as a system of rights—​lies the capacity for purposiveness in the relation of will to will. Kant calls this capacity ‘the right of humanity in our own person’.44 Humanity, in contrast to animality, signifies our capacity, as inhabitants of the natural world along with and in distinction from other animals, to set pragmatic and moral ends for oneself.45 Humanity is described as a right because it figures in the juridical relationship of one person to another. One might suppose that by associating the concept of Right with this capacity Kant should be committed to the notion that rights are inapplicable to those who lack the capacity. That supposition, however, would be mistaken. Humanity is a normative idea, not a psychological or physical one.46 It does not provide a criterion, rooted in a physical ability of some kind, for any particular person’s eligibility to hold rights. Instead, the idea of humanity signals the applicability to human beings of the systemic connection between rights and reciprocal freedom. This freedom is expressed through the wills of the interacting parties, that is, through the exercise of their purposiveness in the pursuit by each of his or her own ends, whatever they might be. Humanity is thus, to use a Rawlsian phrase,

44 Kant (n 3) 32 [6:236], 36 [6:240]. 45 The meaning of ‘humanity’ is much discussed in treatments of the second formulation (the formula of humanity) of the categorical imperative in Kant’s Groundwork for the Metaphysics of Moral; see H Allison, Kant’s Groundwork for the Metaphysics of Morals: A Commentary (OUP 2011) 209–​18; J Rawls, Lectures on the History of Moral Philosophy (Harvard UP 2000) 187–​90. 46 Compare J Rawls, Political Liberalism (Columbia UP 1993) 86–​88 on the moral psychology of the person within the conception of justice as fairness.

Rights  45 the moral power that defines the conception of the person implicit within a system of rights.47 Humanity so conceived is not concerned with the particularities of any specific person’s will. Its focus, rather, is on the significance of the will as such for the normative character of a system of rights as such.48 The will’s freedom figures here not because the ability to exercise it at a given moment or over a given period of time preconditions anyone’s right-​holding status, but because without it the system of rights is not normatively intelligible. Understood as a moral power, humanity is generic to human beings. The relation between humanity and human beings is this: humanity is the human being understood from the standpoint of the capacity for freedom that normatively underpins the system of rights; the human being is the physical embodiment of that capacity in the natural world.49 Just as humanity marks the applicability to human beings of the normative connection between the will as such (not any particular wills) and rights as such (not any particular rights), so the system of rights that humanity grounds is available to human beings as such and not merely to those who have particular abilities or competences. And just as the notion of humanity provides the general condition for the possibility of rights, so human beings constitute the correspondingly general conception of the possible holders of rights. The Kantian approach makes the individual capacities of specific persons irrelevant to their eligibility to be right-​holders or to the definition of a right. Instead, it identifies the category of beings characterized by the possibility of self-​ determining agency, and thus by the possibility of interacting with one another on the basis of their reciprocal freedom. This possibility is present to all human beings as a result of the very fact that they are human beings, although the competence to exercise one’s freedom does not exist for each human being individually at all times.50 Accordingly, all human beings have the standing to rank as holders of rights. They retain this standing regardless of whether at any particular time an individual is an infant or an adult, asleep or awake, impaired or unimpaired, competent or incompetent. Thus, despite the apparent affinity between the will theory and the Kantian gloss on corrective justice, significant differences exist between them. The will theory aims to connect the right-​holder’s control of a right to the necessary and sufficient conditions for the existence of a right, whereas the Kantian approach is concerned 47 For the relationship between Kant’s idea of the right of humanity and Rawls’s two moral powers see A Ripstein, ‘Private Order and Public Justice: Kant and Rawls’ (2006) 92 Va LR 1391, 1399. 48 Compare the structurally similar observation that in the first Critique; Kant is not concerned with ‘whether we can have this or that bit of knowledge’, but with the question of ‘[w]‌hat is the relation between the general form of what is and the general form of knowledge’. J Conant, ‘Why Kant is Not a Kantian’ (2016) 44 Philosophical Topics 75, 83. 49 Kant (n 3) 35 [6:239]. 50 As Kant puts it, the freedom from constraint by another’s choice is the innate right that ‘belongs to every human being by virtue of his humanity’; ibid 34 [6:237]. Compare also ibid 140 [6:363] (respect for humanity is ‘respect for the species’).

46  Ernest J. Weinrib with the normative significance of rights as markers of equal reciprocal freedom. The will theory ties the existence of rights to the exercise of one’s will and therefore faces the challenge of accounting for, or dismissing, the rights of those who are incompetent to exercise their rights, whereas the Kantian approach regards human beings, whatever their competence, as nonetheless eligible to hold rights.

5.  Conclusion Being foundational to contemporary law, the idea of rights has over the last century justifiably generated a vast and complex philosophical literature. This literature has focused on two tasks. The first, associated primarily with the achievement of Hohfeld, has been to work out and apply a typology of the various kinds of legal entitlements as well as of their correlative disentitlements and their opposites. The other, giving rise most prominently to the interest theory and the will theory, has been to formulate the necessary and sufficient conditions for rights. This chapter has dealt with the connection of corrective justice to these issues while stressing the normative role of rights within the corrective justice approach. The starting point of my argument has been the contention that, if they are to be adequate to the correlative structure of the parties’ relationship, the rights of private law have to be conceived in Kantian terms. The Kantian conception of rights thus carries forward the normative thrust of corrective justice by presenting the private law relationship as the expression of equal reciprocal freedom. When actualized within a legal system, equal reciprocal freedom manifests itself through different levels of specificity—​from the most general level of the objective order of value, to the various subjective rights and their correlative obligations, and then to the specific entitlements that converge with the elements of the Hohfeldian typology. Moreover, the Kantian gloss on corrective justice differs from the will theory in being concerned not with a right’s necessary and sufficient conditions but with the functioning of rights as normative components of the practical reason characteristic of private law. Understood in this way, rights are essential to the coherent expression of human freedom in one person’s legal relationship with another.

3

Ownership 1.  Use and Exclusivity In the preceding chapter, I suggested that, under its Kantian gloss, corrective justice integrates the various Hohfeldian entitlements into a unified conception of a right. Ownership, for example, brings together the owner’s liberty to use the owned object and the owner’s claim-​right to exclude others from it. As a distinctive normative idea, ownership relates owner and non-​owner through the exclusivity of the owner’s control over what is owned. Within this general conception of ownership as a right, the owner’s liberty to use and the claim-​right against uses by others are the Hohfeldian entitlements governing the specific actions that pertain to what is owned. Indeed, without this combination of use and exclusivity, the very idea of ownership would be unrecognizable. This positioning of Hohfeldian entitlements within the Kantian conception of a right raises issues of its own. It is one thing to suggest that a right conceived in Kantian terms integrates the relevant Hohfeldian entitlements; it is another to show the grounding of this integration with respect to a given right. Ownership is a case in point. Although it is clear that ownership combines use and exclusivity, the relationship between the two and the terms on which they come together remain a matter of continuing dispute among property theorists. Among the questions raised in this debate are: Does the right to exclude function instrumentally as a rough and ready proxy for promoting the owner’s interest in a thing’s use?1 Or does the distinctiveness of ownership consist in the right to exclude, with the power to determine the thing’s use being merely the consequence of the fact that no-​one else is in a position to interfere with what the owner decides?2 Or is the notion of use conceptually independent of the idea of ownership as such?3 Or, conversely, has the idea of exclusion itself been overemphasized, in view of the permitted uses by non-​owners that one finds in contemporary legal systems?4 Taking up the problematic connection between use and exclusivity, this chapter sets out the conception of ownership that fully integrates them within a regime of reciprocal freedom.

1 HE Smith, ‘Property as the Law of Things’ (2012) 125 Harv LR 1691. 2 TW Merrill, ‘Property and the Right to Exclude’ (1998) 77 Neb LR 730, 741. 3 S Douglas and B McFarlane, ‘Defining Property Rights’ in J Penner and HE Smith (eds), Philosophical Foundations of Property Law (OUP 2013) 219. 4 H Dagan, Property: Values and Institutions (OUP 2011) ch 2.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0003

48  Ernest J. Weinrib Broadly speaking, one can conceptualize the relationship between use and exclusivity in two ways.5 The first way can be termed ‘separatist’, for it treats the two as separate notions that the idea of ownership conjoins. Use and exclusivity are regarded as conceptually independent of each other, coming together within the phenomenon of property through historical happenstance or because of the goal that this conjunction serves. The well-​known image of property as a ‘bundle of rights’6 is an example of the separatist approach: each stick in the bundle is separately significant, with the bundle itself being a collection of components that can be shuffled around or combined in various ways to produce different social effects. The second way can be termed ‘integrative’, because it views use and exclusivity as integrated aspects of ownership. On the integrative approach, neither the right to use nor the right to exclude is normatively intelligible apart from the idea of ownership that they jointly constitute. Each of them contributes to the meaning of ownership as a unified idea, and ownership in turn gives meaning to each of them as its constituent aspects. The normative character of the right to exclude and the normative character of the right to use are not independent of each other. Rather, each is intrinsic to a unified account of the basis of ownership. Nor are they brought together for instrumental reasons to promote a separately justifiable goal. Instead, they belong together as the mutually interconnected aspects of a non-​instrumental conception of ownership. This chapter deals with Kant’s paradigmatically integrative account of the relationship between use and exclusivity.7 For Kant, use and exclusivity are not merely different sticks in a bundle of incidents. They are the reciprocally conditioned aspects of ownership as a right within a regime of reciprocal freedom. On the one hand, the right to use property presupposes the exclusive right of owners to have what they own available for their use. On the other hand, the right to exclude arises out of an inquiry into the conditions under which persons can be legally related to one another through a thing’s usability. The Kantian account differs from the standard contemporary approaches to property. Accounts of property are thought to be either nominalist or essentialist.8 A nominalist account treats property as a purely conventional label that, having no fixed meaning, can be attached to whatever the legal system wishes. An essentialist account, in contrast, seeks to disclose the necessary and sufficient conditions that make up the irreducible core of property. Both kinds of account seek, in their different ways, to identify what property is. Kant’s account is neither nominalist nor 5 The comments that follow apply the distinction between accidental and intrinsic unity to the idea of ownership. On that distinction see EJ Weinrib, The Idea of Private Law (HarvardUP 1995, OUP 2012) 33–​36. 6 DB Klein and J Robinson, ‘Property: A Bundle of Rights? Prologue to the Property Symposium’ (2011) 8 Econ J Watch 193. 7 See also the illuminating treatment in A Ripstein, ‘Possession and Use’ in J Penner and HE Smith (eds), Philosophical Foundations of Property Law (OUP 2013) 156. 8 Merrill (n 2) 734–​39.

Ownership  49 essentialist. He is not interested in determining when the property label can legitimately be attached. And just as he does not articulate the necessary and sufficient conditions of rights generally,9 so he does not specify the necessary and sufficient conditions for regarding something as property. Rather, his focus on the status and implications of ownership as a rightful legal category. His account is normative: by situating ownership within the conceptual and institutional framework of the reciprocal freedom of all, it elucidates why persons have a right to make things their own and thereby to impose correlative legal obligations on others. This chapter works its way towards Kant’s integrative account by first discussing, in section 2, an example of the separatist approach. This is the example of Grotius, who (in common with many early modern thinkers) postulated the right to use as the original right from which the fuller conception of property characterized by the right to exclude evolved. I will then in section 3 invoke Kant to undermine the viability of Grotius’ original right to use. Finally, I will present Kant’s integrative account in two stages, dealing first with his conception of ownership (section 4) and then with the relationship between ownership so conceived and acquisition (section 5). The concluding section will sketch the implications of the Kantian approach for a theory of property and then relate the present chapter to the two that follow.

2.  Grotius’ Account Grotius’ account of property—​sometimes considered the first modern theory of property10—​treats the right to use as the root of the comprehensive notion of ownership to which the right to exclude belongs. The right to use is ownership’s paramount aspect, even to the extent of allowing it under certain circumstances to trump the restrictions of exclusion. Grotius asks us to imagine a primeval world in which everyone has an original right to use things. Starting with this primeval world, Grotius describes a process of evolution from the common possession of all to a private right of property in which owners have a right to exclude. By common possession, Grotius means not that everything was owned by everybody, but that nothing was owned by anybody, so that everyone had a right to use anything. In this original and universal right to use, ‘each man at once could take whatever he wished for his own uses, and consume whatever was capable of being consumed’.11 In Samuel Pufendorf ’s subsequent formulation, ‘all things lay open to all men, and belonged no more to one

9 See ch 2, s 4. 10 A Mossoff, ‘What Is Property? Putting the Pieces Back Together’ (2003) 45 Ariz LR 371, 379. 11 H Grotius, De Jure Belli ac Pacis Libri Tres, vol 2 (FW Kelsey tr, Clarendon Press 1925) 186. I have modified the translation slightly.

50  Ernest J. Weinrib than to another’.12 Over time and in response to various historical developments—​ the dispersal of mankind, the desire for a higher standard of living, the lack of the justice and kindness required for a fair division—​everyone agreed that common possession was to be superseded by the right of each person to have as his own what each had occupied. Consequently, a new form of right arose: the private property of individuals. Except under very particular circumstances such as necessity, private property carried with it the owner’s right to exclude others. Thus, Grotius’ account traces a progression from communio to dominium. At the beginning of this progression, everyone is entitled to use everything. The allocation of this right is as broad as possible, in that everyone has it; but the right itself is as thin as possible, in that all that anyone has is a right to use. The evolution of private property symmetrically transforms both the right’s allocation and its content. The allocation of the right contracts to the particular persons who own particular things, but the right itself becomes as thick as possible, because owners can now exclude others from what they own. In Grotius’ account, the right to use exists prior to and independently of the right to exclude, which is added with the advent of private property. Grotius postulates an apparently straightforward distinction between the primeval right to use and the later right to exclude. However, in one situation this distinction becomes blurred. In the primeval society, a question can arise concerning the effect on others of one person’s taking something for one’s own use. Let us suppose that in the primeval society you pick an apple from a tree and are about to take a bite, when I snatch the apple from your hand and consume it. Am I unjustly interfering with your right to use, or am I merely exercising my own equivalent right? On the one hand, a right to use that binds no-​one else and can be so easily frustrated may appear to be no right at all. On the other hand, the status of the apple as being open to all and belonging no more to one person than to another does not change because you have the apple in your hand. In his early treatment of this issue, Grotius invokes two complementary laws of nature, one of which applies to a person’s own activity of using, and the other to a person’s relation to the similar activity of others. The first authorizes the original and universal right to use: ‘It shall be permissible to attach to oneself (adiungere sibi) and to retain those things that are useful for life.’13 The basis of this law of nature is that, because God gave the world to humans generally and not to individuals, things can be put to use only through individual possessory acts. This individually self-​regarding law of nature enshrines a Hohfeldian liberty. The effect of

12 S Pufendorf, De Jure Naturae et Gentium Libri Octo (CH Oldfather and WA Oldfather trs, Clarendon Press 1934) 537. 13 H Grotius, Commentary on the Law of Prize and Booty (MJ van Ittersum ed, Liberty Fund 2006) 23 (translation modified).

Ownership  51 such a liberty is that you are under no duty to me to abstain from picking the apple. However, because one person’s liberty is not correlative to another’s duty, your act of picking the apple from the tree does not obligate me to anything. Considered on its own, then, the first law is indifferent to whether I in turn exercise my liberty by picking my own apple or by snatching from you the apple that you have picked. Grotius’ explanation of the right to use confirms its status as a Hohfeldian liberty. He remarks that ‘this attribute the Scholastics chose to describe as a concept of fact but not of law,’ and he goes on to distinguish it from occupation in developed law, where ‘ “use” carries with it a privative force with respect to others’.14 If grabbing the apple from another is wrongful, then use indeed has the ‘privative force with respect to others’ that Grotius denies. The permissibility of grabbing the apple from another is also implied by his characterization of use as a concept of fact but not of law. That terminology refers back to the Franciscan controversy. The Franciscans, striving to re-​enact in their lives the absolute poverty of Christ and the apostles, renounced property.15 Their opponents charged that this renunciation could not be effective, because a person who renounced property could not continue to make use even of things necessary for one’s survival. In response, the Franciscans insisted that use did not imply a property right in the thing used. Eating an apple, for instance, was merely a ‘simple use of fact’;16 that is, an act that was licit but unconnected to any assertion of right. If the Grotian universal right to use was an instance of the Franciscan use of fact, then picking the apple from the tree exemplified only a Hohfeldian liberty, as that was all that Franciscan use could be.17 Accordingly, being deprived of that apple by another was not an injustice. In contrast, a second law of nature prohibits my seizing the apple from you. It stipulates: ‘Let no one seize what has been seized by another.’18The basis of this law is that it ‘circumscribes [the previous one] within a just limit’,19 as human welfare requires the mutual harmony that consists in each person’s abstention from what others are holding. Thus, at the end of the day Grotius allows for both the permissibility of taking things for one’s use and the impermissibility of taking from another. Summing up his position in his mature work, he asserts that ‘[t]‌he enjoyment of

14 ibid 315–​16 (translation modified). 15 On this controversy see B Tierney, The Idea of Natural Rights (Scholars Press 1997) 93–​203; MD Lambert, Franciscan Poverty: The Doctrine of the Absolute Poverty of Christ and the Apostles in the Franciscan Order 1210-​1323 (Franciscan Institute 1998) 134; MJ Schermaier, ‘Dominus Actorum Suorum: Die Willenstheoretische Begründung des Eigentums und das Römische Recht’ (2017) 134 ZRG Germ Abt 49, 65–​76. 16 William of Ockham, ‘The Work of Ninety Days’ in William of Ockham, A Letter to the Friars Minor and Other Writings (AS McGrade and J Kilcullen eds, CUP 1995) 22; J Robinson, William of Ockham’s Early Theory of Property in Context (Brill 2013) 217–​25. 17 B Tierney, ‘Hohfeld on Ockham: A Canonistic Text in the Opus nonaginta dierum’ in WP Müller and M Summar (eds), Medieval Church Law and the Origins of the Western Legal Tradition (Catholic University of America Press 2006) 365. 18 Grotius (n 13) 27 (my translation). 19 ibid (my translation).

52  Ernest J. Weinrib this universal right was then in place of private ownership (vice proprietatis); for whatever each had thus seized for himself, another could not snatch away from him except by an unjust act’.20 The artful formulation ‘in place of private ownership’ (vice proprietatis) professes its consistency with the role of use in the primeval society (the right to use is not proprietas but only vice proprietatis), while importing the duty that proprietas imposes, to abstain from what others have. The second law of nature thus creates a grey zone between the communio of the right to use and the dominium of the right to exclude. Its effect is to introduce some measure of exclusivity into a pre-​proprietary context that is otherwise free of it. Within Grotius’ primeval society, picking an apple from a tree does not count as an appropriation that thereby bars others from the appropriated object; appropriation does not yet exist. It is merely the exercise of a liberty rooted in the Franciscan notion of use of fact. Qua liberty, its correlate under Hohfeldian analysis is not a duty but the absence of a claim-​right to preclude the liberty’s exercise.21 Nonetheless, Grotius’ second law of nature matches the liberty to use with a duty on others to abstain from the thing being used. Thus, even before the inception of a regime of property, the right to exclude characteristic of such a regime is operative in conjunction with that liberty. To that extent, the priority and independence of the right to use are compromised.22 Nor is the scope of the liberty itself clear. When the first law of nature authorizes the ‘attaching to oneself ’ and the ‘retaining’ of an apple, does it refer to two separate 20 Grotius (n 11) 186 (translation modified). Although Grotius’ accounts of property in the two works differ in some respects, I am assuming that this statement in the De Jure Belli ac Pacis encapsulates the effects of the laws of nature in the De Jure Praedae; otherwise, the statement in the De Jure Belli ac Pacis would be unsupported. 21 WN Hohfeld, ‘Some Fundamental Legal Conceptions as Applied to Legal Reasoning’ (1913) 23 Yale LJ 16, 33, insisting that it was a mistake to treat a duty as correlative to a liberty. 22 This tension in Grotius’ account was noticed by his early critics, who alleged a contradiction between the non-​existence of private property in primeval society and the supposed injustice of taking what another has already taken. Pufendorf purported to refute this criticism by claiming that ‘whatever each man has seized for his own uses becomes proper to him by a previous pact’. See Pufendorf (n 12) 553; see also 537. This was certainly not Grotius’ view in the De Jure Praedae—​the relevant passages of which were unknown to Pufendorf—​where Grotius traced the injustice to a law of nature, not to a pact. More fundamentally, the refutation (which Pufendorf claimed was ‘comprehensible even by men of ordinary intelligence’) is not without difficulty. Pufendorf ’s refutation is part of his general position that ‘assuming an original equal faculty of men over things, it is impossible to conceive how the mere corporal act of one person can prejudice the faculty of others unless their consent is given, that is, unless the pact intervenes’ (ibid 539). However, it is hard to see how a pact (or any number of pacts) can create in rem rights unless they already exist. An in rem right involves an obligation to abstain from the owned object that is generally incumbent on every non-​owner of that object. A pact or contract, on the other hand, creates an obligation that flows from one specific person to another. No aggregate of specific obligations can add up to a general obligation. Even if everyone in the world subscribed to a contract with every owner in which each person was obligated not to interfere with what each owner owns, those obligations would be binding on them only as particular contracting parties. That the corresponding right was against the whole world would merely reflect the contingency of who had subscribed to the contract; it would not be a consequence of the proprietary nature of the right. If others subsequently came into the world (later generations, for instance), the contract would not be binding on them. Pufendorf ’s position, that proprietary rights can be created through contract, seems to involve a kind of normative alchemy.

Ownership  53 actions or to one continuing action? To put the question more concretely, if after taking the apple from the tree, you put it down intending to consume it at a future time, is that apple still reserved to you under the second law of nature because you retain it even though it is no longer attached to you? And if the latter is the case, does the perpetuation of my duty depend on your continuing with the use you originally intended, or is any foreseeable—​or even any possible—​use sufficient? What if, having seized apples, the use you want to make of them is to store them to trade for the pears that someone else might have seized? At this point, the distinction between the second law of nature in the primeval society and the right to exclude in the developed notion of property becomes difficult to discern. Grotius cryptically remarks about the second law of nature: ‘Hence the distinction of ownerships and that famous Mine and Thine.’23 This comment leaves uncertain the extent of the gap, if any, between communio and dominium. These questions arise from the looseness of the link between the two laws of nature. As their divergence from the typology of Hohfeldian correlatives indicates, the two laws are not intrinsically connected. Instead, Grotius presents them as directed to different ends, the first to one’s own good and the second to the good of others.24 In order to have the second law limit the first, he brings them together through an amorphous invocation of the mutual harmony required for human welfare. Conjoining the liberty authorized by the first law of nature with the duty imposed by the second, Grotius effectively conceives of the universal right to use as comprising a bundle of rights that is instrumentally ordered towards his notion of human welfare. Both of the components of this bundle, each deriving from a different law of nature, are then taken into the fully developed notion of private property. From the Kantian standpoint, Grotius’ account of the primeval society—​and, therefore, of the developed regime of property that succeeds it—​has the hallmarks and the deficiencies of a separatist explanation of ownership. First, it constructs the legal relationship between the person using the thing and a prospective interloper out of disconnected rather than integrated components. Secondly, it grounds rights to things in human welfare, a non-​correlatively structured notion, rather than in the correlatively structured idea of reciprocal freedom. And thirdly, it treats these rights instrumentally as means for promoting welfare rather than non-​instrumentally as constitutive of interaction on fair and equal terms. These three deficiencies are interrelated, because disconnected components can be combined only instrumentally and only in the service of non-​correlatively structured goals.25

23 Grotius (n 13) 27 (my translation): ‘[h]inc dominiorum distinctio celeberrimumque illud Meum et Tuum’. 24 Grotius (n 13) 27. 25 EJ Weinrib, Corrective Justice (OUP 2012) 302–​11.

54  Ernest J. Weinrib

3.  Is There an Independent Right to Use? Nonetheless, I want to suggest that Grotius was essentially correct in holding that one cannot snatch an apple out of another’s hand, but on a different (and Kantian) ground. This suggestion, however, is not friendly to Grotius, because it calls into question the very existence of the right to use as Grotius conceptualizes it. For Kant the reason that, in a world without property, I wrong you if I snatch the apple from your hand is that, by disturbing the disposition of your fingers around the apple, I infringe your innate right in your own person. Under this right you are entitled to the space that your body occupies.26 Correlatively, the apple’s attachment to your body while it is in your hand places me under an obligation not to subject you to the movement that my snatching it would cause you. So long as you hold the apple, it is shielded from my action; once you put it down, it becomes available to me. My seizing the apple from your grasp is, accordingly, a trespass to your person. It is not a trespass to your property, because ex hypothesi there is no property. The Kantian analogue to Grotius’ primeval community is a world in which juridical relationships reflect nothing but each person’s innate right. In this world, the co-​existence of one person’s action with another’s freedom requires only that no-​ one violate the innate right of another. If your grasp of the apple limits my action, it must be because your innate right in your body, and not any right that you have to the apple as property, makes it wrongful for me to take it. Property is absent both from Grotius’ primeval community and from Kant’s world of innate right. Kant’s version of the world without property, however, is considerably more extreme than Grotius’. If we think of property as including both a right to exclude and right to use, Grotius’ primeval society lacks property in the sense that, although it has the right to use, it has only a restricted version of the right to exclude, geared to an unspecified range of interferences with use. In contrast, Kant’s world of innate right lacks property in the sense that it lacks both the right to use and the right to exclude. The absence of a right to use does not mean that using something is impermissible. Kant has no difficulty, for instance, with a person’s eating apples in the world of innate right.27 The point, however, is that, although by eating apples one uses them, one does not thereby exercise a right to use them. For Kant, a right connects the freedom of one person to the action of another through the object of the right. In the case of the snatched apple, it is not the apple that is the object of the right but the apple-​holder’s bodily integrity. Only the disposition of the apple-​holder’s fingers around the apple is juridically significant. What these fingers are holding and what the apple-​holder is doing with what they are holding do not figure in the legal

26 27

I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 55 [6:262]. ibid 42 [6:248], 44 [6:250].

Ownership  55 relationship between the apple-​holder and the apple-​snatcher. When the apple is snatched, the holder’s legal complaint is not about the apple or its use. It is about the interference with the fingers. The difference between Grotius and Kant reflects their different theoretical aspirations. Grotius attempts to show how, starting from the thinnest conception of the legal relationship between human beings and things (namely, the right of humans to use them), a fuller conception involving the right to exclude others from their use might have evolved. He takes it for granted that persons can be legally connected to things. His interest is in whether this connection operates through a common right or through an individual right. This interest aligns with his conviction, in the context of the commercial activity of his day, that the open seas are always subject to a common right to use and can never become private property.28 Kant’s aim is more radical. He wants to disclose how it is possible for any legal relationship to connect persons through things. For him, an account of our reciprocal freedom has to bridge the gap that exists between the innate right of humanity in our own person and the acquired rights, such as the right of property. The reason that the innate right figures in legal relationships is straightforward: the innate right is the subjective right that corresponds to the concept of Right, in which the action of each co-​exists with the freedom of all. In a world of innate right, persons are related to each other directly. Relating persons through things, however, is another matter, and requires additional argument. From Kant’s point of view, Grotius’ account assumes what has to be shown, that one’s connection to a thing can have juridical significance. Grotius considers use to be the precursor of property: ‘from the initial seizure follows use and then property’.29 Given the aim of Grotius’ account, it is not surprising that he characterizes the original and universal right to use as standing in the place of property, and then sets out a process of incremental movement from the primeval use to the later notion of property. The universal right to use and the developed notion of property are thus the thinnest and thickest ways of making things juridically relevant to persons. So far as Kant is concerned, this relevance itself is what has to be established, not the movement from its thin to its thick form. Thus, in the world of innate right, no right to use things exists, because use adds nothing that is not already contained in a person’s innate right. Under innate right things can be used, but a thing does not count as the object of a right. Nor does use function as a ground for placing others under an obligation to abstain from using something that is already being used. Accordingly, in no way does the permissible use of a thing anticipate or stand in the place of property. From a Kantian standpoint, Grotius gets one thing right but a more important thing wrong. Insofar 28 Grotius (n 11) 186, 190. See generally P Borschberg, Hugo Grotius, the Portuguese and Free Trade in the East Indies (NUS Press 2011). 29 ‘Apprehensio . . . unde usus et mox dominium secutum est’; Grotius (n 13) 24.

56  Ernest J. Weinrib as use involves attaching the used thing to oneself, Grotius is correct to hold that interference by others with use is unjust. Such interference would infringe the user’s innate right. However, Grotius’ premise that property develops out of a pre-​ existing right to use that obligates others to abstain from what is being used is misconceived. On the Kantian approach, there is no such right. Kant’s treatment of innate right draws attention to the equivocal nature of use. Use can refer either to an action that operates upon a thing or to an action that, through its operation upon a thing, attests to a juridical relationship between the actor and others with respect to that thing. Both kinds of use may be permitted, in the sense that, taken on its own, the action of making use of something may be the exercise of a Hohfeldian liberty. However, because property is a legal concept that relates persons through things, only the second kind of use is relevant to an account of property. What matters is the use of something as property, and not merely use on its own. Accordingly, so far as Kant is concerned, Grotius’ suggestion that use is the precursor of property is a conceptual dead end.

4.  Usability In relating persons through things, ownership raises two issues: on what basis is a person legally so connected to a thing that the thing can count as the object of a right, and how does a person’s connection to a thing come to have a legal effect on other persons? The first of these issues goes to the owner’s right in the thing, the second to the obligation imposed on others through the creation of that right. Both the right and the obligation give expression to the reciprocal freedom of the participants in the relationship. How they do so is the theme of the next two sections of this chapter, which deal with each of these issues in turn. I first address the legal connection between person and thing. The distinction between persons and things is that things can be used but persons cannot. Under the concept of Right, the co-​existence of one person’s action with another’s freedom characterizes the objective order of value for legal relations. Corresponding to this objective order of value is the subjective right that consists in each person’s innate right, which precludes constraint by the will of another. As the expression of the right of humanity in one’s own person, the innate right is incompatible with one person’s being a mere means for others, for to be a means is to be subject to the constraints of another’s will. Persons are thus not usable by other persons. Things, in contrast, have no innate right. As the possible objects of persons’ wills, they are eligible to serve as the means through which persons pursue their purposes. The underlying idea of the Kantian account is that, in a normative order of reciprocal freedom, the freedom to pursue one’s purposes goes beyond the exercise of one’s innate right, and thus beyond the use of things only insofar as one is physically connected to them. Because things are usable, they can be the objects of

Ownership  57 rights that are distinct from a person’s innate right. The argument is the obverse of the Kantian critique of Grotius presented in the preceding section. That critique pointed out that the use permitted in a world of innate right implies no right in the things that are being used. Continuing that thought, the present section contends that a right in the things that are being used is the concomitant of extending reciprocal freedom beyond innate right. This contrast between Grotius and Kant goes hand in hand with another. While Grotius focuses on what he takes to be the natural laws governing particular uses, Kant understands the right to put things to particular uses as a juridical implication of the idea of usability. That idea abstracts from the particularity of specific uses to a more general idea: the normative possibility of using that is presupposed in those specific uses. Ownership is the legal concept that pertains to the usability of things. It is important to note that the normatively relevant dimension of use is usability and not usefulness.30 Usefulness refers to the function of a thing, given its specific qualities, in satisfying the particular needs and wants of the person using it. Usability, in contrast, reflects the legitimacy—​entailed by the distinction between persons and things—​of a thing’s being subjected to someone’s use. Whereas usefulness is geared to the particularity of the thing and of the needs or wants that the thing might satisfy, usability is a formal idea that abstracts from particular uses to the rightfulness of use as such. Usability denotes the eligibility of things to be the means through which persons pursue their purposes as free beings. Just as needs and wants, like all welfare considerations, are not the basis of rights within the Kantian approach, so usefulness is not the basis of ownership. Thus, ownership is not at root a utilitarian device for maximizing the aggregate satisfaction of needs and wants across society, nor is it normatively grounded in being a repository of incentives that operate on needs and wants in order to promote economic efficiency. Instead, it is an expression of the role of the usability of things in a regime of reciprocal freedom. Usability lies at the heart of the Kantian account of ownership.31 A regime of reciprocal freedom limits one person’s action by the freedom of another. A thing, however, is devoid of freedom, and therefore imposes no limit on the action of the person using it. If this were not the case, then a person’s action would be restricted by things rather than by the freedom of others. This would produce the self-​contradictory situation in which something usable could not be put to use, even if using a thing were consistent with everyone’s freedom and thus wronged no-​one. Through the concept of ownership, the law recognizes usability as such. It does this by treating whatever is usable as the possible object of a right that is categorically different from the innate right in one’s person. This non-​innate right does not depend on the physical attachment of the thing to the right-​holder. Nor does

30 31

cf Ripstein (n 7) 181: ‘The law of property is not the solution to a problem about usefulness’. Kant (n 26) 46 [6:246].

58  Ernest J. Weinrib it depend on the particular use that the right-​holder makes of the object of the right; determining that use is wholly within the freedom of the thing’s owner. Thus, the owner has exclusive control over the thing without needing to be present to physically hold and manipulate it. The owner then has ‘an open-​ended set of use-​ privileges’32 in relation to the thing owned; the owner, for example, can sell it, rent it, or authorize others to work on it or even to destroy it. Indeed, only through the possibility of such non-​physical control can the status of a thing as something usable be treated as legally significant. For if usability were a matter solely of physical connection between person and thing, it would be an otiose notion; a person’s use would be protected by the innate right itself without reference to the thing’s usability. Ownership, accordingly, is the condition of a thing’s being subject to the owner’s control without its being physically connected to the owner. It allows me as owner to ‘limit others through my mere power of choice without doing so in accordance with my innate right of the inviolability of my person’.33 Within a regime of reciprocal freedom, freedom alone governs the connection between the owner and the thing owned. Because the thing has no freedom but merely the eligibility to be used, it cannot restrict what the owner can do with it. The thing falls within the owner’s complete control to do or not do with it what he or she wishes. Now one might object that reference to the inability of things to restrict the owner’s actions is misleading. Perhaps persons, not things, could rightfully limit the scope of usability. Thus it has been claimed, against Kant, that under conditions of extreme scarcity property is not in fact compatible with everyone’s freedom because one person’s use of some vital thing precludes someone else’s equally vital use of that thing. On this view, ‘legitimate action requires sufficient abundance that one person’s use (benefit) is not (at least not directly) someone else’s vital injury (deprivation)’.34 The error in this objection is that it introduces extreme scarcity and sufficient abundance in order to make need relevant to reciprocal freedom, and then assumes that a justification of property that is indifferent to need is incompatible with that freedom in circumstances where need becomes urgent. However, freedom in the Kantian account is always relational, as are the considerations that underlie the rights that are the legal manifestations of that freedom. Need, therefore, being applicable solely to the person in need without in itself relating that person to anyone else, is alien to the Kantian analysis of relations between one person and another.35 In effect, the objection understands Kant’s argument as being concerned with usefulness, that is, with the role of things in satisfying needs and wants; it then contends that equal freedom cannot be achieved 32 J Harris, Property and Justice (Clarendon Press 1996) 72. 33 I Kant, Lectures and Drafts on Political Philosophy (F Rauscher ed, CUP 2016) 289 [23:287]. 34 KR Westphal, ‘Do Kant’s Principles Justify Property or Usufruct?’ (1997) 5 Jb R & Ethik/​Ann Rev L and Eth 141, 153. 35 Kant (n 26) 27 [6:230].

Ownership  59 under conditions of scarcity, when a thing’s usefulness to one person comes at the expense of its usefulness to someone else. The objection is misconceived. Kant is dealing not with usefulness but with usability; that is, with the eligibility of things to serve as the means of purposive action. Usability is indifferent to need and unaffected by conditions of scarcity.36 In being concerned not with particular uses but with the eligibility for being used, the notion of usability corresponds to the Kantian conception of the person who does the using. Just as persons are characterized by the capacity for purposiveness regardless of their particular purposes, so things are characterized by the capacity for being used regardless of particular uses. The owner’s potential for action on a thing is mirrored in the thing’s potential for being acted upon. Because the legal category of ownership is based on the usability of things and not on any of their particular uses, a person can have ownership of a thing even during periods of time when he or she is neither actually using it nor intending to use it. The control that the owner has over the thing is in play when law treats the thing as being exclusively available to the owner regardless of physical possession. Another’s interference with the thing or unauthorized use of it counts as a wrong, regardless of whether the owner has been impeded in any actual or prospective use. The notion of usability, then, has a manifold significance for the rightfulness of ownership as a legal category. First, usability is a normative idea that refers not merely to the physical power to use something but to the eligibility of such use from the standpoint of right. Secondly, the rightfulness of using what is usable reflects the normative incapacity of what is usable to impose a limit on a person’s freedom. Conversely, because freedom fills the entirety of the space within which its operation is not rightfully limited, the freedom with respect to which the action of others must co-​exist is not confined to one’s innate right, but rather extends to the kind of right—​‘ownership’—​that reflects one’s rightfully acquired control over what is usable. Thirdly, usability is the mark of what can be owned: things and not persons. Slavery, therefore, lacks rightfulness in that, inconsistently with their innate right, it treats persons as usable. Fourthly, usability is the basis and not merely an incidental feature of the right of ownership. Ownership is the distinctive juridical idea that makes things the objects of rights because they are usable. In this respect, ownership differs from the innate right, which shields use from interference by others only if the thing being used happens to be attached to the body of the person using it. Fifthly, usability is the aspect under which things are owned. To own something is to have a right to it as something usable, that is, to have the right to determine the particular uses to which it is to be put, and even to determine that it is not going to be put to any particular use. Or to formulate this last point another

36 This conception of property in private law does not mean that, as a matter of public law, the state has no duty to relieve poverty; see ch 5.

60  Ernest J. Weinrib way, not putting something to a particular use is itself a particular use, in that it is an exercise of the ownership right that flows from the usability of things. Such a conception of ownership integrates the right to exclude and the right to use. The right to exclude refers to the owner’s entitlement to control the owned thing by having it uninterruptedly available for whatever use he or she might want, even if no particular use is in prospect. The right to use refers to the owner’s entitlement to convert the usability of the owned thing into the particular use that he or she desires to make of it. Use and exclusion thus form the positive and negative poles of the idea of usability, postulating respectively the subjection of the thing to the owner’s will and its subjection to no-​one else’s will. The implication of this polarity is that, on the Kantian view, neither use nor exclusion is conceptually prior to or independent of the other.37 Nor is it the case that the right to exclude is a means for promoting the owner’s interest in use, or that the right to use is a consequence of the right to exclude.38 Accordingly, use and exclusion are not two separate sticks in the supposed bundle of rights conferred by ownership. They are, rather, different aspects of the usability of things that is inherent in the conception of ownership within a regime of reciprocal freedom. The right to exclude is one of the forms that usability takes; that is, it is one of the ways in which usability, as a rightful possibility, can legitimately be realized in legal relationships. Similarly, the right to use and the other incidents of property (such as the right to possess, the right to manage, the right to the capital, the right to the income, and the right of transmissibility, and so on)39 are also forms of usability. These various forms of usability come together in what has succinctly been termed the owner’s authority to set the agenda for the owned thing,40 because the idea of usability implies no restriction on how owners incorporate things into their purposive activity. The Kantian account presents as an integrated expression of reciprocal freedom the conception of ownership that is routinely found in the canonical texts of Western law across the leading jurisdictions.41 Its uniting of use and exclusivity figures, for instance, in William Blackstone’s famous definition of the right of property as ‘that sole and despotic dominion which one man claims and exercises over the external things of the world, in total exclusion of the right of any other individual in the universe’,42 as well as in the classic French definition of property as ‘the 37 Contrast the view of Douglas and McFarlane (n 3). 38 As is claimed by Smith (n 1) and Merrill (n 2) respectively. 39 AM Honoré, ‘Ownership’ in AG Guest (ed), Oxford Essays in Jurisprudence (Clarendon Press 1961) 107–​28. 40 L Katz, ‘Exclusion and Exclusivity in Property Law (2008) 58 UTLJ 275, 278. 41 On the situation in Jewish law see the important article by B Porat, ‘Ownership and Exclusivity: Two Traditions, Two Visions’ (2016) 64 Am J Comp L 147. Porat’s analysis shows that, in effect, Jewish law has a right to exclude (which is effective when exercised) but not a right of exclusivity (which renders the unauthorized use of another’s property wrongful even if the owner does not actively exercise his right to exclude). 42 W Blackstone, The Oxford Edition of Blackstone’s Commentaries on the Laws of England: Commentaries on the Laws of England: Book II: Of the Rights of Things (S Stern ed, OUP 2016) 1.

Ownership  61 right by virtue of which a thing is absolutely and exclusively submitted to a person’s will and conduct’.43 Especially pertinent—​because it provides an explicit formulation of the integrative approach—​is the description of ownership by the great German jurist Bernard Windscheid.44 Windscheid defines ownership of a thing as ‘the right that in itself makes the will of the right-​holder decisive for the thing in the entirety of its relationships’.45 Ownership so conceived is whole and unrestricted, with a positive side under which the owner may provide for the thing in accordance with his will, and a negative side that bars anyone else from providing for the thing without the owner’s will.46 To be sure, Windscheid notes, one can specify the various entitlements47 available by virtue of ownership: the entitlement to use, the entitlement to exclude, the entitlement to claim the thing from an unlawful possessor, the entitlement to alienate. In words that anticipate and reject the bundle of rights approach he continues: However, one should not say that ownership is composed of an aggregate of individual entitlements, or that it is a tying together of individual entitlements. Ownership is the plenum of the right in the thing, and the various individual entitlements are only the externalizations and manifestations of this plenum.48

5.  Acquisition One might object that this conception of ownership seems in one important respect to be in tension with the reciprocal freedom that it purports to express. The tension is between ownership as an idea and acquisition as a process. In this section, I want to explain this tension and show how the Kantian account resolves it. The tension is this. On the one hand, if the argument in the preceding section is sound, ownership is consistent with the reciprocal freedom of all. On the other hand, ownership is always acquired by performing some act, whether it is unilateral, as in the case of original acquisition, or bilateral, as in the case of derivative acquisition, that binds those who are not parties to it. This is especially striking in original acquisition, which is accomplished by a person’s taking control of the unowned thing; original acquisition apparently has nothing relational about it, and 43 C Aubry and C Rau, Cours de Droit Civil Français (5th edn, Paris 1897) 256, s 190. 44 B Windscheid, Lehrbuch des Pandektenrechts, Neunte Auflage (Frankfurt am Main 1906) 856–​ 60, s 167. 45 ibid 858 n 5. 46 ibid 857 n 1a. 47 The word that Windscheid uses is ‘Befügnisse’, literally ‘authorizations’, the same word that Kant uses for the entitlements that fall under the innate right; see ch 2, nn 12 and 23. 48 Windscheid (n 44) 857. Similarly, Charles Demolombe, after referring to ownership as ‘a sovereign power, a complete despotism’, notes that the owner’s entitlements to usus, fructus, and abusus, when united in the owner’s hands ‘make up for him only one whole, and only a single right’ (ne forment pour lui qu’un seul tout, et qu’un droit unique). C Demolombe, Cours de Code Napoleon IX (Paris 1852) 485.

62  Ernest J. Weinrib yet has relational effect. The question that arises from acquisition is: how can an action or transaction in which they did not participate create in others an obligation to abstain from what the owner thereby acquired? Can it be that ownership is consistent with reciprocal freedom but that acquisition—​the mechanism for becoming an owner—​is not? This would mean that ownership is a legitimate condition that can never legitimately be brought about. Put in Kantian terms, the question involves the compatibility of acquisition with Right’s objective order of value and with its concomitant subjective right, the innate right. The innate right is the right to independence from constraint by the will of another. One of the authorizations of innate right is innate equality, the immunity from being bound by others more than one can in turn bind them.49 Acquisition seems inconsistent with this authorization, because with respect to any particular acquired object, acquisition binds everyone else to the acquirer without the acquirer being in turn bound to them. Acquisition, accordingly, seems to be inherently unrightful, an exercise in subjecting others to the constraints of the acquirer’s will in violation of their innate right. The Kantian response to this problem is that ownership is legitimate when one person’s acquisition is related to acquisitions by others through a general system of acquisition. Acquisition is problematic for innate equality only if each instance of acquisition is regarded as standing on its own rather than as partaking in a system of acquisition that applies to all acts of acquisition by all acquirers over time. If seen in isolation, any instance of acquisition appears to bind others more than they in turn bind the acquirer. However, this is not the case if acquisition is part of a system. A system functions as an integrated normative totality that enables any acquirer to regard acquisition by others as entitling his own. Each act of acquisition is legitimated by every other actual or potential act of acquisition. Within a system of acquisition, an acquirer does not bind others more than they bind him. Rather the opposite: their potential or actual binding of him is the condition for his being able to bind them. Acquisition is thereby relational, with no person’s acquisition being normatively prior to or independent of anyone else’s. Because each acquisition is tied to the availability of acquisition to others and no particular acquisition counts as normatively independent, the system of acquisition signifies reciprocity among all potential acquirers, and reciprocity signifies conformity to the innate right of everyone. The subjection of each person within this system is not to the will of another, but to the system itself as the juridical manifestation of reciprocal freedom. To fulfil its legitimating function, the system of acquisition has to be viewed as a totality, and not merely as a plurality, of acquisitions. Regarded as a plurality, acquisitions are an aggregate of individual instances that have no greater legitimacy in sum than each has on its own; regarded as a totality, acquisitions are interconnected



49

Kant (n 26) 34 [6:237].

Ownership  63 with each other to constitute a single whole that endows each instance with a legitimacy that it would individually lack. What matters to the argument about the normativity of acquisition is not merely that everyone can be an acquirer, but that all the acquisitions taken systemically are the ground for the legitimacy of any particular acquisition. One might be tempted to think that acquisition is sufficiently grounded simply through its relation to the acquirer’s freedom: because every individual exercises his power of acquisition only because of his or her status as a person, all persons have this power and therefore none of them can deny it to others.50 However, this contention yields only a plurality of individual acts in which any acquirer can meet a supposed obligor’s challenge that he is being bound more than he binds by saying that ‘I can acquire and you (and everyone else) can acquire’. This response would multiply and not resolve the inconsistency with innate equality. What applies to each acquisition taken on its own would apply to all of them taken severally, resulting not in reciprocal freedom but in reciprocal servitude. In contrast, the Kantian account treats the sum of acquisitions as a systemic totality, not as a plurality. Each acquirer can then say ‘I can acquire because you (and everyone else) can acquire’. The system of acquisition, not the aggregate of all the possible individual acts of acquisition, renders any particular acquisition binding on everyone else. The site for the system of acquisition is the state. For Kant, the state is the entity that through its public institutions relates the totality of persons under its authority to its members as individuals.51 Understood in the light of the reciprocal freedom that ought to animate it, the state is inherently systemic. It integrates the persons who form it into a people, the institutions through which it acts into a constitutional structure, and the norms to which it gives effect into a functioning legal system. Within the state so understood, nothing—​no individual, no institution, no legal category, no norm, no legal practice—​exists in isolation. Every constituent of legality is related to every other through the totality that they together comprise. Acquisition is one aspect of the state’s systemic character, in that each acquisition is legitimated through its participation in the system of acquisition. Thus, the apparent tension between ownership as an idea and acquisition as a process is dissolved by the movement of thought from a condition in which public institutions are absent to one in which they are present. Kant terms these two conditions the ‘state of nature’ and the ‘civil condition’, respectively. In the former condition, ownership is understood as a normative possibility that accords with the reciprocal freedom of all. The link between usability and ownership belongs to the state of nature, because its articulation requires no reference to legal institutions.

50 P Benson, ‘Misfeasance as an Organizing Normative Idea in Private Law’ (2010) 60 UTLJ 731, 788; see also NW Sage, ‘Original Acquisition and Unilateralism: Kant, Hegel and Corrective Justice’ (2012) 25 CJLJ 119. 51 Kant (n 26) 97 [6:311].

64  Ernest J. Weinrib However, ownership cannot be operative in legal relations unless things can be acquired; and things cannot be acquired consistently with reciprocal freedom unless acts of acquisition belong to a system of acquisition. The state of nature has no resources for regarding acts of acquisition in this way. Only in the civil condition, when acts of acquisition are governed by legal institutions, can they be understood as belonging to a system of acquisition. Accordingly, for reciprocal freedom to be realized, ownership as a conceivable concept of reciprocal freedom is postulated in the state of nature, and rights of ownership as effective legal realities are acquired and enjoyed in the civil condition. The hallmark of the Kantian account of property is its seriatim treatment of ownership and acquisition. It moves from the question of how to have something external as one’s own to the question of how to acquire something external.52 Unless ownership is a viable concept under the principle of right, acquisition as the process through which ownership is achieved is normatively unthinkable. Similarly, unless acquisition is consistent with the freedom of all, ownership cannot be actualized. Both are necessary to an account of ownership. The Kantian account of ownership, accordingly, has two stages. Arguing from the usability of things, it first establishes ownership as a legal category compatible with the reciprocal freedom of all. Because things, being usable, cannot restrict freedom, and because freedom fills the entire space of its unrestricted operation, ownership is a necessary category within a system of reciprocal freedom. The Kantian account then enquires into the conditions for the legitimacy of acquisition, again as a matter of reciprocal freedom, in view of the difficulty posed by the obligation that acquisition imposes on strangers to the acquiring act. It deals with this difficulty by situating acts of acquisition within the systematic reciprocity available to acquirers in the civil condition.

6.  Conclusion This chapter has been concerned with the relationship between use and exclusivity within the notion of ownership. Are these two incidents of ownership to be regarded as conceptually separate from each other and, if so, which one is paramount? Or are they to be regarded as the integrated aspects of a single juridical phenomenon? The chapter has presented versions of each of these possibilities, culminating in Kant’s integrative account of use and exclusivity as intrinsic to the legal relationship of persons through things within a regime of reciprocal freedom. That the chapter has brought us to the Blackstonian conception of ownership as the despotic dominion over things may be disappointing to some. Has this



52

Kant (n 26) 41 (6:245), 52 (6:258) (emphasis in original).

Ownership  65 conception of ownership not been discredited by centuries of theorizing about property, by the rise of industrial capitalism and the welfare state, and by the contemporary consensus of property scholars? To those troubled by such thoughts, I offer some sketchy remarks of consolation. This chapter has dealt with the normative basis of ownership as such: within a system of rights, what is the justification for legally relating persons through things? The answer to this question does not yield a complete theory of property law, but is merely the initial stage of such a theory. In the Kantian approach, the significance of property for a system of rights is elucidated within a conceptual progression that is tied together by the idea of reciprocal freedom. In this progression, each stage presupposes and then complements the previous one. The root question is how, within a regime of reciprocal freedom, can persons, considered as persons and thus in abstraction from the particular projects that they each might have, relate to one another through things at all? To answer this question one starts with ownership, that is, with the legal category that reflects the sheer usability of things in abstraction from the particular uses that various individuals might make of them. This is the most abstract and unqualified form of property. Accordingly, instead of proceeding (as Grotius thought) by accretion from use as the most anaemic form of property, one begins with the most robust form, which other forms of property diminish or qualify. The answer to the question about the nature of ownership as such leads, therefore, to a second question: how is ownership, so understood, related to the complexities of a system of property that allows for various diminutions of ownership, whether diminutions from the emoluments of ownership, as when the owner’s entitlement to use or alienate is restricted (restrictive covenants or the rule against perpetuities are examples) or diminutions in ownership itself (for instance, leasehold or the variety of estates in land).53 The answer to this question leads to a third question: given that ownership is normatively legitimate, what is the relationship between one owner and another, as evidenced, for instance, by the law of private nuisance or of private necessity? This is a question not about the right of ownership as such—​that question has already been answered at the outset—​but about the operation of that right within a system of rights. This, then, leads to a fourth question: given that a system of rights requires a state, what is the relationship between owners and the state? On what basis, in other words, can the state qualify, burden, abridge, or terminate an owner’s right? One thinks in this connection of taxation or the police power or land use controls or the role of reciprocal freedom in underpinning a social minimum and arrangements of social insurance. This, in turn, leads to a fifth question: given that a state is regulated by a set of constitutional norms, what counts as property for purposes of constitutional protection?



53

This terminology and classification are taken from Pufendorf (n 12) 534.

66  Ernest J. Weinrib As the German Constitutional Court has pointed out,54 the answer to this question is much broader than the answer about ownership that the Civil Code provides to the first question. This fifth question then leads to a sixth: even for property that falls under the constitution’s protection, when is the right to such property justifiably limited by, or in need of reconciliation with, other constitutional rights and values?55 Under the Kantian approach, theoretical clarity requires that these questions be handled sequentially. This means that, although the issue of ownership discussed in this chapter is not the whole story about the juridical relationship of persons through things, it is indispensable to that story. The sequencing of the questions also means that the answers to the subsequent questions do not undermine but rather presuppose the answer to the first question. As this book is not about property but about reciprocal freedom more broadly, I will not pursue these questions here except as they bear on the exposition of my theme. In that connection, the material in this chapter looks both backwards to the preceding chapter and forwards to what follows. The preceding chapter contended that the Kantian conception of rights provides the content appropriate to the correlative structure of the relationship between plaintiff and defendant in private law. Within any private law relationship, such rights are the organizing and unifying principles of the specific kinds of jural entitlements that Hohfeld elucidated. Accordingly, when conceived along Kantian lines ownership integrates the owner’s liberty to use the owned thing and the owner’s claim-​right to exclude others from its use. This chapter has presented the argument that leads to this conclusion. That argument highlights the role of the civil condition in reconciling acquisition with the reciprocal freedom of all. This role opens on to other issues: More generally, what is the significance for private law of the civil condition, with its systematizing character and its array of public institutions? And how does the notion of conceptual sequencing just mentioned contribute more broadly to the elucidation of reciprocal freedom? In the following chapter I turn to the first of these issues, and in the subsequent chapter to the second of them.

54 BVerfGE 58, 300 (Groundwater, 1981). 55 See eg the treatment of the right to property and the right to housing under the South African constitution in Port Elizabeth Municipality v Various Occupiers 2005 (1) SA 217 (CC) paras 14–​23.

4

Public Right 1.  From Private Right to Public Right In this chapter, I deal with the significance for private law rights of their actualization through the public institutions of a functioning legal system. Kant’s legal philosophy distinguishes between the categories of rights that a person might have against others in accordance with their reciprocal freedom and the public institutions that guarantee those rights. As juridical markers of reciprocal freedom, rights comprise the legal concepts under which the freedom of the right-​holder co-​exists with the freedom of those placed under a corresponding obligation. Among these rights are the standard rights of private law to bodily integrity, to property, and to contractual performance. On the Kantian approach, the content of the rights, the mode of acquiring them, the consequent obligations that the various rights impose on others, and the internal logic by which property rights are good against the whole world (in rem) and contract rights are good against the parties to the contract (in personam) all emerge from an analysis of how the action of one person can be consistent with the freedom of another. The normative character of these rights is discernible even apart from the operation of the public institutions that secure them. However, if rights are to be enjoyed and not merely envisioned, they must be made effective in a system of positive law that determines their specific meanings and that safeguards them from infringement. In performing this role, the legal system neither grounds the normative validity of the rights nor provides merely a remedial rubber stamp for rights that are conceivable independently of it. Rather, while presupposing the rights that it secures, it brings to bear its own distinctive normative commitment to the public and systematic nature of legal relations. The interweaving of this normative commitment with the correlatively structured rights of private law is the theme of this chapter. The chapter thus focuses on the public character of private law.1 This involves no paradox. Mention of the public character of private law is not an invitation to see private law as a repository of instrumental or utilitarian considerations that aggregate or trade off the interests of those subject to it. Instead, this public character is animated solely and entirely by considerations relevant to the achievement 1 For a previous treatment of this theme see EJ Weinrib, The Idea of Private Law (Harvard UP 1995, OUP 2012) 218–​22.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0004

68  Ernest J. Weinrib of reciprocal freedom through a system of private law rights that operates (as it must) within the framework of public institutions. Understood in this way, the public character of private law is integral to the internal normative dynamism of private law. Kant’s account of this public character presents a movement to a system of positive law from an imagined condition in which persons interact without the availability of public institutions. Kant calls this imagined condition the ‘state of nature’, in contrast to the ‘civil condition’ marked by the presence of public institutions. Although conceivable in the state of nature, rights cannot be fully realized there, because the state of nature lacks the systematicity and the publicly authoritative institutions of a functioning legal order. Systematicity is needed if each person is to enjoy the most extensive freedom that can co-​exist with the similar freedom of all other persons across the entire ensemble of legal rights and obligations; institutional authoritativeness is needed if a person’s rights are to be determinate and publicly knowable domains of freedom secure from usurpation by others. Together, systematicity and publicly authoritative institutions allow a regime of reciprocal freedom not merely to be conceivable as a rightful possibility but also to exist as an ordered and effective juridical reality. The Kantian state of nature is a device for imagining the interactions that can conceivably be informed by the parties’ reciprocal freedom, even if one does not consider the role of authoritative public institutions. These interactions involve five ideas. First, each person is endowed with the innate right to be independent of the constraints of another’s will, because the innate right is the subjective right presupposed in any conception of interaction based on reciprocal freedom. Thus, in accordance with one’s innate right, each person has a right that others cannot interfere with one’s bodily integrity. Secondly, freedom is conceived as extending beyond the exercise of one’s innate right; it includes also the use of things that are distinct from another’s body, because such use does not interfere with the innate right of anyone. From the legitimacy of such use—​what in the preceding chapter I termed ‘usability’—​arises the possibility that things, as well as one’s body, can be the objects of rights. Such rights to things are not innate but acquired. Thirdly, the different kinds of acquired rights are conceived as reflecting the different ways in which the reciprocal freedom of the interacting parties can manifest itself. One’s freedom can be exercised in the pursuit of one’s purposes either independently, or consensually with another, or on behalf of another. Corresponding to these are the right of ownership, the right to contractual performance by another, and the set of entitlements and duties that pertain to fiduciary relationships (such as the relationship of parent and child). Fourthly, each such right is conceived as being acquirable through the mode of acquisition that befits the kind of right it is. A right to contractual performance, for instance, requires the promisor and promisee to manifest a united will regarding the promise.2 Similarly, because ownership subjects

2

I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 63 [6:272].

Public Right  69 something to the exclusive control of the owner, original acquisition is achieved by acting as an owner,3 that is, by bringing an unowned thing under one’s control through an act that indicates one’s intention to exclude everyone else from it.4 Fifthly, to present these conceptual possibilities as underpinning juridical interactions based on rights that each have their distinctive scope and logic, persons are imagined to perform acts that would count as acts of acquisition if they were not in the state of nature. Because true acquisition is not yet available, such acts give persons merely provisional rights that would become conclusive on the transformation of the state of nature into a civil condition. Thus, a person who has been the first to take control of an object with the intention of excluding others has, to the extent possible within the state of nature, taken the requisite steps to gain provisional ownership of it. Taken together, these five ideas present the state of nature as modelling how the rights of private law (and the obligations correlative to them) constitute a distinctive morality of their own, even apart from the institutional considerations that figure in a functioning system of law. The point of setting up the state of nature is not only to provide an account of the normative character of the rights of private law, but also to underline the necessity to move beyond the state of nature to the civil condition. The phenomenon of ownership provides an example of the limitations present to the state of nature. Among these limitations are the following. First, ownership cannot bind others consistently with their reciprocal freedom unless it is situated within a system of acquisition; in the state of nature, however, purported acquisitions are discreet individual events rather than constituents of a system. Secondly, what constitutes the control required for original acquisition must be authoritatively specified (if I enclose land within a fence do I control the land enclosed or the land on which the fence stands?); in the state of nature, however, no authoritative specifying body is present. Thirdly, disputes about ownership have to be resolved in accordance with the reciprocal freedom of each party; in the state of nature, however, no institution exists to adjudicate disputes or enforce judgments. Fourthly, ownership carries with it the liberty to use what is owned; in the state of nature, however, no mechanism exists for dealing with uses by different owners that are incompatible with one another. Evident as these deficiencies are, the state of nature lacks the resources to remedy them. The problem with the state of nature is that its rights function only through the parties’ unilateral wills. In the absence of public institutions, the specification of the entitlements that make up any right and the taking of measures to secure the right or to exact compensation for its infringement are referable only to the private judgment and the private power of one or the other of the parties. Although the 3 This is Peter Benson’s helpful formulation; see P Benson, Justice in Transactions: A Theory of Contract Law (Belknap Press 2019) 344. 4 Kant (n 2) 52 [6:258].

70  Ernest J. Weinrib parties are related to each other through their rights and obligations, no mechanism is available authoritatively to expound and compel adherence to the normative implications of the relationship. Thus, although the rights are conceived as expressions of reciprocal freedom, their operation is not yet consistent with the parties’ equal standing. The civil condition addresses this problem by systemically relating all the parties and their rights to one another and by providing the publicly authoritative institutions that allow the rights to function in a manner that is true to the reciprocal freedom that grounds them. Kant calls the ensemble of rights that pertain to the state of nature ‘private right’. In contrast, ‘public right’ refers to the totality of laws that pertain to the civil condition. In the civil condition individuals are related to one another through their participation in an entity—​the state—​that relates everyone to everyone else through institutions that publicly promulgate, interpret, and enforce their rights. Public right thereby systemically connects and institutionally secures the rights of the individuals in the civil condition.

2.  Omnilaterality The civil condition is structurally different from the state of nature. In the state of nature, the parties to a legal relationship are conceived as being related to each other bilaterally as bearer of a right and obligor of a correlative obligation. In contrast, the civil condition encases the relationship of right-​holder and obligor in a framework that links everyone to everyone else through the legal system that all share. In the civil condition, the bilateral relationships of private right are incorporated within a public right that is omnilateral, in the sense that it connects each person to all and all to each.5 So far as liability is concerned, the most salient feature of the civil condition is the role of publicly authorized courts. A court is the instrumentality through which the duly appointed delegates of the community as a whole resolve controversies between its members. In accordance with its omnilateral function, it draws out the juridical implications of the entire system of rights for any particular controversy, thereby subjecting any given set of litigants to the requirements of that system. Because it speaks in the name of the legal system as a whole, a court is both disinterested and impartial as between the litigants. It is disinterested, in that it has no stake that aligns it with either of the parties. It is impartial, in that it brings to bear a normative perspective under which the justifications for liability operate 5 Kant (n 2) refers to omnilaterality at 53 [6:259–​60] and 56 [6:263]. At 99 [6:314], he says of the legislative aspect of omnilaterality that ‘each decides the same thing for all and all for each’. On omnilaterality in the property context see the instructive article by L Austin, ‘Possession and the Distractions of Philosophy’ in J Penner and HE Smith (eds), Philosophical Foundations of Property Law (OUP 2013) 182, 191–​95.

Public Right  71 within a system of rights to embrace the relationship as a whole rather than either of the parties separately. As a result, rights are not only conceivable as expressions of reciprocal freedom, but can also actually be enjoyed as such.6 In adjudication, both the bilateral relationship between the parties and the omnilateral relationship among members of the civil condition contribute their respective normative dimensions. For the bilateral relationship, the normative dimension consists in the parties’ subjection to the correlatively structured bases of liability. For the omnilateral relationship, the normative dimension consists in every member’s subjection to the state’s lawful authority as it acts in the name of the citizenry as a whole. In adjudication, a court combines these two dimensions by projecting its own omnilateral authority on to the parties’ bilateral relationship. The court thereby extends the significance of its decision beyond the specific dispute, making it a norm for all members of the state. In the Kantian account, the state of nature and the civil condition are interdependent and mutually complementary. Formulated sequentially as moments in an articulated unity, neither makes any normative sense without the other. Without the civil condition, the rights of the state of nature could not be actualized; without the state of nature, there would be no rights for the civil condition to actualize. The movement from the state of nature to the civil condition is conceptual, not historical; it points not to the process by which the state actually emerged, but to the sequence in which the various aspects of a system of rights have to be considered if the interconnection between them is to be intelligible in an ordered way. The sequence reflects the progression of thought from general to specific with respect to both the substance and the enforceability of rights. As a matter of substance, the rights postulated for the state of nature themselves specify the content for the most abstract characterization of the concept of Right, that is, for the totality of conditions under which the action of one person can co-​ exist with the freedom of another.7 In determining that abstraction, the state of nature discloses a set of subjective rights (such as the right to personal integrity, to property, and to contractual performance) that conform to the reciprocal freedom of right-​holder and obligor but are themselves abstractions that require further determination. The civil condition renders those abstractions still more determinate through public institutions, specifying (for instance) how property is acquired or how the parties’ wills come together to form a contract or how these rights function in the omnilateral context of public right.8 As a matter of enforceability, public right also makes determinate the conditions for employing coercion to secure a party’s rights. The authorization to use coercion accompanies even of the most abstract characterization of the concept of Right,

6

ibid 92 [6:306]. See ch 2, s 2. 8 On this latter point see ss 3 ff below. 7

72  Ernest J. Weinrib because conduct incompatible with reciprocal freedom can be resisted for the sake of such freedom.9 The rights in the state of nature specify more particularly the kinds of circumstances in which the freedom of one person imposes a correlative obligation on another; they thereby also specify the kinds of circumstances where coercion is, in principle, an apt response to the breach of that obligation. Through its further determination of rights, the civil condition more precisely specifies the occasions for the legitimate coercive enforcement of those rights. In addition, its public institutions also provide publicly authorized mechanisms for such enforcement under law, so that coercion is no longer the exercise of the unilateral will of the aggrieved party. With these further determinations of both the substance and the enforceability of rights, the civil condition elaborates the rights of the state of nature in the light of the reciprocal freedom that stands at the most general pole of the entire sequence. In moving from general to specific, the sequence from the state of nature to the civil condition also moves from the conceivable to the actual. The rights conceivable in the earlier stage of the sequence are presupposed in the latter stage; in turn, the latter stage provides the conditions under which those rights effectively function as constituents of a system of reciprocal freedom. Thus, the civil condition comes not to abolish the rights conceivable in the state of nature but to fulfil them, that is, to fulfil the idea of reciprocal freedom that animates them and that they inchoately express. Conversely, the civil condition would have no material on which to operate unless those rights were normatively available for it to systematize and secure. The civil condition, accordingly, does not inscribe its norms on a blank slate. Rather, it subjects the private right of the state of nature to the omnilaterality of public right.10 9 Kant (n 2) 28 [6:231]. 10 For a different interpretation of Kant see A Brudner, ‘Private Law and Kantian Right’ (2011) 61 UTLJ 279. Brudner contends that under public right, private right has no continuing or even residual normative force. As an interpretation of Kant, Brudner’s analysis is unsatisfactory on several counts. First, Kant expressly affirms that private right and public right have the identical ‘matter’, that is, that the kinds of rightful possession (property, contract, and domestic right) are the same in both (Kant (n 2) 93 [6:306], 98 [6:313]). Kant sees public right as a unity of form and matter, with the matter being taken over from private right and the form consisting in the publicness of public right (ibid 92 [6: 306]). Secondly, Brudner regards Kant’s statement that public right ‘alone determines what is right, what is rightful and what is laid down as right’ (ibid 59 [6: 267] as showing that nothing of independent force necessarily survives from the state of nature (Brudner 297). This ignores the technical meaning of ‘determines’ in Kant, which refers to the specification of an abstraction; see I Kant, Logic (RS Hartman and W Schwarz trs, Dover Publications 1974) 105 (I, 15). In the passage quoted, ‘rightful’ is precisely Kant’s characterization of the norms of the state of nature (as contrasted with ‘what is laid down as right’, which refers to the norms of positive law in the civil condition). Public right determines what is rightful by specifying the abstractions of the state of nature. For example, in the state of nature original acquisition requires control, but public right (that is, the positive law of a given jurisdiction) tells us what constitutes control for this purpose. The sentence thus states the opposite of the conclusion that Brudner draws from it. Thirdly, in accordance with his thesis that private right has no residual force under public right, Brudner (at 299) asserts that, under Kant’s conception of public right, a person whose property has been expropriated has no right to compensation. Kant, however, is explicit that such a right exists. See Kant (n 2) 145 [6:369].

Public Right  73 In performing its function of authoritative determination and enforcement, public right brings to bear two normative ideas that are distinctive of its omnilateral character. The first of these is publicness. Being omnilateral, public right must be marked by legal norms capable of being known by all, by legal processes available to all, and by methods of fact determination that are accessible to all. Free and equal persons could not be bound by a principle of action that depended on its being concealed from them. Such a principle could not possibly express their freedom; free persons have access to knowledge of the rights that they are supposed to enjoy, aware that these rights reflect their freedom and not the grace of their rulers. Nor could such a principle express their equality; concealment prevents assurance that the state is respecting each and every person as an end, rather than merely manipulating them, or some of them, as means. Accordingly, state institutions are public not merely because their actions are binding on everyone, but also because their reasons for action are capable of being made public to everyone. Because the idea that legal norms be publicly knowable applies to all norms regardless of their content, one may think of publicness as the formal aspect of the omnilaterality of public right. The second idea informing public right is systematicity. This is the substantive aspect of the omnilaterality of public right, because it bears on the relation of the norms to one another, to the institutions from which they arise, and to the legal community whose members are all subject to them. Kant defines public right as ‘a system of laws for a people, that is, a multitude of human beings . . . which, because they affect one another need a rightful condition under a will uniting them . . . so that they may enjoy what is laid down as right’.11 Public right is a unifying idea that has its own integrating conception of a people, of its laws, and of its institutions. A people is a multitude of human beings who are related to one another by virtue of belonging to the same commonwealth. The interacting persons are, accordingly, not an aggregate of individuals but members of a political unit that expresses a united will through a system of laws that are binding on everyone. The legislature, executive, and judiciary perform the functions that respectively actualize this united will: formulating the laws, carrying them out, and giving judgments that award to each person what is due under them. The laws, in turn, are not a collection of disparate rules, but a systematic union of norms. Taken in its entirety, public right is a whole that embraces and systematically connects the interacting persons, the terms on which they interact, and the institutions that determine and enforce those terms. The adjudication of liability manifests both publicness and systematicity. First, a court exercises its authority in a public manner by exhibiting justifications for liability that are accessible to public reason. Juridical concepts, such as property and



11

ibid 97 [6:311].

74  Ernest J. Weinrib contract, form the basis for a process of reasoning that is open to all and that is applied to factual evidence that on reasonable investigation can be openly produced and made patent to all. Opacity or secrecy at any point is a legitimate ground for criticism or requires special justification. Secondly, the court’s decision partakes of the systematicity of the entire legal order. This has both an institutional and a doctrinal aspect. The institutional aspect arises from the differentiation between the legislative, executive, and adjudicative functions of the state and, thus, between the various institutions that serve these different functions. Public right requires a court to conform to this system of coordinate institutions by acting within its competence as an adjudicative body and by not usurping the role of other state institutions. The doctrinal aspect is that the reasoning of any decision forms part of a coherent pattern of reasoning across decisions. Although the court decides as between two particular parties, the significance of its decision is not confined to those parties alone but is binding on everyone. And although the court adjudicates a particular ground of liability, its reasoning has to fit within the entire ensemble of the concepts and principles that govern the interactions of all parties. The public and systematic qualities of public right are closely connected, as pertaining respectively to the form and content of public right. Consequently, a deficiency in one is usually associated with a deficiency in the other. If a decision is reached through a process of reasoning not open to all, there is no assurance that the decision is within the adjudicative competence or forms part of a coherent pattern of reasoning across decisions. Conversely, decisions beyond the court’s institutional competence do not evince the distinctive kind of public reasoning characteristic of the adjudicative process; nor are they based on evidence available to judges or within their institutional capacity to access and to assess. Similarly, a decision that is inconsistent with other decisions leaves opaque the real basis on which disputes of that sort are adjudicated. Thus, public right seamlessly elaborates the correlatively structured rights and obligations of private right. Private right normatively connects reciprocal freedom to the large abstractions of private law—​the rights to personal integrity, property, contractual performance, and so on. Public right then provides the determinations for these large abstractions by specifying the principles, standards and rules of varying generality without which the abstractions of private right could not be operative in interaction under law. Within the institutional context of the court, the rights and obligations of private right, as well as the principles, standards, and rules through which public right articulates their meaning in particular circumstances, constitute a domain of public reason. Moreover, these rights, principles, standards, and rules participate in the legal order’s systematicity. Institutionally, they are within the court’s adjudicative competence because they deal with justice between the parties rather than with distributive issues requiring political decision. Doctrinally, they form a coherent pattern of reasoning, because they operate

Public Right  75 not only within any given relationship but also across relationships, and not only as the components of discreet holdings but also as constituents of a concordant body of private law doctrine. Public right allows the rights conceivable in the state of nature to be enjoyed through the operation of public institutions. In any legal controversy the kind of right at issue sets the internal logic of the claim; public right’s omnilaterality endows that logic with the dimensions of publicness and systematicity.

3.  ‘A Common Fault of Experts on Right’ I now propose to set out a notable consequence of the different kinds of determination that figure in the transition from the state of nature to the civil condition. The rights of the state of nature are abstractions that are then made determinate in the civil condition. Only in the civil condition can this determination occur, because determination is the work of an authoritative positive law that binds everyone through the omnilateral effect of public right. Such determination takes two forms. The first form of determination—​the one assumed in the preceding section—​is the specification of the right taken on its own in accordance with its internal logic. For example, ownership of an unowned object is acquired through the assertion of control; the public right of a given jurisdiction specifies the kinds of acts that constitute legally effective assertions of such control in different circumstances. Similarly, the right to contractual performance emerges from the concurrence of the wills of the contracting parties; public right specifies how this concurrence is established, as is evidenced in the common law by the doctrines concerning offer, acceptance, and consideration. In these examples, the determination follows through on the normative character of the particular right considered on its own. In this and the following section of this chapter, I elaborate on a second form of determination. This form of determination involves specifying how the right functions in relation to other aspects of the legal system as a whole, in view of the role of publicness and systematicity. The scope and internal logic of the right remain what they were as a matter of private right. However, because public right insists on the realization of its own distinctive normative ideas of publicness and systematicity, considerations specific to these ideas govern the law’s treatment of particular rights. Consequently, under certain circumstances a court may rightfully adopt a principle at odds with the internal logic of the right as it was conceived in the state of nature. In Kant’s account, this perhaps surprising difference between the scope and internal logic of a right as a matter of private right, and the adjudicative treatment of the same right as a matter of public right becomes evident when facts crucial to the transaction are not publicly ascertainable. An opposition then arises between the inner logic of the parties’ rights, on the one hand, and the public character of the

76  Ernest J. Weinrib parties’ interaction and of the court’s consequent decision, on the other. To such situations, two contrasting perspectives, each normatively valid, apply, one concerned with what is right in itself as a matter of private right, and the other with what is publicly right. The latter prevails, because rights cannot be enjoyed beyond the capacity of a court to adjudicate specific cases through a public process that deals with the publicly ascertainable aspects of the parties’ interaction. An illustration that Kant discusses is the doctrine, familiar to the common law as well, that a sale in market overt transfers title even if the vendor has no title to give. Market overt operates as follows. It is a commonplace of the law of property that only a person who has title to a thing can transfer that title: nemo dat quod non habet. Assume, however, that the thing is sold by a thief or a borrower who has no right to sell it. Under the doctrine of market overt, a purchaser for value without notice of the vendor’s defective title could retain the thing even against the true owner, provided that the purchase occurred in an open market. Kant’s analysis of this notoriously problematic situation is that the two opposing notions—​that one cannot transfer what one does not have, and that the purchaser in an open market can acquire a title that the vendor does not have—​are both valid, but from different points of view. The first notion accords with what is intrinsically right as a matter of reason when one focuses on ownership as a juridical category in abstraction from the institutions of public right. Because an owner has a right against the whole world, ownership cannot be affected by a putative sale by a non-​ owner to a purchaser, however innocent. The second notion, however, reflects the publicness requirement of public right. If the law insisted that a vendor have good title, a purchaser would have to verify the entire chain of title—​an investigation that ‘would go on to infinity in an ascending series’.12 Because such verification is effectively impossible, a legal system that required it would be unable to fulfil a primary function of public right, to guarantee secure acquisition. Instead, public right contents itself with allowing transfer by a non-​owner to confer title on an innocent purchaser if the transaction observes the conditions of publicness present in an open and publicly regulated market. Through its public quality, an open market both creates a mechanism for securing the purchaser’s acquisition and provides an opportunity for identifying the goods as misappropriated by the seller. By having recourse to the doctrine of market overt, a judge determines ownership on the basis of what is publicly ascertainable, with the result that ‘what is in itself a right against a person, when brought before a court, holds as a right to a thing’.13 In offering this analysis of market overt, Kant is not proposing a rule. Historically, the idea of market overt has existed in many variations.14 A legal system that employs this idea would have to decide on its specific contours, based on (among other

12

ibid 89 [6:301]. ibid 89 [6:302, emphasis in original]. 14 D Murray, ‘Sale in Market Overt’ (1960) 9 ICLQ 24. 13

Public Right  77 factors) whether the publicness of an open market gave a reasonable opportunity for unmasking the infirmity of the seller’s title. One such decision would have to specify what constitutes a market overt. Before the abolition of market overt in England, for example, every shop within the City of London qualified as a market overt with respect to the kind of goods that it normally sold,15 whereas under the German civil code market overt is largely restricted to goods sold at a public auction.16 Another such decision would concern special conditions applying to special kinds of goods. Kant takes the stolen horse as paradigmatic. English law, in contrast, made special provisions for horses, because their mobility allowed them to be spirited away to markets far beyond the scrutiny of true owners or of their neighbours. Or perhaps, as under modern conditions where market overt has been abolished, the geographic diffusion of the market and the ease and impersonality of transactions make the idea of market overt inapposite. Kant’s justification for market overt does not predetermine any of these decisions. Rather, it points out the existence of a distinctive normative space informed by publicness, and it situates that space within the entire domain of normative considerations applicable to the transaction between the parties. How a particular system of positive law fills or ought to fill this space is another matter. What, then, is the relation between what is right in itself and what is publicly right? What is publicly right provides the court with a new principle of decision based on the omnilateral standpoint of a public institution. It does not, however, transform what is right in itself. Public right, Kant remarks cryptically, ‘contains no further or other duties of human beings among themselves than can be conceived [in private right]; the matter of private right is the same for both’.17 The matter of private right refers to the various kinds of rights that one can have in external things. For Kant there are three such kinds of rights: property rights, rights to contractual performance, and rights with respect to household relationships. Each of these kinds of rights has its distinctive normative nature, in accordance with which the internal logic of the right and its correlative obligation operates. Public right does not change this internal logic. Indeed, mistaking considerations of publicness for what is right in itself, Kant alleges, is ‘a common fault of experts on right’,18 that is, of persons conversant with the positive law who lack a true understanding of its normative foundations.19 Since Kant’s time, the prevalence of realism and instrumentalism in legal studies has made this ‘common fault of experts on right’ even more appealing. The legal realist ascribes decisive importance to the point at which the legal dispute makes contact with the coercive apparatus of the state. In Karl Llewellyn’s famous words,

15

P Smith, ‘Valediction to Market Overt’ (1997) 41 Am J Legal Hist 225, 242. Murray (n 14) 48. 17 Kant (n 2) 93 [6:306]. 18 ibid 85 [6:297]. 19 ibid 26 [6:229]. 16

78  Ernest J. Weinrib ‘law is what officials do about disputes’.20 Realists are of the view, therefore, that one cannot think of a right apart from the way the court enforces it. From the fact that considerations of publicness determine the principle of decision in a given case, a realist would conclude that those considerations are constitutive of the right itself. In contrast, Kant’s view is that law as a normative practice and the process of adjudication would themselves make no sense unless the legal categories applicable to the dispute were already immanent in the interaction of the parties as self-​ determining beings and were therefore available for the court when the case came before it. Publicness merely adds what is necessary for the court to function as a public institution, even to the extent of changing the decision, but publicness does not transform the nature of the underlying right. As for the instrumentalists, the contrast with Kant goes both to structure and to content. From the structural standpoint, Kant understands law as a sequence of ideas in which one first identifies the concept of freedom that pertains to law, then works out the various rights expressive of this freedom in the conceptual space of the state of nature, and finally posits the public institutions necessary for the enjoyment of the rights. The stages of the sequence comprise a conceptual ordering that articulates in a progressively more adequate form the conditions under which the freedom of one person can co-​exist with the freedom of others. Instrumentalists, in contrast, have difficulties with sequenced ideas. Once the underlying goal or combination of goals is posited, there seems little reason to deny it general scope. Accordingly, if the law’s publicness is considered instrumentally valuable in the service of some goal, that goal would also be the relevant to the analysis of such fundamental concepts as property and contract. With respect to content, the contrast with Kant can be illustrated by supposing how an instrumentalist might approach the doctrine of market overt. Blackstone observed that, without a doctrine of market overt, ‘all commerce between man and man must soon be at an end’.21 From this one might readily infer that the point of the doctrine is to facilitate commercial activity. Then, because the doctrine applies to contracts for the sale of property, one might conclude that the facilitation of commercial activity is also the goal that justifies the protection that the law affords to property and contract. From the Kantian perspective, this line of reasoning confuses the consequences of the law with its justification. The facilitation of commercial activity, not being a correlatively structured consideration (as corrective justice requires), is a poor justification for property and contract. On the Kantian approach, rights in property and contract are the juridical markers of reciprocal freedom. Nor is the economic account of the doctrine of market overt satisfactory from the Kantian perspective. The normative significance of the doctrine is not the economic goal that it serves

20

21

K Llewellyn, The Bramble Bush: Our Law and Its Study (rev edn, Oceana Publications Inc 1951) 12. W Blackstone, Commentaries on the Laws of England (Clarendon Press 1765–​1770) 2.449.

Public Right  79 but the condition of publicness that it exemplifies in accordance with what public right requires. The idea that ‘the matter of private right’ remains unaffected by an opposing judgment responsive to the need for publicness has two implications. First, judgments from the standpoint of public right do not justify a revision in the basic categories of private right, such as property and contract, despite the inconsistency of these judgments with the internal logic of those categories. Such a revision would undermine that internal logic by introducing considerations that do not pertain to it. Secondly, because the judgment about publicness supervenes upon legal categories that remain intact and continue to structure the relationship between the parties, the judgment should bear the imprint of the category to the extent possible. Far from opening the door to a wide-​ranging instrumentalism at odds with the nature of private law, the judgment of public right should vary the result that would follow from the internal logic of the basic categories only to the extent necessary to achieve publicness. In market overt, for instance, the doctrine should not effect an absolute transfer of title to the purchaser (as was the case under English law) but should allow the true owner to regain title by reimbursing the purchaser for the price paid (as in Jewish law22 and many continental systems23). The true owner would then have the power to reassert the ownership that was never properly terminated for an object that has sentimental value to her or that turns out to have a greater value than was reflected in the price.24

4.  The Effect of Systematicity Kant’s discussion of the common fault of experts on right is devoted to demonstrating that the publicness requirement of public right can transform the principle of decision, while leaving the internal logic of the right unaffected. Turning now to the other aspect of public right, the systematicity of law, I want to suggest (although Kant did not advert to this) that systematicity can have a similar effect. The basic idea is this. The driving impulse of the Kantian approach to law is to vindicate the possibility that law can conform to the requirements of reciprocal 22 J Karo, Shulchan Aruch (1565) 356.1. 23 Murray (n 14). 24 This was the solution reached by the Supreme Court of Israel in the fascinating case of Cnaan v The United States Government, 57(2) PD 632 (2003). The defendant had purchased a painting at a flea market. The painting turned out to have been the work of a distinguished Israeli artist that had been stolen while on tour. Even though the Israeli sales legislation gave unencumbered title to the purchaser in an open market, the majority of the Court held that the true owner could trace its property into the vendor’s power to rescind due to mutual mistake. Accordingly, the Court concluded that no sale had taken place and that therefore the market overt rule was inapplicable. It nonetheless required the true owner to reimburse the purchase price to the purchaser and to compensate the purchaser for the expenses incurred in investigating the painting’s provenance.

80  Ernest J. Weinrib freedom. To this end, Kant moves through a series of conceptual stages, starting with the concept of Right as the ensemble of conditions under which the freedom of one person can co-​exist with the freedom of everyone else, then proceeding to the subjective rights of private right, and finally integrating these rights into a public and systematic totality of persons, norms, and institutions. Accordingly, Kant neither begins nor ends with a collection of specific rights. Although he regards rights as juridical manifestations of freedom and therefore as necessary for a free society, his broader intention is to explore the totality into which rights fit. This means that the specific rights that arise in the state of nature do not exhaust the normative space comprising this totality. Although a person holds the kinds of rights postulated for the state of nature, those rights now operate within a public and systematic framework that has distinctive requirements of its own. Indeed, given the primacy of the parties’ unilateral wills in the state of nature, those rights can be effective as the manifestations of reciprocal freedom only if incorporated into the civil condition, where they operate in accordance with the distinctive normative ideas that inform public right. Therefore, if public right treats these rights in ways that diverge from the internal logic postulated for them by private right, then that internal logic has to cede. We have already seen that for Kant this is explicitly the case with respect to the publicness aspect of public right. My suggestion is that this is also the case with the systematicity aspect. In the following sections I illustrate this by reference to both contract and property. With respect to contract, I point to two doctrines (inducing breach of contract and assignment) that extend the effects of contract from the contracting parties to third parties. With respect to property, I consider two doctrines (the privilege to preserve property and liability for private nuisance) that modulate the entitlements to exclude and to use that are indigenous to ownership. Needless to say, this is not an exhaustive catalogue of the instances of this phenomenon.

a.  Inducing Breach of Contract Because of its apparent inconsistency with the nature of contract, the tort of inducing breach of contract has long been a puzzle to legal commentators. A contract links two parties through a consensually assumed set of mutual rights and obligations. The effect of the tort of inducing breach of contract is to extend to the rest of the world the obligation to respect the contract. The tort thereby ‘inexplicably convert[s]‌the in personam right created by the law of contract into an in rem right for purposes of tort law’.25 This conversion leads some to regard the cause of action as ‘quasi-​proprietary’,26 and even to read the characteristics of property back

25

26

J Neyers, ‘The Economic Torts as Corrective Justice’ (2009) 17 Torts LJ 162, 164. Zhu v Treasurer of New South Wales [2004] HCA 56.

Public Right  81 into the contractual right.27 These moves mask the difficulty by recourse to an unilluminating label or compound it by importing the uncertainty about the tort into the contract itself. The Kantian explanation of this tort draws on the omnilateral significance of public right.28 In the state of nature a contract binds only those who are parties to it, creating a right to performance in the promisee and placing the promisor under a correlative obligation. No-​one, however, can have the assurance that his or her rights will be respected. Moreover, in the absence of such assurance, one could not be relied upon to carry out one’s own contracts, for to treat one’s own contracts as binding without the assurance that everyone does likewise would be to subordinate oneself to the will of others. Public right cures the ineffectiveness of contractual rights in the state of nature by creating a system of omnilateral assurance through institutions of adjudication and enforcement that represent the general will of all. This system of omnilateral assurance of course requires that courts hold the contracting parties to their obligations. But because public right relates each person to every other person through the system of laws that all share, it also requires more. When everyone is united under a system of laws that assures the rights of all, everyone is obligated to respect everyone else’s contractual rights. Because a court operates under the authority of public right, it is not merely a private arbitrator of a private arrangement between the promisor and the promisee. Rather, it has the public function of making everyone secure in their rights against everyone else. This function would be unfulfilled if parties external to the contract could procure violations of another’s contractual rights at their will. Accordingly, whereas in the state of nature the parties to a contract are not secure even against each other, public right makes their rights secure against everyone by attaching liability not only to a breach of contract by the other contracting party, but also to the procuring by third parties of such a breach. Thus, public right makes the contract a juridical object for everyone, thereby creating a system of reciprocal assurance that relates all to all. From this Kantian explanation, one can readily understand why liability for interfering with the contract is based on intent and excludes negligence.29 The point of the tort is to provide assurance to a contracting party that no-​one, not even a stranger to the contract, may act inconsistently with the recognition of the contract’s juridical significance. Essential to the Kantian conception of this wrong is that persons who commit it act on the implicit principle that they are free to disrespect contracts to which they are not parties. Liability responds to the wrong in order to provide the assurance that no-​one, whether a party to the contract or not, 27 R Epstein, ‘Inducement of Breach of Contract as a Problem of Ostensible Ownership’ (1987) 16 JLS 1 28 S Byrd and J Hruschka, ‘Kant on “Why Must I Keep my Promise?” ’ (2006) 81 Chi-​Kent L Rev 47, 68–​71. My formulation of the Kantian argument is based on Kant (n 2) 44 (6:255–​56). 29 But compare Neyers (n 25) 174.

82  Ernest J. Weinrib can regard another’s contractual right as a nullity. Hence, the tort requires knowledge of the contract’s existence and an intention to interfere with its performance, because one cannot regard as a nullity something that one does not know exists and that one’s action does not target. Mere negligence with regard to the benefits that would accrue to another under a contract does not imply a refusal to treat the contract as an object of respect. This explanation of the tort of inducing breach of contract has two similarities to Kant’s treatment of market overt. First, the additional layer of analysis that reflects the omnilaterality of public right leaves intact the normative connection between contract and in personam rights. The explanation thereby obviates the need to read proprietary notions back into the category of contract in order to provide a basis for the apparently in rem character of the tort. Indeed, to do so is to commit the ‘common fault of experts on right’ that Kant noted in his observations on publicness. Secondly, the explanation bases itself on the requirements of a system of rights, not on the commercial advantages that might flow from protecting contracts against third parties. Economic analysts of law have discussed whether and under what conditions the tort of inducing breach of contract contributes to economic efficiency, especially in the light of the apparent tension between this tort and the efficient breaches of contract that they think the law should facilitate.30 From the Kantian perspective, nothing about the normative foundation of the tort hinges on this issue. Rather, the tort is a juridical reflection of the systematicity of law as a juridical phenomenon.

b.  The Assignment of Contract Rights Like the tort of inducing breach of contract, the possibility of assigning contractual rights extends the effects of a contract to persons who are not parties to the contract. An assignment allows the benefit of the contract to move from the obligee to a third party, who is then entitled to demand payment or performance from the obligor and to give a good discharge. The omnilaterality of assignment consists in the obligee’s power to assign to—​and thereby place the obligor into a legal relationship with—​anyone. The benefit due under a contract is thus not only an element in the transaction between the two contracting parties; through assignment, the benefit can also become the subject matter of a further transaction (and through sub-​assignment, of yet a further transaction), thereby entering the stream of commerce and linking the obligor to any potential navigator of that stream. Because of

30 F McChesney, ‘Tortious Interference with Contract vs ‘Efficient’ Breach: Theory and Empirical Evidence’ (1999) 28 J Legal St 131.

Public Right  83 the possibility of assignment, contract rights are bilateral phenomena that contain within themselves the seeds of their own omnilaterality. This omnilateral aspect of contract is particularly notable when the contract right is to a payment. Alienability through the assignment of contract rights allows a debt not only to be owed to the specific creditor who is party to the original contract, but also subsequently to be treated as part of a system of credit in which debts circulate through the economy. Like all markets, the market for such debt is an omnilateral mechanism of exchange, because the value that the market establishes for any item that any person holds reflects a quantitative comparison with the value of all the other items that all the other persons hold. This comparison transcends the bilaterality of contractual relationships. In the history of English law, the difference between the bilateral and the omnilateral aspects of the contractual relationship is reflected in the different attitudes towards assignability at common law and at equity. The common law’s historic antipathy to assignability was rooted in the idea that a contract ‘presupposes a personal relation between two individuals’, and ‘a personal relation in the very nature of things cannot be assigned’.31 Courts of equity did not share this preoccupation with the bilaterality of the contractual relationship, but were able to create new equitable rights that enabled third parties to enjoy the benefit of assignment.32 In Kantian terms, the common law courts started out by adopting the perspective of private right, whereas the courts of equity were more sensitive to the systematizing aspirations of the civil condition. A principal task of the law governing assignment is to allow the omnilaterality of assignment to co-​exist with the bilaterality of contract.33 The leitmotif of this co-​existence is that assignment cannot materially prejudice the obligor’s position under the contract. Assignment is ineffective if it would ‘materially change the duty of the obligor, or materially increase the burden or risk imposed on him by his contract, or materially impair his chance of obtaining return performance, or materially reduce its value to him’.34 Several doctrines work towards preserving the position of the obligor. First, the assignee’s power to enforce the assigned right is subject to the equities—​the defences, limitations, and set-​offs—​as they exist between the obligor and the assignor prior to the obligor’s receipt of notice of the assignment. Secondly, if on construction of the contract the benefit involves the personal skill of, or personal confidence in, the obligee (the relationship of author and publisher is a paradigmatic example35), the benefit is unassignable without the obligor’s consent. Thirdly, the contract itself can stipulate that the benefit is not to 31 JB Ames, Lectures in Legal History (Harvard UP 1913) 210. 32 J Edelman and S Elliott, ‘Two Conceptions of Equitable Assignment’ (2015) 131 LQR 228. 33 In this respect, assignability differs from negotiability, which functions solely as a matter of public right divorced from the bilateral relationship in which the negotiable instrument arose. 34 Restatement of the Law, Second, Contracts (1981) s 317(1)(b). 35 Griffith v Tower Publishing Co [1897] 1 Ch 21.

84  Ernest J. Weinrib be assigned, thereby accommodating the obligor who does not want the burden of remaining alert to the receipt of a notice of assignment or who wishes to continue to be able to set up defences against the obligee.36 The point of such non-​ assignment clauses has been graphically described: ‘[A]‌debtor should be able to choose to make sure that if he or she goes to sleep next to Portia, he or she will not wake up entwined in the arms of Shylock.’37 The upshot of these doctrines is that the possibility of assigning a contractual right enlarges the reach of the contract beyond the contracting parties while still respecting the priority of the bilateral relationship between them. Within a system of reciprocal freedom, the idea of public right provides the normative space for this enlarged reach. On the one hand, a contract bilaterally links two parties in a transaction that expresses their reciprocal freedom. On the other hand, assignment operates omnilaterally to provide for the movement of the contractual benefit from the obligee to someone else. The bilateral and omnilateral dimensions of contract come together in the assignment doctrines that preclude prejudice to the obligor. These doctrines allow for the enjoyment of the most extensive freedom compatible with the equal freedom of parties who have contracted with each other.

c.  The Privilege to Preserve Property Even though ownership signifies the owner’s exclusive control of what is owned, the law nonetheless permits a person to use (and even to damage) another’s property in order to preserve one’s own. In the common law, the two most famous cases dealing with the privilege to preserve one’s property involve boats that are moored at docks in a storm. In the first of these cases, the court held that the dock owner must tolerate what would otherwise be a trespass; he could not, relying on the exclusivity of his property right, prevent the boat from remaining moored during the storm.38 In the second of these cases, the court held that, although the crew acted reasonably in keeping the boat attached to the dock, the owner of the boat was nonetheless liable for the damage to the dock caused by the boat’s pounding against it.39 These cases have occasioned much commentary, including suggestions that the liability for the damage to the dock has radical implications for the fault-​ based nature of tort law or for the divide between misfeasance and nonfeasance. In contrast, I suggest that the privilege arises as a matter of public right in the Kantian sense, leaving the underlying nature of tort liability unaffected. The point of the

36 R Goode, ‘Inalienable Rights’ (1979) 42 MLR 553. 37 A Tettenborn, ‘Assignments, Trusts, Property and Obligations’ in JW Neyers, R Bronaugh, and SGA Pitel, Exploring Contract Law (Hart Publishing 2009) 267, 281. 38 Ploof v Putnam 81 Vt 471, 71 A 188 (1908). 39 Vincent v Lake Erie Transportation Co 109 Minn 456, 124 NW 221 (Minn 1910).

Public Right  85 privilege is to deal with the systematic relationship between owners when circumstances endanger the property of one of them. The question of whether one can damage another’s property to preserve one’s own has a venerable history. The classic instance mentioned in writings of the Roman jurists was whether, in order to save one’s house from a spreading fire, one could create a firebreak by tearing down a neighbouring house. Roman jurists split on this issue.40 One position was that warding off a fire was no defence to a tort action for wrongful damage. Another position was that a private person could tear down a neighbouring house only if the fire had already reached that house, so that the house was doomed to destruction in any case. A third position was that, even if the house was not doomed, no tort liability existed, on the ground that damage done out of legitimate fear was not wrongful. Sifting through this diversity of opinion in the seventeenth century, the German legal thinker Samuel von Pufendorf reconceptualized the entire issue: A necessity that touches our own property apparently allows one the permission to destroy or appropriate the property of another, but with the following restrictions: that the threatened loss to our property . . . cannot be averted in any more convenient way; that we do not destroy another’s article of greater value for one of our own of less value; that we make good the value of the article if it would not have been lost anyway.41

Pufendorf ’s formulation was subsequently incorporated into the German Civil Code section on necessity: The owner of a thing is not entitled to prohibit the interference of another person with the thing if the interference is necessary to avert a present danger and the threatened damage, compared to the damage arising to the owner from the interference, is disproportionately great. The owner may demand compensation for the damage incurred.42

The Pufendorf formulation, especially as restated in its modern German form, indicates the normative structure of the privilege of using another’s property to preserve one’s own. Pufendorf assumes that preserving the endangered thing is a proper purpose, and that one is permitted to interfere with another’s proprietary right in the execution of this purpose, provided that the interference conforms to the criteria of necessity and proportionality. To a reader familiar with modern

40 Digest 43.24.7.4 (Ulpian); 9.2.49.1 (Ulpian). 41 S Pufendorf, De Jure Naturae et Gentium Libri Octo (CH Oldfather and WA Oldfather trs, Clarendon Press 1934) 308. 42 BGB 904.

86  Ernest J. Weinrib constitutional law, the relationship between the property right and the privilege is strikingly similar to the relationship between an entrenched constitutional right and a limitation of that right in accordance with proportionality analysis. Pufendorf has, as it were, formulated a private law version of the ‘Oakes test’43 for the justified limitation of the owner’s proprietary right. Once one conceives of the privilege in this way, the obligation to compensate for the damage done to the sacrificed property poses no problem. Necessity both animates and limits the justification. In order to preserve the endangered object, it is necessary that the defendant use the plaintiff ’s property, even to the extent of injuring it, if need be. It is not, however, necessary that the defendant be relieved of responsibility for the damage to the thing used. To leave the damage uncompensated would allow the defendant to leave a permanent mark on the plaintiff ’s property. This would be beyond the scope of the justification, which allows temporary use only to the extent necessary to preserve one’s own property during the emergency. Moreover, in permitting a right to be infringed, a justification limits the right but does not negate it—​indeed, if the justification did negate the right, they would be incapable of co-​existing within the same system of law. Under the privilege, the defendant commits no wrong in using the plaintiff ’s property for the justified purpose and therefore cannot be prevented from using it. Nonetheless, the property used remains the embodiment of the plaintiff ’s right. Accordingly, the defendant must pay for the damage done to the plaintiff ’s property through the defendant’s use of it.44 In drawing attention to proportionality in constitutional law, I am of course not suggesting that private law somehow anticipatorily borrowed a principle from modern constitutional law. Rather, the possibility of justifying an infringement of a right arises in many contexts, including private law and constitutional law. Justification always signifies both that a right has been infringed (that is, that something occurred that needs to be justified) and that this infringement is nonetheless not wrongful in the circumstances (that is, that the infringement is justified and not merely excused). Accordingly, it would not be surprising to find that justifying arguments exhibit a common structure wherever they appear. The difference between Pufendorf ’s formulation and modern constitutional law lies not in the structure of justification, but in the legal complexity of the situations to which they respectively apply. Modern constitutional law has developed an explicit and sophisticated formulation of the structure of justified infringements of constitutional rights—​that the infringing legislation be for a proper purpose, that it employ means suitable and necessary for this purpose, and that the benefit of achieving the purpose be proportionate to the effect on the infringed right. Pufendorf ’s formulation is a simpler version of this structure because it is

43 44

R v Oakes [1986] 1 SCR 103. Compare A Ripstein, ‘Tort Law in a Liberal State’ (2007) 1(2) JETL 1.

Public Right  87 addressing a simpler problem. Whereas the justification for infringing a constitutional right involves situating a statutory provision, which can have almost any content, within the entire constitutional order, Pufendorf ’s formulation has the narrow object of situating an infringement of a property right within a regime of property rights. The constitutional inquiry into proportionality is complex because one must compare, without any obvious common metric, the intensity of the legislation’s interference in the right with the importance of achieving the legislation’s purpose. Under Pufendorf ’s formulation the object preserved and the object sacrificed can readily be compared because value provides here, as it does always, the relational criterion for the quantitative comparison of different things. The possibility of justifying the infringement of a right is a reflection of the systematicity of public right. By allowing specific rights to be limited through arguments that justify their infringement, the law reveals that it does not regard those rights as absolute. Rather, the law’s concern is for the entire system of rights. Justifications work to adjust the effects of rights so that rights fit within the totality of conditions under which the freedom of all can co-​exist. This is the case in constitutional law, where rights are limited by principles underlying the constitutional order as a whole. In a much more modest way, it is also the case with Pufendorf ’s formulation. Understood in the light of Kant’s conception of public right, Pufendorf ’s formulation treats the rightfulness of the attempt to preserve property as implicit in a system of property. In the state of nature an owner’s property right operates unidirectionally to allow the owner to prevent others from using the property. When considered as part of a system of property rights, however, an owner’s property right is modulated by the presence of an adjacent property right. Pufendorf ’s formulation treats as justified an act that preserves to the extent possible the embodiments of both parties’ property rights. The idea of ownership is the genesis both of the owner’s entitlement to exclude and of the proper purpose for justifiably limiting that entitlement through the privilege that others have to preserve their property.45 As set out in the preceding chapter, ownership serves as the legal category that reflects the usability of things; the juridical implication of usability is that an owner has exclusive control over what is owned. In signalling a thing’s rightful eligibility to be used, usability is a normative idea that is operative on physically existing things. Pufendorf ’s formulation of the privilege to preserve property states the terms on which the usability of one thing by one owner co-​exists with the usability of another thing by another owner when the property of one of them faces destruction. If the law did not recognize this privilege, one owner would be able to exercise his entitlement to exclude, 45 Compare the description by Dickson CJC in R v Oakes (n 43) para 64 of the relation between right and limit in the constitutional context: ‘The underlying values and principles of a free and democratic society are the genesis of the rights and freedoms guaranteed by the Charter, and the ultimate standard against which a limit on a right and freedom must be shown, despite its effect, to be reasonable and demonstrably justified.’

88  Ernest J. Weinrib even if that would result in another’s property being destroyed and thereby becoming incapable of use. In contrast, the privilege, under which the owner of endangered property can temporarily ignore the exclusivity of another owner’s entitlement while nonetheless being bound to compensate for the resulting damage, enables the two interacting owners each to realize the usability of their respective things in the most extensive way possible.46 The idea that that an owner’s privilege to act for the preservation of endangered property involves a justified suspension of another owner’s entitlement to exclude has several implications. First, the privilege is directed solely towards the preservation of property, not towards the creation of opportunities for gain. It may well be that the defendant can dramatically enhance the value of her property by temporarily encroaching on her neighbour’s—​for example, by placing a crane on it to facilitate the construction of a high-​rise building. Such action is not protected by the privilege. Secondly, in order to be shielded by the justification the act has to be made for the justified purpose. The negligent destruction of another’s property that turns out to save one’s own is an unjustified wrong.47 Thirdly, the fact that the act is justified means that it is permissible, not obligatory. It is not obligatory on the owner of the saved property, because he is as free to save his property or not as he is to use it or not. Nor, a fortiori, is it obligatory on the owner of the sacrificed property. Because the act is permissible, the owner of the sacrificed property must abstain from preventing it, but is under no affirmative duty to save the endangered property. Thus, the privilege conforms to the standard notion, evidenced in the distinction between misfeasance and non-​feasance, that private law postulates no obligatory ends. One can put this last point in another way. In the civil condition usability requires the systematic alignment of everyone’s exercises of their rights of ownership. It is not an invitation to a kind of utilitarianism in which usability is a good to be maximized. If it were, everyone including the dock-​owner in the case of the battered dock would be under an obligation to save the boat. Far from being utilitarian, the usability of things is the basis for a right of ownership and for a justified limit on that right within a system of reciprocal freedom. Usability imposes no duty on anyone to undertake any positive act. Rather, the argument about ownership that it generates allows a person in the civil condition to be the owner of property, and in the circumstances of the privilege, to preserve the physical usability of one’s own property, even in the face of another owner. As with the examples of market overt, inducing breach of contract, and assignment, the privilege regarding the preservation of property is an example of the 46 A similar explanation accounts for easements by necessity. 47 Justinian’s Digest, D 43.4.7.4 (Ulpian) mentions tearing down a house when there was no fire but a fire subsequently arises that would have allowed the house to be torn down to create a firebreak. Labeo holds that one is liable for damage wrongfully caused, because one evaluates ‘non ex post facto sed in praesenti statu’.

Public Right  89 operation of public right to modify the principle of decision that would hold as a matter of private right. The logic of the concept of ownership gives the owner an entitlement to exclude. As is the case with all justifications, the privilege formulated by Pufendorf does not affect the scope or basis of the underlying right. It merely limits its operation in particular circumstances. To see in the privilege a ground for reconsidering the fault-​based nature of tort law or the divide between non-​feasance and misfeasance is to commit what Kant stigmatized as the ‘common fault of experts on right’. With respect to the privilege to preserve one’s property, public right makes available the normative space for distinguishing between the scope of the right as identified and defined in the state of nature, on the one hand, and the operation of the right within the civil condition’s system of rights, on the other. The scope of the right reflects the role of ownership in providing juridical recognition for the sheer usability of things. Scope, accordingly, refers to the owner’s unrestricted and exclusive dominion over the thing owned, including the entitlement to use the thing and to exclude others as the owner wishes. The operation of the right refers to the way the right of one owner interacts with that of another. When property is in danger of imminent destruction, the privilege to act for the preservation of one’s property bars another owner from interfering with the preservation effort through the exercise of the otherwise standard entitlement to exclude. This privilege has the legal structure of a justification for limiting an ownership right: the right continues to be valid and grounds a claim for compensation should injury occur.48 Thus, the privilege forms part of a system that provides the interacting owners with the most extensive realization of the usability of the things that they each own.

d.  Private Nuisance The distinction between the scope and the operation of an owner’s right is also central to the law of private nuisance. A private nuisance is an interference with the use and enjoyment of land. In the central case of interferences that emanate from a neighbour’s land, the law of nuisance protects the use that one person might make of his or her property while restricting the other person’s use. Yet the law’s impingement upon the parties’ respective uses when these turn out to be incompatible does not affect the underlying idea that ownership gives unrestricted dominion over

48 The notion that a justified limitation on a right does not affect the right’s scope is routine in constitutional jurisprudence. See A Barak, Proportionality: Constitutional Rights and their Limitations (CUP 2012) 19: ‘The modern theory of constitutional rights . . . draws a fundamental distinction between the scope of the constitutional right and the extent of its protection. The scope of the constitutional right marks the right’s boundaries and defines its content. The extent of its protection prescribes the legal limitations on the exercise of the right within its scope.’ As Barak observes at 27, for most constitutional rights ‘[t]‌he extent of their protection is narrower that their scope’.

90  Ernest J. Weinrib what is owned. The scope of the owner’s right—​in Kantian terms, how that right is identified and defined in the state of nature—​goes hand in hand with the operation of that right within a system of rights—​Kant’s civil condition—​that requires one person’s use to be accommodated to the right of another. While both the privilege to preserve property and the law of nuisance presuppose the distinction between the scope and the operation of an owner’s right, they respectively endow that distinction with a different form. In the case of the privilege to preserve property, one person is using the property of another—​in Pufendorf ’s words, acting ‘to destroy or appropriate’ it—​in order to save his or her own. Considered on its own, the use of property without the owner’s consent is wrongful unless it can be justified, that is, unless grounds can be adduced for regarding this apparently wrongful act as rightful after all. The privilege accomplishes this by exhibiting the way in which necessity and proportionality combine to constitute a justification of an otherwise wrongful act. In the case of nuisance, in contrast, the situation involves the use not of another’s property but of one’s own. Because land is bounded and immoveable, the inescapable contiguity of the neighbouring owners and the freedom inherent in ownership for land to be used as its owner wishes give rise to the possibility that one owner’s use may interfere with another’s. Considered in themselves, neither of these uses is wrongful, but they are incompatible when juxtaposed. The resolution of the resulting ‘troubles de voisinage’ (as they are termed in French law)49 calls not for a privilege that justifies what is otherwise wrongful but for a body of doctrine that sets out the terms on which one owner’s use has to adjust to another’s. The law of nuisance supplies that body of doctrine. The contrast between ownership, which renders the owned thing totally subject to the will of the owner, and nuisance, which restricts the owner’s use in order to avoid disturbing a neighbour, has long aroused perplexity, especially among civilian jurists.50 How can the absoluteness of the owner’s power to use be reconciled with the limitation of that power through the law of nuisance? The law governing the way that ownership operates seems to be at odds with what ownership is. The transition from the private right of the state of nature to the public right of the civil condition addresses this perplexity. Instead of directly juxtaposing ownership and nuisance, the Kantian account arranges them sequentially, first establishing the normative validity of its conception of ownership in the state of nature and then dealing with the role of the civil condition in actualizing ownership so conceived. Neither of the poles in the transition from the state of nature to the civil condition effaces the other. The usability of things leads to the conception of ownership as a right that subjects what is owned to the despotic will of 49 J Gordley, ‘Immisionsschutz, Nuisance and Troubles de Voisinage in Comparative and Historical Perspective’ (1998) 6 ZeuP 13. 50 ibid.

Public Right  91 the owner. However, such ownership cannot be effective without the systematicity that characterizes the civil condition. Unless the rights of owners are systematically ordered towards one another, ownership as dominion over the owned thing would be self-​stultifying in the face of incompatible uses of different properties, because one owner’s use would interfere with another’s. The extent to which owners could realize the usability of their respective lands would be hostage to the conduct of neighbouring owners. Instead of demarcating areas of reciprocal freedom, the ownership of land would expose each owner to domination by other owners, as the more obtrusive uses would render the less obtrusive ones unsustainable. Through their unrestricted powers to use the land they each own, owners would be engaged in a war of all against all, so that none of them could be secure in his or her use.51 Thus, the very conception of ownership as the recognition of usability requires that use be limited by the law of nuisance when uses by interacting owners cannot co-​exist. On the Kantian account, nuisance law, like ownership, is the expression of the reciprocal freedom of the parties involved. The preceding chapter described how ownership satisfies reciprocal freedom. What about the law of nuisance? The tort of nuisance deals with the competing uses of land against the background of the notion of usability that, on the Kantian analysis, underlies the phenomenon of ownership. The gist of the tort of nuisance, as a ‘tort against land’, is the defendant’s interference with the land’s usability, that is, with its capacity to be used and enjoyed by its owner for ordinary purposes.52 The particular uses that the parties make of their respective lands are therefore evaluated with reference to the ordinariness of each party’s use relative to the other’s. Under the law of nuisance, when uses are incompatible an ordinary use triumphs over one that is not ordinary.53 Accordingly, at the doctrinal level the law establishes several contextual indications of what is ordinary among owners of adjacent properties. One such indication is the overarching entitlement to non-​interference with ‘the ordinary comfort physically of human existence’.54 A second is reference to the standard of the locality to determine what is ordinary within the relevant area.55 A third disallows claims based on the plaintiff ’s abnormal sensitivity.56 A fourth is found in law’s adoption of the principle of ‘live and let live’ for the minor disturbances that 51 ibid 17, where Gordley quotes the observation of Charles Demolombe, that if all proprietors could ‘invoke their absolute right, it would be clear that none would have one in reality . . . It would be war! It would be anarchy!’ 52 D Nolan, ‘The Essence of Private Nuisance’ in B McFarlane and S Agnew (eds), Modern Studies in Property Law, vol 10 (Hart Publishing 2019) 71; D Nolan, ‘ “A Tort Against Land”: Private Nuisance as a Property Tort’ in D Nolan and A Robertson (eds), Rights and Private Law (Hart Publishing 2011) 459. 53 EJ Weinrib, The Idea of Private Law (OUP 2012) 192–​94. 54 Walter v Selfe (1851) 64 ER 849, 852. 55 Sturges v Bridgman (1879) 11 ChD 852, 865 (CA). 56 Rogers v Elliot, 15 NE 768 (Mass SJC 1888).

92  Ernest J. Weinrib inevitably accompany ordinary use and that are thus a normal part of the mutuality of interactions among owners.57 These doctrines facilitate the assessment of the parties’ respective uses of their land in the light of what counts as an ordinary use in their situation. Parties whose uses conform to what is ordinary treat one another equally as owners, because they each allow the other what they take for themselves. Conversely, a use that is not ordinary is inconsistent with such equality, because it attempts unilaterally to set the terms of the relationship between one owner and another. Accordingly, if owners’ uses are incompatible, nuisance law requires the use that is not ordinary to give way to the ordinary uses upon which it impinges. In this way, nuisance law arranges for the co-​existence of neighbouring owners on terms of their reciprocal freedom with respect to the uses to which they put their property. Thus, nuisance law provides a special set of principles for dealing with the operation of ownership rights in land while leaving unaffected the scope of those rights. On the Kantian account I am presenting here, the normative space for the distinction between scope and operation is created by differentiating between private right as modelled in the state of nature and public right as actualized in the civil condition. The civil condition has a systematizing dimension that comes into play when the uses to which owners put their land—​authorized though these uses may be by the conception of ownership taken on its own—​are not mutually compatible. The law of private nuisance is the expression of this systematizing dimension. By setting out contextual criteria of ordinary use and by requiring that owners whose uses are not ordinary bring themselves into line with the ordinary uses of their neighbours, the law of private nuisance, like the privilege to preserve property, allows each owner the most extensive realization of the usability of what is owned consistent with the reciprocal equality of other owners.

5.  Conclusion In this chapter, I have set out the implications for rights of their being situated within an authoritative system of public institutions. Those institutions presuppose the omnilaterality of legal relations in what Kant calls ‘the civil condition’. In exploring the significance of this omnilaterality, I have drawn attention not merely to the role of institutions in determining and enforcing the rights, but to the ideas of publicness and systematicity that introduce distinctive considerations into the doctrinal corpus of private law. What, then, is the character of private law when subjected to the institutional guarantees of public right? In the Kantian view, those institutions are the products



57

Bamford v Turnley (1862) 122 ER 27, 33.

Public Right  93 and representatives of the united will of all, which connects everyone to everyone else. The omnilaterality of this will not only grounds the authority and legitimacy of courts, as well as of other public institutions; it also informs the judicial treatment of relations among right-​holders. Under public right the publicness and systematicity of the legal order as a whole warrant the adjustment of one person’s rights and freedoms because of the presence of someone else’s. Right-​holders thereby become reciprocally determining participants in the legal system. The reciprocal determination of elements that exist independently of one another but are nonetheless combined together into a single whole is the essential feature of Kant’s conception of community.58 On the model of Kant’s famous characterization of morality as forming a kingdom of ends,59 one might regard public right as transforming private law into a community of rights.

58 I Kant, Critique of Pure Reason (P Guyer and A Wood trs, CUP 1998) B111–​13. 59 I Kant, Groundwork of the Metaphysics of Morals in the Cambridge Edition of the Works of Immanuel Kant: Practical Philosophy (Mary J Gregor tr, CUP 1999) 4:433.

5

Distributive Justice 1.  The Pluralism of Justice So far, this book has elaborated the theoretical implications of understanding private law as corrective justice. Starting with the notion that liability in private law corrects an injustice inflicted by one person on another, the preceding chapters have inquired into the structure of the private law relationship and into the content adequate to that structure. Chapter 1 argued that the structure refers to the correlative normative positions of the potential parties to a finding of liability. Chapter 2 claimed that the content for this structure is found in the correlativity of right and obligation within a Kantian conception of reciprocal freedom. Illustrating the contention that the Kantian idea of a right comprehends and unifies different Hohfeldian categories, Chapter 3 set out the normative argument for understanding ownership as integrating the liberty to use and the claim-​right to exclude in a way that is expressive of the reciprocal freedom of all. Chapter 4 then explored, again from the standpoint of reciprocal freedom, the effect on private law of a regime of authoritative institutions with its dimensions of publicness and systematicity. In this chapter, I consider the question of how corrective justice fits more broadly into the legal order. The chapter moves beyond private law to the wider legal world, where private law’s grounds of liability co-​exist with other arrangements (workers’ compensation is an example) that submit particular areas of social life to state regulation. Whereas private law actualizes corrective justice, these other arrangements are instances of distributive justice. The issue for this chapter is: how are these two kinds of justice related? As Aristotle noted long ago,1 both distributive justice and corrective justice are structural ideas. Whereas corrective justice refers to the structure of relationships in which one party can be held liable to the other, distributive justice refers to the structure of relationships in which a benefit or burden is divided among several recipients. For instance, workers’ compensation in many jurisdictions is a double exercise in distributive justice: employers are subject to the burden of paying into a fund in accordance with an industry’s accident history, and employees receive benefits from that fund in accordance with their workplace injuries. Structurally,



1 Aristotle, Nicomachean Ethics, 5, 2–​4.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0005

Distributive Justice  95 any particular distribution involves the conjunction of three elements: a benefit (or burden) to be distributed, a set of persons to whom that benefit is to be distributed, and the criterion that governs the distribution. The criterion acts as a unifying principle for linking the benefit to the persons, thereby relating every participant in the distribution to every other participant in an ordered way. The tighter the connection between the criterion, on the one hand, and the persons and benefits that it governs, on the other, the more coherent is the distributive arrangement. Conversely, a distribution is incoherent to the extent that it operates under-​inclusively or over-​inclusively, giving participants either more or less than they merit under the criterion of distribution, or dividing among the participants a benefit that fails to fulfil the criterion’s purpose, or employing a criterion that does not felicitously link the benefit being distributed and the persons among whom it is being divided. Distributive justice is thus a structure through which persons can coherently be related to one another with respect to whatever is being distributed in accordance with its governing criterion. As with corrective justice, distributive justice has its implicit conception of coherence and fairness. The coherence consists in the ordering that the distributive criterion imparts to the relationship among the participants in the distribution. The fairness refers to each person’s equal standing within the distribution in terms of that criterion. Because both coherence and fairness in the distributive context are functions of the interconnection between the distributive criterion, the benefit and the participants, it is the case for distributive justice no less than for corrective justice2 that no incoherence can be fair and no fairness can be incoherent. Thus, as structural ideas, corrective justice and distributive justice represent divergent ways in which legal relations between persons can be fair and coherent. The contrasts between them are manifold. First, whereas corrective justice pertains to the immediate interaction of two persons as the doer and sufferer of the same injustice, distributive justice mediates an interaction among persons by relating them through a distributive criterion. Secondly, corrective justice focuses on whether an inconsistency with the plaintiff ’s right can be attributed to the defendant; distributive justice, in contrast, marks out a benefit or burden to be distributed by reference to a criterion that determines each person’s share of that benefit or burden. Accordingly, corrective justice governs the interaction between plaintiff and defendant by the motto ‘To each his own’; distributive justice governs the interaction between participants by the motto ‘To (or from) each according to the criterion of distribution’. Thirdly, corrective justice and distributive justice employ different internal logics. Corrective justice features the logic of correlativity, under which the law elaborates and applies concepts that place the parties in normative positions that are the mirror images of each other. Distributive justice employs the logic of



2

See ch 1, s 6.

96  Ernest J. Weinrib comparison, in which burdens or benefits are determined by how persons stand to one another with respect to the features that the distributive criterion makes relevant. Fourthly, corrective justice links only two persons, because correlativity is a pair-​wise operation; distributive justice, can link an indefinite number of persons because comparison implies no numerical limit on who is compared. These interconnected contrasts show that corrective justice and distributive justice are categorically different and mutually irreducible. The immediacy of an interaction cannot be understood in terms of mediation by a distributive criterion, nor is what one has as one’s own the same as what should be received through a distribution, nor is the logic of correlativity intelligible as a comparison, nor can a relationship limited to two persons be open to an indefinite number. Corrective justice and distributive justice each give shape to legal justifications that can operate fairly and coherently in the relationships that instantiate them, but they do so in completely dissimilar ways. From this an important implication follows. If one thinks of justice in terms of justificatory structure, that is, in terms of the fairness and coherence of the justifications for particular legal arrangements, there is no justice tout court. Rather, justice is a pluralism of different kinds of justificatory structure. Aristotle himself affirmed this when he originally introduced the distinction between corrective and distributive justice. ‘It seems’, he remarks, ‘that justice and injustice are spoken of in many senses’.3 In Aristotle’s account, justice is an omnibus term referring to the norms that govern one person’s relation towards another, but the categorical distinction between corrective justice and distributive justice indicates that the preposition ‘towards’ has radically different meanings. Given this categorical distinction, how are we to understand the relationship between corrective and distributive justice? In particular, is there any way of nonetheless conceiving of them as forming a juridical unity within the state as a whole? By ‘juridical unity’, I mean this: as mentioned earlier,4 the positive and the juridical are two aspects of legality, the positive pertaining to the validity of law, the juridical pertaining to the fairness and coherence implicit in the structure of particular legal relationships. In its positive aspect, the unity of a legal system comes from the connection of all its legal norms to a single source of validity.5 In its juridical aspect, the unity of a legal system reflects the singleness of the justificatory grounding for all legal relationships. The positive aspect deals with the form of legal norms as valid, whereas the juridical aspect deals with the content of legal relationships as justified by reference to their fairness and coherence. In previous chapters, I have presented corrective justice as having a distinctive conception of the juridical, under which the correlatively structured relationships of private law participate in a public

3 Aristotle, Nicomachean Ethics 1129a26; compare 1130b6: ‘the justices are plural’. 4 5

See ch 1, s 7. H Kelsen, General Theory of Law and State (A Wedberg tr, Harvard UP 1945) 110–​11.

Distributive Justice  97 system of rights grounded in the reciprocal freedom of all. Now I ask: Can we also understand the state’s exercises of distributive justice as expressive of reciprocal freedom? And are the two forms of justice normatively connected within a system of rights? The juridical unity of the legal order presupposes an affirmative answer to these questions. At first sight, the pluralism of justice evidently challenges the notion that the legal order has a juridical unity. Given the different structures of justification at play in different kinds of legal relationship, juridical unity cannot consist in the uniformity of justifications across the entire range of legal phenomena. In this respect, the operation of a system of rights that incorporates both corrective and distributive justice differs from utilitarianism or the economic analysis of law, where a single decision criterion is applied to every situation. Nor can it be supposed that one of the structural ideas must be illegitimate, so that the other should be regarded as occupying the entire space of justification under law. A proponent of corrective justice would then be a libertarian for whom corrective justice is the only justice there is, whereas a proponent of distributive justice would be a collectivist for whom every legal operation without exception instantiates some pattern of distribution. Aside from being inconsistent with the legal praxis of modern states, the supposition that only one form of justice is legitimate is theoretically unwarranted. Because they are categorically different structural ideas, neither corrective justice nor distributive justice refers to the other. Consequently, neither offers a reason to eliminate the other. In this chapter, I suggest that there is nonetheless a conception of juridical unity that encompasses both corrective and distributive justice and establishes the normative connection between them. This is the unity of a conceptual sequence.6 Viewed as the components of a system of rights, corrective justice and distributive justice participate in a progression, the stages of which are tied together by the idea that inherent in reciprocal freedom is the mutual independence of persons. This conceptual sequence moves from its starting point in the reciprocal freedom of innate right to the laws and institutions that actualize corrective and distributive justice. Each stage presupposes and complements the previous one without invalidating it, so that the reciprocal freedom with which the sequence begins remains thematic throughout the entire progression. A conceptually sequenced argument such as this one has certain requirements. It must indicate not only the sequence of concepts but also why the concepts belong together, why they refer to certain normative considerations rather than others, why these different sets of normative considerations should be kept separate and in the sequenced order, why the sequence cannot stop at the first stage 6 On the role of sequence in ordering different values see GP Fletcher, ‘The Right and the Reasonable’ (1985) 98 Harv L Rev 949, 950–​54, distinguishing between ‘flat’ and ‘structured’ legal thinking; see also J Rawls, Political Liberalism (Columbia UP 1993) 259–​62, discussing unity by appropriate sequence.

98  Ernest J. Weinrib but must go on to the subsequent ones, and why the subsequent ones do not simply efface the earlier ones. Accordingly, this chapter sets out why the system of rights is not limited to corrective justice, what conception of distributive justice has a sequenced position within the system of rights, and how that conception nonetheless leaves corrective justice intact at the sequence’s earlier stage. By focusing on the idea of a conceptual sequence, this chapter elucidates a feature of the Kantian approach that has already been mentioned in previous chapters. The discussion of ownership in Chapter 3, for instance, emphasized the importance of considering ownership and acquisition sequentially. Ownership, it was contended, had to be conceivable as a viable legal category under the principle of right before one addressed the problem of how its acquisition was compatible with the reciprocal freedom of all. In Chapter 4, the sequence evident in the treatment of ownership was presented more broadly as a movement from the state of nature to the civil condition, and thus from the bilateral logic of the private law rights to the omnilateral institutional context in which those rights operate publicly and systematically. In those chapters, the idea of a conceptual sequence was invoked to deal with different aspects of private law. The present chapter has a more radical ambition: to show the relevance of conceptual sequence for the pluralism of justice, and thus for bridging the divide between private law and public law while nonetheless leaving the two distinguishable. On the methodological plane, the unity of a system of rights can be explicated only through a conceptual sequence. The root idea of a system of rights is the reciprocal freedom of all. Such freedom is an abstraction that is inseparable from juridical relations and that gets realized only through the operation of legal institutions. It is rendered determinate through concepts of greater and lesser generality until it can be actualized in a legal order and applied to the specific interactions of specific persons. Until this point of application is reached, each stage in the sequence requires specification in a further stage, and every such further stage seeks consistency with the stages that preceded it. The result of this process of progressive specification is that the system of rights as a whole is the sequential articulation of a totality animated by and realizing the idea of reciprocal freedom. Because of the intimacy of its connection with the juridical relations necessary to realize it, reciprocal freedom is on a different methodological grid than are consequentialist principles such as the principle of utility. The principle of utility states a goal that consists in maximizing whatever promotes the greatest happiness for the greatest number. This goal is not intrinsically dependent on law; the institutions and doctrines of the legal order are merely contingent means for contributing to the goal’s achievement. Being conceptually independent of law, the principle of utility provides the external measure of the desirability of any element of law and of the conduct of every official and every individual. Unity consists in the reference of the components of the legal order to the exogenous master principle that they all serve.

Distributive Justice  99 Reciprocal freedom does not work like that. At its most abstract, reciprocal freedom refers not to a material interest whose maximization counts as the criterion of success, but to the form of a relationship between persons in which their freedom is compossible. In rendering this form into more determinate arrangements, the legal order is not the means for the achievement of reciprocal freedom but the medium through which it exists and in which it exhibits its meaning. Its normative dynamism comes not from outside but from within its own articulation and development. In the absence of reference to a material interest at a foundational level, the consequentialist methodology gains no foothold. Reciprocal freedom does not, therefore, stand to the legal order as a consequence that the legal order produces but as the normative idea that is intrinsic to its character as a juridical phenomenon. In order to operate as the governing idea for juridical relations, reciprocal freedom has to be determined through the specification of the legal categories and institutions that are in conformity with it. Without such specification, reciprocal freedom can be neither understood nor enjoyed. Accordingly, the point of the Kantian approach is to show how legal categories and the public status of legal officials figure as conceptual stages in reciprocal freedom’s unfolding. Far from having a meaning that is independent of law, reciprocal freedom becomes intelligible only through the elucidation of the various rights that express it and of the various legal institutions that make those rights effective. Between the general idea of reciprocal freedom and its application by public officials to the particularities of social life lie a series of intermediary concepts that are specifications of that general idea.7 The very abstractness and formalism of reciprocal freedom make such specifications indispensable to freedom’s operation within a functioning legal system. Moreover, given the diversity of these intermediary concepts—​some of which are substantive and some of which are institutional, some of which pertain to private law and some of which pertain to public law—​they cannot each be developed directly out of reciprocal freedom but must be developed seriatim out of one another. Hence, the sequence in which the intermediary concepts are elucidated is crucial to our understanding both of the system of rights that they generate and of the idea of reciprocal freedom that they explicate. In other words, because reciprocal freedom is a principle that pertains to actual human interactions, the intermediary concepts flesh out both how reciprocal freedom is realized and what reciprocal freedom means within a particular system of rights. If, then, both corrective justice and distributive justice are present within this conceptual sequence, the unity that binds them together cannot be either of the diverse kinds of justifications whose structure they respectively represent. Rather, the unity must reside in the internal integration of the sequence itself as it comprehensively realizes the reciprocal freedom of all. The operation of this sequence, 7 I Kant, ‘On a Supposed Right to Lie from Philanthropy’ in I Kant, Practical Philosophy (M Gregor tr and ed, CUP 1996) 614–​15 (8:429–​430).

100  Ernest J. Weinrib extending from the state of nature to the kind of distributive arrangements characteristic of the modern state, is the subject of this chapter.

2.  Reciprocal Independence In this conceptually ordered sequence distributive justice supplements private law while leaving it intact. The point of this supplementation is to enable the system of rights as a whole to realize the reciprocal independence of all persons in their relations with one another. This independence underlies distributive justice no less than corrective justice, because in a legal order based on rights, reciprocal independence informs all legal relationships, whether those between one person and another or those between a person and the state. Kant designates independence from others as the defining characteristic of reciprocal freedom. He describes the freedom pertaining to the innate right—​the subjective right at the core of the system of rights—​as ‘independence from being constrained by another’s choice’.8 This independence is the opposite of the subordination that consists in being subject to or dependent on another’s will. The integrity of one’s body from invasion by another is the necessary condition of such independence, because for human beings no freedom is possible to the extent that our bodies, the organs through which we act, are subject to the control of others. The innate right, however, is not the right to something considered on its own. Just as dependence and independence are always on and from something, so innate right is a relational term that refers to the co-​existence of one’s freedom with everyone else’s. From its semantic composition one might think that independence (‘Unabhängigkeit’ in German) is just the negation of dependence. However, Kant’s exposition of the various entitlements under innate right9 makes it clear that independence has both a negative and a positive side. On the negative side is the freedom from being constrained by another’s will or from being non-​reciprocally bound by another. The negative side precludes others from doing with you what they wish. The corresponding positive side is having one’s fate in one’s own hands by being possessed of a power to act in accordance with one’s own will so far as one’s relations with others are concerned. Put in the terminology of Roman law where the phrase refers to not being subject to another’s legal power (alieni iuris, as the Romans would say), the positive side of independence is that everyone is sui iuris. This means that everyone counts juridically as one’s own person, and therefore, in interacting with others, can assert oneself both in word and in deed. Asserting oneself in word means that one can say whatever one wishes, whether true or untrue,

8 9

I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 34 [6:237]. ibid 34 [6:237–​38]. See ch 2, s 3.

Distributive Justice  101 so long as the utterance does not injure another’s rights. Asserting oneself in deed means that doing wrong is a matter of one’s conduct and not of one’s existence, of what one has done rather than of than of who one is. Persons who are reciprocally independent in these ways carry on their lives on their own initiative and responsibility, being both unconstrained in their rightful conduct by others’ wills and answerable for their wrongful interferences with others’ rights. The negative side of independence, accordingly, precludes one from being a means for others or from treating others as means, whereas the positive side consists in the entitlement to assert one’s worth as an end in relation to others who are also ends. In this way the actions of all are compatible with everyone’s freedom to make one’s way in the world on one’s own terms.10 Thus, the positive and negative sides of independence form the mutually constituting aspects of the innate right as a single unified right. In the state of nature this idea of independence has a circumscribed range of application for two interlocking reasons. First, the state of nature is a condition in which juridical relationships are always and only bipolar, because the omnilateral relationships of the civil condition have not yet arisen. The sole juridical issue is whether one person has infringed another’s innate right; this issue can be framed only as between two persons. Although each person’s innate right is binding on every other person, its validity is not ‘against the whole world’—​there is as yet no legal world that can be regarded as a totality—​but against each and every individual person in turn. Secondly, acquired rights (including property rights) that are conclusively and not merely provisionally binding do not exist in the state of nature, because the acquisition of such rights is consistent with the innate right only through the omnilaterality of the civil condition.11 Consequently, given the absence of the mutual assurance about acquired rights that comes from the civil condition, encroaching on what another possesses is not an injustice to the possessor.12 The result of the combination of these two circumstances is that independence is violated only when one person invades or threatens to invade another’s bodily integrity.13 In the civil condition, the situation changes dramatically on both counts. While the innate right remains intact, other kinds of rights have accrued; and relationships can be omnilateral as well as bipolar. In addition to a right to bodily integrity, persons can now conclusively acquire rights to property, to contractual performance, 10 ibid 32 [6:236]. 11 See ch 3, s 5. 12 Kant (n 8) 93 [6:307]. 13 This sentence is a slight overstatement for the sake of clarity. In addition to bodily integrity, the innate right includes the entitlements (mentioned above in ch 2, s 3) to communicate one’s thoughts, to be one’s own person, to be subject only to obligations that are reciprocal, and to be considered beyond reproach. Derived from the last of these is the entitlement to one’s good name. These qualifications do not affect the implications that I am drawing from the contrast between the state of nature and the civil condition.

102  Ernest J. Weinrib and to fiduciary entitlements. They are now related to each other not only through the constraint that one applies or does not apply to another’s body, but also through the rights and obligations consequent on their acquisitions. And because these rights are entrenched in the positive law of their civil condition, juridical relationships are not merely bipolar but also publicly and systemically omnilateral. This transformation has two effects on the idea of independence. First, the addition of non-​provisional and publicly enforceable acquired rights to the catalogue of subjective rights extends the range within which right-​holders can carry on their lives on their own responsibility and initiative; concomitantly, it increases the possible wrongs that a person can suffer at another’s hand. Independence is now actualized not only through each person’s right of bodily integrity but also through a new set of bipolar relationships that center on the rights one has acquired and on the wrong that would be done to a right-​holder by damage to or unconsented use of the objects of those rights. In particular, the conception of ownership as an unlimited submission of a thing exclusively to the owner’s will14 endows the right to one’s person and the right to one’s property with a similar bipolar significance for the relationship between the right-​holder and every other person in the world. For rights both innate and acquired, their bipolar operation is the realm of corrective justice. Secondly, however, this expansion of the range of independence both by adding rights and by allowing them to be omnilateral creates a new problem of subordination. This problem becomes evident when the state of nature is compared to the civil condition. In the state of nature for all its limitations, the innate right represents a rigorous norm of mutual independence. As a juridical idea, independence in the state of nature reflects the range of action to which our reciprocal freedom entitles us, whereas subordination consists in being subjected to the wrongful actions of another. The only form of subordination pertinent to the state of nature is the wrong that consists in one person’s body being subject to another’s constraining will. In all other interaction, however unsatisfactory their lives may be in other respects, the interacting parties are fully independent from one another. This independence is unaffected by the availability or unavailability of material goods. Such goods are equally open to all. In the absence of interference with anyone’s body, everyone’s use of the things of the world in pursuit of one’s self-​ determined end is an exercise of independence consistent with the independence of others. For example, although the apple in your hand is not accessible to me because my seizing it would affect the disposition of your fingers around it and therefore violate your innate right, the apple in the tree remains available. And, although I may starve if the apple in the tree is beyond my reach, that misfortune is not the



14

See ch 3, s 4.

Distributive Justice  103 product of a violation by anyone of reciprocal freedom. Your failure to fetch me the apple that I could not get for myself is not a wrongful act; it is a mere non-​feasance. Although my body may deteriorate as a consequence of your inaction, you have not subjected it to your will. Nor can I rightfully compel you to use your greater height to secure the apple for me, because that would constrain you to my will and thus wrongfully trench on your independence. The reason that material goods do not affect independence in the state of nature is that the state of nature models only the juridical relationships that are operative in it. It is not concerned with the fate of its imagined participants. The issue is not whether I will starve unless you get me the apple from the tree. Because acquisition cannot occur in the state of nature, the apple is not a component of our juridical relationship. Therefore, I cannot make a rightful demand on you with respect to the apple or any other thing. Rather, the issue is whether the terms of our interaction as bearers of innate right infringe on our independence from each other. In the state of nature, persons relate only to each other and only through the embodiments of their physical existences. With respect to their bodies they interact as equals, using their bodies as the organs for their independent purposes, subject to being reciprocally limited by the bodies of others. The misfortune of my starvation in the wake of your inaction is not the consequence of a wrong that you have committed. Nor can it be regarded as a systematic defect inherent in the structure of our relationship: in the state of nature, aside from one person’s action on another, there is no system to which a defect can be ascribed. Accordingly, no-​one can be excluded from whatever material goods there are, but the sufficiency of those material goods for anyone’s needs and the ability to access them are irrelevant to the only kind of subordination that matters in the state of nature. The transition from the state of nature to the civil condition, however, presents a new challenge to the independence of persons. Now the state makes possible and secures legal relations among persons through things. The law’s recognition of ownership allows the owner to exclude others even from things to which the owner is not physically attached. Although I may be in danger of starving to death unless I get an apple, you may be the owner of the apple and, therefore, have exclusive right to it even when you put it down or add it to your pile of apples. The state thereby enables you to shut me out from what would otherwise be accessible. The existence of property in the civil condition introduces the possibility of accumulation, and with it the possibility of subordination for those whose sphere of action is confined to what is left over. For some of them, what is left over may be insufficient for making one’s way in the world on one’s own. Indeed, their very survival may be dependent on the compassion or generosity of others, or they may be able to purchase their survival only by subjecting themselves to others as their bondsmen.15 Even if they have suffered no wrong against their persons and 15 An example is the ancient Roman practice of nexum; see A Berger, ‘Encyclopedic Dictionary of Roman Law’ (1953) 43 Trans Amer Phil Soc 333, 595.

104  Ernest J. Weinrib against such property as they have, their condition may be so desperate that they find themselves at the mercy of another’s will, exposed to the practical domination of others even in a legal system that forbids anyone from being their legal owner or dominus. Thus, whereas the state of nature is a world of pure independence in which subordination to another’s will is exemplified by wrong, the civil condition is a world of more extensive but adulterated independence, in which subordination can arise through the systemic operation of rights. This form of subordination to the will of others is unconnected to injustice within the bipolar relationships of private law. The gist of subordination in this context is not that someone has wronged the bodily integrity or the acquired rights of others as a matter of corrective justice, but that the entire system of rights has resulted in some persons’ being subordinated to others. Such subordination may be consistent with the most scrupulous observance of the rights of contract, of property, and of bodily integrity. Indeed, private law itself—​for example in contracts between parties who stand on a grossly unequal footing—​may be the instrumentality that reflects and then further entrenches the practical constraint exercised by the stronger party on the weaker one.16 Subordination may thus emerge from transactions that are unimpeachable from the standpoint of corrective justice and from injuries that are suffered without anyone else being at fault or civilly liable. A system of rights cannot be indifferent to the form of subordination that it itself has enabled. Under a system of rights, the reciprocal independence of persons has to inform every juridical relationship: not only the bipolar relationships of corrective justice, but also the omnilateral relationships in the civil condition that run between each person and the totality of all other persons. Juridical relations now exist not only between individuals but also between an individual and the state constituted by this totality. Even though in creating this possibility of dependence the system of rights does not wrong persons from the standpoint of corrective justice, it nonetheless fails them from the broader standpoint of its own systemic rights-​protecting function if it does not anticipate or alleviate the dependences that actually come into being.

16 West Coast Hotel v Parrish 300 US 379, 398–​99 (1937): ‘What can be closer to the public interest than the health of women and their protection from unscrupulous and overreaching employers? And if the protection of women is a legitimate end of the exercise of state power, how can it be said that the requirement of the payment of a minimum wage fairly fixed in order to meet the very necessities of existence is not an admissible means to that end? The legislature of the State was clearly entitled to consider the situation of women in employment, the fact that they are in the class receiving the least pay, that their bargaining power is relatively weak, and that they are the ready victims of those who would take advantage of their necessitous circumstances. The legislature was entitled to adopt measures to reduce the evils of the “sweating system,” the exploiting of workers at wages so low as to be insufficient to meet the bare cost of living, thus making their very helplessness the occasion of a most injurious competition . . . The exploitation of a class of workers who are in an unequal position with respect to bargaining power, and are thus relatively defenseless against the denial of a living wage, is not only detrimental to their health and wellbeing, but casts a direct burden for their support upon the community.’

Distributive Justice  105 Accordingly, reciprocal independence continues to be normative for the civil condition, albeit transformed from what it was in the state of nature. In an implicit criticism of Hobbes, Kant insisted that the legitimacy of the state requires that there be no sacrifice of one’s innate freedom in the transition from the state of nature to the civil condition; one merely gives up one’s wild lawless freedom to find freedom undiminished under law.17 Because freedom pertains to relations between persons, it is not a concept that is frozen in the state of nature or restricted to the bilateral legal relationships within which a corrective justice wrong can occur. Rather, freedom adjusts itself to the new form of relationship—​omnilateral rather than bilateral—​that characterizes the civil condition. If it were otherwise, then the civil condition would, after all, entail a diminishment of freedom, that is, of one’s independence relative to others. Persons entering the civil condition would find that they had moved from independence to potential dependence, from a condition in which no subordination was rightful to one in which subordination was a rightfully created possibility. Systemically, the idea of independence includes more than freedom from interference with rights in one’s body and one’s property. If independence is to remain exhaustive of all juridical relationships, including the omnilateral ones of the civil condition, the system of rights requires a conception of independence that goes beyond the one present in the state of nature. Contemporary constitutional law calls this more capacious conception of independence ‘human dignity’,18 a term that refers in the most abstract way to the normative idea that governs the relationship between an individual and the commonwealth of right-​holders that constitutes the civil condition. In this enlarged conception of independence, the positive side of innate right, that one is entitled to be one’s own master (sui iuris) and therefore to assert oneself in one’s relations with others, encompasses the power to make one’s way in life in at least a minimally decent and satisfactory manner.19 The point of this is that persons who can make their own way in life do not have to rely on others, and are therefore not subject to or dependent on the will of others. In some jurisdictions, this enlarged conception of independence is treated as including a constitutional right to a ‘dignified human existence’.20 In others, it has historically been reflected in the 17 Kant (n 8) at 101 [6:316]. For an illuminating history of the idea that political society preserves rather than abridges natural rights see D Edelstein, On the Spirit of Rights (University of Chicago Press 2019). 18 A Barak, Human Dignity: The Constitutional Value and the Constitutional Right (CUP 2015); J Weinrib, Dimensions of Dignity: The Theory and Practice of Modern Constitutional Law (CUP 2016). 19 In Kant’s treatment of the family, a child has an innate right to the care of parents, who are supposed to exercise this care to develop him ‘pragmatically, so that in the future he can look after himself and make his way in life’ (Kant (n 8) 71 [6:281]). Kant’s thought seems to be that children are born with their innate right, but the parents are under a duty to see that children, when they mature, are able to realize the independence to which their innate right entitles them. 20 Asylum Seekers Case 1 BvL 10/​10 (2012) (Germany); Barak, (n 17) 288–​92 (Israel); Johannesburg v Mazibuko [2009] 3 All SA 202, para 17 (CA) (South Africa). Article 25 of the Japanese Constitution entrenches ‘the right to maintain the minimum standards of wholesome and cultured living’; see

106  Ernest J. Weinrib state’s police power, regarded as a ‘necessity growing out of the fundamental conditions of civil society’ to protect ‘the lives, the health, and the safety of the public against the injurious exercise by any citizen of his own rights’.21 Regardless of its particular legal formulation or its constitutional status, the underlying idea is that in actualizing the system of rights the civil condition protects the independence of the individual not only by securing the rights of private law, but also by mitigating the various dependencies to which the civil condition may give rise, so that each person can take charge of his or her own life in at least a minimally satisfactory way. In mitigating those dependencies, the state must address the extreme deprivations that might lead persons to subordinate themselves to the power of others. Although within the relationships of corrective justice, basic material needs lack normative standing, they nonetheless are significant as a matter of distributive justice to the forms of subordination that a system of rights is tasked with alleviating. This difference in the role of basic material needs accords with the differing organizing principles that animate the two forms of justice. Corrective justice reflects the correlativity of the normative positions of two particular parties; need has no standing in corrective justice, because the need of one particular person is not correlative to the resources of another particular person. In contrast, distributive justice involves a comparison of any number of persons in accordance with a distributive criterion; as a property comparable across persons, need can figure in such a criterion. The point of effecting distributive justice based on this criterion is not to satisfy a particular need as such, but to create the conditions under which persons are capable of independently setting and acting towards their own ends, despite being excluded from what others own. Thus, on the Kantian view the significance of basic material needs is not that their fulfilment makes a person’s life go well in its own terms, but rather that deprivation leads one person to become dependent on another. In this vein, Dieter Grimm, in his summation of the modern history of rights, has noted that the ‘decisive effect’ of mass poverty was the rise of the dependency of the poor: Having only their labour at their disposal, those lacking in means were compelled to accept the conditions of the wealthy in order to subsist in a situation in which labour was plentiful. Considered in formal terms, both parties were merely exercising their contractual freedom. Considered materially, one side was in a position to arbitrarily dictate terms, while the other could only accept these or perish . . . [P]‌recisely as a consequence of private autonomy, the system permits the accumulation of social power and in this way continually produces threats to freedom.22 Minimum Standards of Wholesome and Cultured Living Case (1948) in W Murphy and J Tanenhaus, Comparative Constitutional Law: Cases and Commentaries (St Martin’s Press 1977)283–​85.

21 22

Lochner v State of New York 198 US 45 at 65 (1905, Harlan J dissenting). D Grimm, Constitutionalism: Past, Present, and Future (OUP 2016)188–​89.

Distributive Justice  107 As is always the case for a system of rights, what matters is the relationship among persons if they are to interact on the basis of their reciprocal freedom, not the welfare or happiness of any person considered on its own. Accordingly, so far as distributive justice is concerned, the reason for attending to basic material needs is not that such needs are inconsistent with persons’ welfare, but that under the conditions of human existence reciprocal freedom cannot be realized unless they are alleviated. In the civil condition, the system of rights, through its protection of acquired rights under private law, creates the systemic possibility of dependence that must then be systemically addressed. Programmes of distributive justice are among the measures through which the state can systemically preserve the independence of individuals. The various programmes focus on the different kinds of activities (workplace injuries, automobile accidents, labour relations, and so on) that can undermine this independence.23 In the modern industrial state, these activities are so large and complex and the legal possibilities for addressing them so varied, that the exercise of distributive justice has to take the form of detailed legislation and

23 For example, the following passage describes the position of workers before the advent of workers’ compensation. It succinctly sets out the economic compulsion forcing them into dangerous work, the frequency and seriousness of employment injuries, the inadequacy of tort law to provide redress, the obstacles to litigation, and the difficulty in receiving damages even when damages were awarded: Economic necessity compelled people to work long hours under conditions more hazardous than they would freely choose and without any opportunity to provide for their personal safety. The physical conditions, equipment, and work methods of the day often increased the likelihood of accidents and decreased the likelihood of recovery in tort. As many as half the workplace accidents were found to result from hazards of the workplace, not from anyone’s fault. When an employer was at fault, he could erect the common law defences of contributory negligence and voluntary assumption of risk to insulate himself from liability. Both of those defences, though generally available, operated with particular harshness in the workplace, because despite the dangers of the workplace and the work tasks, there were no real alternatives open to the worker. In addition, employers alone could assert the defence of common employment to neutralize claims based on the fault of a worker’s fellow employees. All of these defences operated as absolute bars to recovery . . . Finally, even successful worker plaintiffs had no assurance of meaningful recovery. Wealthy employers could also exhaust the worker’s resources in a series of appeals. Legal fees and other costs reduced the amount realized from any award. Impecunious employers often could not pay the judgments against them. For slow developing industrial disease claims, workers might find that their employers had gone out of business or disappeared. Medwid v The Queen in the Right of Ontario (1987) 48 DLR (4th) 272, at 280 (HCJ), affirming the constitutionality of provisions of Ontario’s Workers’ Compensation Act that removed from a certain class of workers the right of action for tort damages against employers other than their own. This paragraph was adopted verbatim from the factum of the Attorney-​General of Ontario, which supplies references for the historical assertions it contains. I am grateful to Lorraine Weinrib, who acted as counsel for the Attorney-​General of Ontario, for bringing this case to my attention and for making her factum available to me. For an account of the labour ideal of independence in the nineteenth century see JF Witt, ‘Toward a New History of American Accident Law: Classical Tort Law and the Cooperative First-​Party Insurance Movement’ (2001) 114 Harv L Rev 690,724–​32. Witt, at 731, cites testimony at the time that previously independent and self-​supporting families were ‘obliged to turn in humiliation and permanent injury to the charitable societies or to relatives and friends’.

108  Ernest J. Weinrib regulation, executed through administrative rather than judicial procedures, and applied to persons who fall within carefully crafted standards of eligibility. One should not suppose that the state’s role in protecting independence merely transfers a person’s dependence from other individuals to the state.24 The relations of individuals with the state stand on a different normative footing than do relations among individuals. Because the obligations that are correlative to the rights of private law are negative only and impose no duties of philanthropy, a person in circumstances of exigency can be only a supplicant for the charity of others; dependence reflects the contingency that such charity might, without any violation of private law, be withheld or withdrawn at will. The state, in contrast, is under an obligation to mitigate the dependencies that systemically flow from the operation of the rights of private law. It fulfils this obligation by enacting appropriate legislation and by creating the appropriate administrative institutions. These mitigating measures cannot be withheld at will, because within a framework of rights the sole will that the state has is to act for a public purpose, and the sole relevant public purpose is that of sustaining the independence of its members. The state’s role in this is not a matter of charity but a normative requirement intrinsic to the rightful exercise of its power. Nor can the administrative organs of the state act at will to withdraw benefits that the state has enacted. These organs are bound by the rule of law and by norms of administrative fairness that reflect the human dignity of every applicant. In making a claim for these benefits an individual does not beseech the administrator’s grace but asserts an entitlement. Accordingly, the threats to independence to which private law may give rise have, as matter of right, no place in the relations between individuals and the state in the civil condition. And if in its activity and processes a state does in fact treat those who are subject to it as its dependents, it lies under an obligation, again as a matter of right, to change its practice in order to eliminate the dependencies. Independence from the constraining will of others thus serves as the unifying theme for the conceptually ordered sequence that goes from corrective justice to distributive justice. The sequence begins with the innate right, for which independence from the constraint of another’s will is the defining characteristic. Because external things are usable by persons (and thus not independent of their wills) acquired rights are consistent with the reciprocal freedom of all persons. Acquired rights in turn require a civil condition, because in a civil condition the systematic availability of acquisition prevents the unilateral imposition of the acquirer’s will on the independence of everyone else. Corrective justice is the kind of justice that is regulative of the bipolar relations that arise from this ensemble of rights and their correlative obligations. Acquired rights, however, create the possibility of forms of systemic subordination that must be systemically addressed by arrangements



24

For this supposition see H Dagan, Property: Values and Institutions (OUP 2011)65–​66.

Distributive Justice  109 of distributive justice. Each stage of this sequence presupposes the stage that preceded it; similarly, none of these stages completes a fully adequate system of rights unless followed by the subsequent ones. The positioning of distributive justice within this sequence conforms to the requirements for sequenced argument described in the first section of this chapter. It was there pointed out that a sequenced argument about distributive justice should set out why the system of rights is not limited to corrective justice, what normative idea allows corrective justice and distributive justice to participate in the same sequence, and why distributive considerations cannot be included in the earlier stage of the sequence. The account presented here answers these questions as follows: The system of rights goes beyond corrective justice, because private law, although grounded in the idea of each party’s independence from the constraint of the other’s will, creates the possibility of systemic subordination that distributive justice alone can address. Distributive justice is posterior to corrective justice, because such subordination arises from a functioning system of private law, even if corrective justice is fully observed. The distributive considerations cannot be backed up into the corrective justice stage, because those considerations are not correlatively structured and therefore cannot fairly and coherently figure within private law. Thus, the argument about distributive justice indicates why corrective and distributive justice belong together despite the categorical distinction between them, how they are each organized around their respective notions of independence, why they are to be kept separate and in sequenced order, and why the sequence cannot stop at corrective justice. Taken as a whole, the entire sequence, with its multi-​staged elaboration of the idea of reciprocal independence, presents the juridical unity of a legal order that incorporates both corrective justice and distributive justice. Despite the pluralism of these two forms of justice, they come together in the unity of a conceptually ordered sequence.25

25 The idea of independence to which this section has been devoted resembles the idea of non-​ domination that has been central to contemporary neo-​ republican thinkers; see, for example, P Pettit, Just Freedom: A Moral Compass for an Unjust World (WW Norton 2014); Q Skinner, Liberty Before Liberalism (CUP 1998). This is not surprising, given Kant’s position in the history of republicanism. See P Pettit, ‘Two Republican Traditions’ in A Niederberger and P Schink (eds), Republican Democracy: Law, Liberty and Politics (Edinburgh UP 2013) 169. However, the Kantian account I have offered differs from contemporary neo-​republicanism in theoretical orientation. Neo-​republicanism is consequentialist, with non-​domination playing the same role as the principle of utility in utilitarianism; P Pettit, Republicanism: A Theory of Freedom and Government (OUP 1997) 99–​102; F Lovett, A General Theory of Domination and Justice (OUP 2010) 159–​63. The Kantian account is not consequentialist; rather, it works out the significance of the juridical relations that constitute law as a distinctive normative phenomenon. Similarly, the neo-​republicans start from a conception of freedom in which non-​domination contrasts with non-​interference. The Kantian account starts from the very idea of coherent juridical relations, which it then elucidates through a sequenced argument that includes both non-​interference (in private law) and non-​domination (in public law).

110  Ernest J. Weinrib

3.  Determining Distributive Justice Like corrective justice, distributive justice is an abstract structural idea that is made determinate through the positive law. As befits the differences between them, however, the process of determination for each of these forms of justice follows a different institutional path. Corrective justice is the structure for the diverse grounds on which an inconsistency with a particular plaintiff ’s right can be ascribed to a particular defendant. The correlativity internal to this structure means that the point of corrective justice is not to promote some goal that is attractive independently of the relationship between the parties, but rather to hold the parties to the normative implications of justice as between them. To this end the positive law works out and applies the various correlatively structured concepts, standards, principles, and rules that make up to the grounds for holding one person liable to another. The task of specifying these concepts, standards, principles and rules is interpretive and adjudicative. It is interpretive in that it seeks to elaborate the series of ideas operating at different levels of generality that discloses how the correlativity internal to the structure of liability figures in the juridical relationship between two particular parties.26 It is adjudicative in that such interpretation is within the province of judges, who directly or in reliance on the work of expert jurists engage in the public reason that authoritatively and systematically works out the meaning of corrective justice for particular transactions and for classes of transactions.27 In comparison with the significance of adjudication, the role played by legislation in corrective justice is relatively minor, either supplementing the stock of private law determinations by, for instance, establishing statutory duties that are relevant to one’s legal obligations to another, or codifying those determinations by setting them out summarily and definitively. In contrast to the unmediated interaction of corrective justice, distributive justice mediates the relation between the participants in a distribution through a criterion of distribution. This criterion indicates the purpose that the distribution seeks to achieve. Distributions can be directed towards different purposes and can vary in scope. For example, a distributive programme that is concerned with alleviating adverse bodily conditions may deal with medical problems generally as in the case of publicly organized medical insurance, or more narrowly with accidentally suffered injuries as in the case of the New Zealand accident compensation programme, or even more narrowly with injuries that result from a particular kind of activity as in the case of workers’ compensation or compensation for injuries suffered by the victims of crime. Thus, whereas corrective justice involves not the selection of a purpose but the explication of the internal normative dynamics of

26 27

See ch 1, s 6. See ch 4, s 2.

Distributive Justice  111 parties’ correlative positions, distributive justice deals with particular distributions that have their particular purposes. This contrast reflects the categorical difference between corrective and distributive justice. Although these forms of justice are abstractions from the particular arrangements that fall under them, they abstract in different ways. The instantiations of corrective justice are the constituents of a totality organized around the idea that the law authorizes the appropriate correction when an inconsistency with one person’s rights can be imputed to another. A single and unified conception of corrective justice exhibits the normative structure immanent to and homogeneously present in the interactions of persons acting for self-​chosen ends. The various grounds of liability are systematically demarcated with reference to one another through the normative considerations intrinsic to each of them. Grounds of liability are thus the articulations of corrective justice as the single unified conception at play in the various circumstances in which an inconsistency with another’s right can occur. In contrast, the instantiations of distributive justice evince the heterogeneity of the different specific purposes and the varying scopes of different distributions. Even if distributions share the general goal of mitigating dependencies (as they would on the Kantian picture being presented here), they pursue that goal through discreet programmes that have no intrinsic mutual connection or alignment. Accordingly, distributions do not articulate a single unified conception, but rather form an aggregate of arrangements that have a certain structure and that share the same general end. Whereas corrective justice is the abstraction representing the totality of private law relationships as an articulated whole, distributive justice is the abstraction representing the multiplicity of possible distributions, each of which has its specific purpose and particular scope. Corresponding to this contrast in their character as abstractions is another that pertains to their respective modes of determination. In the case of corrective justice, determination involves specifying the meaning of corrective justice for a given legal relationship in which the parties are viewed as the correlatively situated with respect to a supposed injustice as between them. To specify in this context is authoritatively to work out at various levels of generality and particularity the most adequate and plausible understanding of what is normatively implicit in the correlative situation of plaintiff and defendant.28 This process of determination is a proper task for the judiciary. In the case of distributive justice, in contrast, determination consists not in specifying meaning in particular circumstances but in choosing a particular distribution from all the possible distributions. Because distributive justice, conceived in terms of addressing conditions of subordination, bears systemically on the state



28

See ch 1, s 2.

112  Ernest J. Weinrib as a whole, the state is required to actualize it. This requirement can, however, be fulfilled in different ways or combinations of ways. One aspect of choosing a particular distribution is to settle upon a particular purpose that is to be collectively recognized. Concomitant to this is the definition of the scope of the distribution and the qualifications for participating in it. Also to be decided (inevitably against a background of scarce resources) are the mechanisms for financing and administering the distributive programme. Decisions on these matters do not specify what is implicit in a certain kind of relationship but rather posit a set of legal arrangements to satisfy the requirement on the state to anticipate or mitigate the circumstances of subordination. In determining distributive justice, the role of the legislature is paramount. Unlike the judiciary, which operates within the confines of the case before it, the evidence relative to that case, and the contingencies of decisions by litigants, the legislature acting under the principle of the separation of state powers has the institutional competence to assess the full range of possible distributions, to articulate the details of the preferred distribution, to provide for its funding, and to set up the regulatory structure that administers it. The legislature also has the authority to commit the polity as a whole to a particular set of distributive arrangements; legislators can then be held accountable through the democratic process for the distributive decisions that they make on behalf of everyone. In determining the distributions that support the independence of all, the legislature has broad latitude. Both in their scope and in the details of their construction, different programmes of distribution can aim at the amelioration or prevention of dependence in different ways. A distributive arrangement can deal with dependence over a broader or narrower range by addressing different activities or conditions. The level of benefits that it dispenses can be higher or lower. The eligibility criteria can be more or less restrictive. Its effect on private law can be more or less pre-​emptive. These determinations are matters of politics for which legislatures are accountable, rather than of legal theory. By mentioning politics I do not mean the free play of individual interests in exercising power or in influencing or controlling collective decisions. Rather, within the Kantian approach, the notion of politics refers to the enterprise of applying concepts of right to human experience; it includes managing the public apparatus of the civil condition in accordance with principles of right.29 Determining distributive justice is political, in that the task calls for a good faith judgment about the measures that, given the contingent context of a particular legal system, would best sustain the independence of all by most felicitously alleviating the systemic dependencies that the legal system would otherwise enable. Citizens may, of course differ about this judgment, but it is one that, at the



29

Kant (n 7).

Distributive Justice  113 end of the day, the legislature authoritatively exercises within its constitutional competence. In making such determinations, the role of legal theory is comparatively modest: to identify the normative space within which such determinations occur and to relate that space to other spaces within the normative ecology of a legal order. In starting from corrective justice, a Kantian theory of rights maps out these normative spaces by working out the stages of the conceptually ordered sequence through which they are related and by elaborating the thematic role of independence in that sequence. In view of this legislative latitude, various mechanisms are available for reducing the prospect of one person’s dependence on another. Without claiming to be exhaustive, I list four of these. One mechanism is for the state to authorize arrangements that foster the development of persons’ independence or that prevent vulnerabilities that would expose them to domination or exploitation by others. An example of the former is the creation of systems of education that give each person the opportunity to develop as a self-​determining moral being and as a self-​ supporting economic actor. Examples of the latter are the recognition of collective bargaining in the employment context and the statutory regulation of residential tenancies, so as to mitigate the power that employers and landlords would have if matters were settled solely on the basis of private ordering. A second mechanism is for the state to provide each person with a social minimum. In the civil condition, this minimum is geared not to subsistence as a biological matter but to the human dignity that consists in the possibility of taking charge of one’s life as an independent person. Like all normative ideas within a system of rights, the dignified human existence at which the social minimum aims is a relational, not a monadic, concept. It refers to persons as interacting beings in the context of the particular society in which they live. It includes the possibility of maintaining relationships with others and of participating in at least a minimal way in the social life that marks human existence.30 What comports with the social minimum for independence varies with a society’s resources and social conditions. Thus, in working within its legitimate and unavoidable latitude, the legislature’s task is to orient the social minimum ‘towards the respective stage of development of the polity and towards the existing conditions of life’.31 A third mechanism is for the state to mandate or to operate social insurance arrangements that deal with the impact of injuries that would otherwise endanger the independence of injured persons. A common feature of such arrangements is that, for injury that prevents one from working they include payments that are based on a percentage of one’s prior earnings rather than being directly related to one’s

30 Social Minimum Case (Hartz IV) 1 BvL 1/​09 para 134 (German Constitutional Court 2010). 31 ibid para 133. See also Barak (n 17) 289: ‘The minimum needed to live with dignity . . . cannot be determined by a medical measurement. It is the product of society’s view regarding the minimum a person needs in order to express her personality in the framework of the society in which she lives.’

114  Ernest J. Weinrib need. On the one hand, such payments are more generous than what would be due under the social minimum. On the other hand, they provide a measure of continuity between a person’s pre-​injury and post-​injury condition, thereby treating independence not merely as a desideratum tied episodically to the occurrence of the contingencies that threaten it, but as a normative idea that it applies over a whole life and reflects the unity of that life. A fourth mechanism involves responding to injury by blending the social minimum or the social insurance with a limited range of tort damages. Many automobile injury compensation programmes and proposals take this form. They carve out a set of injuries for which an accident victim receives benefits on a no-​fault basis while retaining tort law (or elements of it) either as a possible alternative to the no-​fault programme or as a separate regime applicable to injuries that are above a threshold of severity. A pervasive goal of these compensation programmes—​ which differ widely in their details as to the ceiling for no-​fault benefits, the floor for tort recovery and the connection between the no-​fault compensation and tort damages—​is to demarcate the circumstances under which some measure of income replacement, as well as medical and rehabilitation benefits, will promptly be available to a significant proportion of automobile accident victims without their having to prove fault through the protracted and laborious process of tort litigation. The compensation programme thus forms part of the network of arrangements that aim to maintain the victim of injury as an independent person. The possibility that social insurance might replace, or at least partially displace, areas of private law has led some to question the viability of tort law’s treatment of personal injuries in comparison with the compensation systems that are its distributive alternatives.32 When considered as a mechanism for compensating personal injuries, tort law has obvious shortcomings: injuries not caused by someone’s fault are not compensated; the individualized determination of fault and damage assessment involves delay, uncertainty, and expense; and even if fault is found or conceded, financially irresponsible or uninsured injurers may be unable to pay the damages that are due to the injured party. The reason for these shortcomings is that, properly understood, tort law is not a mechanism of compensation at all, but rather a system of redress for one person’s wrongful infringement of another’s rights. Individualized determinations of fault are integral to such a system of redress, in which compensation follows only from, and to the extent of, the wrongful infringement by a particular defendant of the right of a particular plaintiff. Because the provision of compensation to all injured persons is not a goal of tort law, the restricted compensation that tort law provides is not a failure of tort law. Tort law and compensation programmes have different functions and therefore do different 32 Among the classic articles on this theme are MA Franklin, ‘Replacing the Negligence Lottery: Compensation and Selective Reimbursement’ (1967) Va L Rev 774; SD Sugarman, ‘Doing Away with Tort Law’ (1985) 73 Calif L Rev 555.

Distributive Justice  115 things. So formulated, the difference between compensation systems and tort law merely rehearses the pluralism of justice. When the two forms of justice are seen as sequentially ordered, it makes little sense to ask as a general question whether tort law should be replaced. The inquiry has to be more fine-​grained, with due attention to the normative role both of tort law and its alternatives and to the particular circumstances in which the question arises. On the one hand, from the standpoint of corrective justice, tort law is a normative phenomenon that strives to actualize its own distinctive virtues: fairness as between the litigants, respect for the rights of others, responsibility for one’s actions, the operation of law within a culture of justification, and the possibility of a coherent public reason. Consequently, tort law can be criticized to the extent that it betrays those virtues, for example, by legal reasoning that is incoherent or prejudicial to one or the other of the parties and by institutional procedures that make standing on one’s rights difficult and expensive, and that thus provide opportunities for abusive conduct through the process of litigation. Corrective justice, therefore, does more than illuminate tort law as a practice merely because that practice exists.33 It also construes tort law as the vehicle of a distinctive set of moral ideas—​ and as intrinsically valuable for that reason. On the other hand, the rights that tort law vindicates and the fault-​based manner in which it vindicates them may be inadequate to preserve the ability of persons to make their own way in life without recourse to the charity of others and without exposure to exploitation by them. This inadequacy is not a defect that undermines the inherent normative character of tort law as such, but rather a consequence that arises in specific circumstances and in connection with specific kinds of activities. In the face of this inadequacy, the appropriate legislative response is to institute a programme of distributive justice to deal with it in its particular circumstances. This may well involve the pro tanto displacement of tort law. Thus, the overall picture that emerges is this: private law (and tort law more specifically) constitutes a system of rights between individuals that has an intrinsic and abiding moral valence. The legislature, however, has the normative authority to set up distributive arrangements that limit or displace private law in particular circumstances. What justifies this authority is the systemic imperative in the civil condition to sustain the independence of all persons in their relationships with one another. The result is a sequential movement of thought in which the private law rights remain the underlying norm, yet in particular contexts they cede to the state’s inherent authority to deal with dangers to this independence.34 33 As asserted in the elegant essay by John Goldberg, ‘Unloved: Tort in the Modern Legal Academy’ (2002) 55 Vand L Rev 1501, 1515–​17. 34 This sequential ordering resembles the approach to the police power (although without its constitutional significance) present in Justice Harlan’s dissenting judgment in Lochner v People of the State of New York 198 US 45 (1905); see Lorraine Weinrib, ‘The Postwar Paradigm and American Exceptionalism’ in S Choudhry (ed), The Migration of Constitutional Ideas (CUP 2006) 84, 103–​10.

116  Ernest J. Weinrib

4.  Conclusion This chapter has focused on several interconnected ideas. The point of departure was the pluralism of justice—​the notion that corrective justice and distributive justice are categorically different structural ideas that cannot directly be integrated into a single overarching structure. From this pluralism arises the problem of the juridical unity of the state as a whole that is expressive of reciprocal freedom while comprehending both forms of justice. The chapter proposed that the distinctive kind of unity responsive to this problem is the unity of a conceptual sequence. This kind of unity features a series of thematically linked stages each of which presupposes and complements the one that precedes it. The contention of this chapter has been that corrective justice and distributive justice participate in such a conceptual sequence. For this sequence, the idea of the reciprocal independence of all persons is thematic. In the correlatively structured relationships of corrective justice, independence refers to freedom from interference with one’s innate right in bodily integrity and with one’s acquired rights, such as rights in property and contractual performance. When, however, the state creates the possibility of accumulation by making the exclusivity of ownership legally effective, it also creates threats to the independence of those whose action is confined to what is unowned by others. In the civil condition, subjection to another’s will is not identical to suffering of an injustice as a matter of corrective justice; and being able to make one’s way as an independent person is not assured by private law alone. Through arrangements of distributive justice the state systemically addresses this consequence of the system of rights. The root idea of a system of rights, that action should be consistent with the reciprocal freedom of everyone, thereby finds its place both in a state’s private law and in its public law.

6

Horizontality: Presuppositions and Functions 1.  The Phenomenon of Horizontality In its path-​breaking Lüth decision in 1958, the German Federal Constitutional Court dealt with the effect on a civil claim of the constitution’s guarantee of freedom of speech. The defendant had called for a boycott of a new film produced by a person with a history of notoriously anti-​Semitic film-​making during the Nazi period. The plaintiff contended that this call for a boycott violated the civil code provision against harming another person contrary to public morals. In holding for the defendant, the Court expounded the relationship between constitutional rights and private law. While affirming that the chief purpose of constitutional rights is to protect individuals and their dignity from encroachment by state power, the Court nonetheless invoked what it called the ‘radiating effect’ of constitutional rights on private law: [T]‌he Constitution does not aspire to be a value-​neutral order. In its section on basic rights it has also erected an objective order of value, thereby giving expression to a principled reinforcing of the effective validity of the basic rights. This system of value, centering on the dignity and on the free development of the human personality within social community, must hold as a constitutional axiom for all spheres of law; legislation, administration, and adjudication all receive guidance and impulse from it. Accordingly, it obviously influences private law as well; no rule of private law may contradict it, and every such rule must be explicated in its spirit.1

This decision and the sentiments that it articulates symbolized and stimulated the most far-​reaching international development in private law since the end of the Second World War.2 Originating in a sophisticated debate in Germany about the reach of its postwar constitution3 and elaborated through the jurisprudence 1 BVerfGE 7, 198 (Lüth 1958) B, II, 1. 2 For a magisterial survey see A Barak ‘Constitutional Human Rights and Private Law’ in D Friedmann and D Barak-​Erez (eds), Human Rights in Private Law (Hart Publishing 2001) 13. 3 J van der Walt, The Horizontal Effect Revolution and the Question of Sovereignty (de Gruyter 2014) 201–​32.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0006

118  Ernest J. Weinrib of the German Federal Constitutional Court,4 the idea that fundamental rights play a role in private law is now such a commonplace among liberal democracies worldwide that it ranks as a distinctive feature of the prevalent model of global constitutionalism.5 Different jurisdictions bring fundamental rights into private law through different legal modalities: through explicit constitutional provision, or through statutory directive, or through the jurisprudence of a constitutional court, or through the jurisprudence of a supreme court of general jurisdiction.6 ‘Horizontality’ is the generic term for this phenomenon in its various forms: constitutional rights (or the fundamental rights enumerated in supranational instruments) are made to apply not only to the vertical relations between the individual and the state but also to the horizontal relations between one person and another. The phenomenon of horizontality significantly extends the traditional and still primary function of constitutional rights, which is to protect persons against the state rather than against one another. Constitutional rights emerge out of the superior legal position that the state has in relation to those it governs. Through its legal organs the state has the authority to command and its subjects have the corresponding obligation to obey. The question then arises: how is this state superiority compatible with everyone’s freedom, given the possibility—​amply confirmed by history—​of arbitrary or even tyrannical exercises of state power? The guarantee of constitutional rights provides the answer. The state is required to exercise its power in a way that respects and protects the dignity of the individual as a bearer of rights; constitutional rights, in turn, are conceived as the legal expressions of each person’s freedom as against the state. Accordingly, when constrained by constitutional rights, the state’s superiority operates within a legal framework that is attentive to the freedom of all. A regime of constitutional rights is thus the response of public law to the problem of authority that public law generates. In view of the contrast between the relationships of public law and those of private law, however, horizontality requires a different account. Unlike the 4 The literature (even in English) is extensive. See eg G Brüggemeier, AC Ciacchi, and G Comandé, Fundamental Rights and Private Law in the European Union, vol 1 (CUP 2010) 253–​324; H-​J Cremer, Human Rights and the Protection of Privacy in Tort Law: A Comparison between English and German Law (Routledge-​Cavendish 2011)153–​225; D Looschelders and M Makowsky, ‘The Impact of Human Rights and Basic Rights in German Private Law’ in V Trstenjak and P Weingerl (eds), The Influence of Human Rights and Basic Rights in Private Law (Springer 2016) 295; BS Markesinis, Always on the Same Path: Essays on Foreign Law and Comparative Methodology, vol 2 (Hart Publishing 2001), 175–​218; J Mathews, Extending Rights’ Reach: Constitutions, Private Law, and Judicial Power (OUP 2018)47–​ 90; S Oerter, ‘Fundamental Rights and their Impact on Private Law: Doctrine and Practice under the German Constitution’ (1994) 12 Tel Aviv U Stud L 7; P Quint, ‘Free Speech and Private Law in German Constitutional Theory’ (1989) 48 Maryland L Rev 247; C Starck, ‘Human Rights and Private Law in German Constitutional Development and in the Jurisdiction of the Federal Constitutional Court’ in D Friedmann and D Barak-​Erez (eds), Human Rights in Private Law (Hart Publishing 2001) 97; G Taylor, ‘The Horizontal Effect of Human Rights Provisions, the German Model and Its Applicability to Common-​Law Jurisdictions’ (2002) 13 KCLJ 187. 5 Kai Möller, The Global Model of Constitutional Rights (OUP 2012)10–​13. 6 These different mechanisms are present in South Africa, the United Kingdom, Germany, and Canada, respectively.

Horizontality: Presuppositions and Functions  119 relationship between the state and those governed by it as a matter of public law, relationships of private law (at least as understood within the corrective justice framework) are characterized not by the juridical superiority of one party to the other but by their transactional equality within the correlative structure of reasoning about liability.7 Understood in this way, private law seems to have no place for the role that constitutional rights play in public law. Yet the currency and broad diffusion of the phenomenon of horizontality among liberal democracies bespeaks a contemporary consensus that constitutional rights are not sealed off from private law. Hence the proliferation of an enormous literature. Hence also the sentiment expressed by one of the most astute and important contributors to that literature, that horizontality poses a ‘fundamental and enduring problem that can scarcely be definitively resolved’.8 Over the years, jurists interpreting the positive law have suggested several different underpinnings for the relevance to private law of constitutional rights. Speaking broadly and ignoring sophisticated admixtures and variations in detail, one can distinguish three approaches: constitutional rights can be regarded as applying either directly to private law relationships just as they do to relationships between the state and the individuals it governs,9 or indirectly through the values or principles represented by the rights rather than through the rights as such,10 or institutionally through the duty on courts as public authorities to act consistently with constitutional rights and to protect them in all contexts, including those of private law.11 These approaches originated in German jurisprudence, in a legal system in which private law derives its authority from a sophisticated, respected, and apparently self-​contained codification, and in which constitutional issues and private law claims are adjudicated by different courts. The entrenched status of the civil code and the distinctive jurisdictional separation of constitutional law from private law readily give rise to the question of whether (and on what basis) the Constitutional Court can intervene in private law disputes at all. The answer provided by the three German approaches, for all their diversity of detail, is that the constitutional court, as the guardian of the constitution, can legitimately intervene when the regular civil courts misapply or fail to apply the appropriate constitutional considerations. Depending on the approach adopted, this may mean the civil courts have either not taken proper account of an applicable constitutional right, or have missed or misconstrued a constitutional value, or have failed to comply with the duty to protect constitutional rights that is incumbent on every public

7 See ch 1, s 3. 8 C-​W Canaris, ‘Grundrechte und Privatrecht’ (1994) 184 ACP 201, 202. 9 HC Nipperdey, ‘Grundrechte und Privatrecht’’ in Festschrift für Erich Molitor zum 75. Geburtstag (Berlin 1962) 17 10 G Dürig, ‘Grundrechte und Zivilrechtsprechung’ in T Maunz (ed), Von Bonner Grundgesetz zur gesamtdeutschen Verfassung (Munich 1956) 117. 11 Canaris (n 8).

120  Ernest J. Weinrib authority. These approaches from the German jurisprudence have influenced or been replicated in legal systems elsewhere, providing general categories for understanding how horizontality works. Although reference to these various approaches is useful and even essential for classifying or interpreting the different ways in which horizontality operates as a matter of the positive law of particular jurisdictions, I do not in this chapter pay much attention to the distinctions between them, let alone attempt to determine which of them is soundest. This is for several reasons. First, there is a growing recognition that the distinction between the direct effect of constitutional norms and other forms of horizontality makes very little practical difference.12 Similarly, the insistence that the courts have an obligation as public authorities to protect constitutional rights in all contexts can hardly rank as a free-​standing basis for horizontality, for it presupposes that those rights are antecedently significant—​either directly or indirectly—​for the relationship between the parties to a private law controversy; otherwise, the courts would not be exercising an adjudicative function.13 Moreover, the fact that constitutional norms can impinge on private law in different ways either within a single legal system or across different legal systems renders vain the effort to reduce them to a single form. Finally and most importantly, the operation of direct effect, indirect effect and judicial obligation present alternative mechanisms through which constitutional rights exert their influence on controversies between private persons. Important as these are, they are merely particular manifestations of the general phenomenon of horizontality itself. This phenomenon raises issues that go beyond the mechanism of its application, to its presuppositions, its function, its scope, and its operation. Those issues concern, respectively, what the connection is between constitutional law and private law that makes horizontality possible, what horizontality is supposed to do (rather than merely how it is triggered), what the boundaries are within which constitutional rights are applicable to private law, and how horizontality works for a controversy within those boundaries. 12 R Alexy, A Theory of Constitutional Rights (J Rivers tr, OUP 2002)355–​58; H Collins, ‘The Challenges Presented by Fundamental Rights to Private Law’’ in K Barker, K Fairweather, and R Grantham (eds), Private Law in the 21st Century (Hart Publishing 2017) 213; M Kumm, ‘Who is Afraid of the Total Constitution? Constitutional Rights as Principles and the Constitutionalization of Private Law’ (2006) 7 German LJ 342; G Brüggemeier and AC Ciacchi ‘Introduction’’ in G Brüggemeir, AC Ciacchi, and G Comandé (eds), Fundamental Rights and Private Law in the European Union, vol 2 (CUP 2010) 8: ‘a diffuse consensus exists that the distinction between direct and indirect Drittwirkung is not appropriate, nor is this term itself helpful to clarify the relationship between fundamental rights and private law’. 13 Compare Brüggemeier’s criticism of Canaris in G Brüggemeier, ‘Constitutionalisation of Private Law: The German Perspective’ in T Barkhuysen and S Lindenbergh (eds), Constitutionalisation of Private Law (Martinus Nijhoff Publishers 2006) 59, 76: ‘[T]‌he entire hyperconstruction of Canaris rests on a misunderstanding, that is, cause and effect are confounded . . . [F]undamental rights do not claim relevance for private law relations just because a state court makes a decision . . .The civil courts clarify the reciprocal rights and duties of the contending private parties . . . [T]he decisive constitutional wrong remains the act of the invading private party.’

Horizontality: Presuppositions and Functions  121 This chapter and the following one address these more general issues. My focus is not on the working of horizontality within any particular regime of positive law (although German law is treated in many respects as paradigmatic), but on thematic aspects, engaged by the very phenomenon of horizontality, of the relationship between private law and constitutional rights. Central to this treatment of horizontality is the idea, introduced in the preceding chapter and further elaborated in this one, that within a functioning legal system human dignity is the equivalent in contemporary jurisprudence of what Kant termed the person’s innate right. For Kant the innate right is the subjective right corresponding to the universal principle that, as a matter of right, each person’s action should coexist with everyone else’s freedom; unless by virtue of their innate right persons were entitled to act independently of the constraint of others, such freedom would be inconceivable. The innate right thus reflects the reciprocal freedom present in the interaction of mutually independent beings who count as ends for one another and not merely as means. As the subjective right inherent in humanity’s capacity for self-​determining action, the innate right is normatively presupposed in every other kind of right. It thus undergirds both private law and public law. Although the innate right is fully and universally valid in the state of nature, the legal terms in which the innate right is rendered concrete are formulated in the civil condition. The civil condition, in turn, is the domain of the public and systematic legal norms through which persons interact as members of a community who all have equal standing to have their innate right (as well as their other rights) respected. In a constitutional system of rights, horizontality renders the fundamental rights enshrined in the constitution relevant to the legal concretization of persons’ innate right in their relations with one another. The idea of human dignity transposes Kantian innate right into contemporary legal discourse by signifying the non-​instrumental value that persons have innately and by virtue of their humanity. Through the phenomenon of horizontality reciprocal freedom expressly includes reference to the human dignity of the persons interacting in the civil condition. So understood, human dignity reflects the person’s being situated in the community that comprises the civil condition. Human dignity is an attribute of the person viewed not as an atomistic individual, but as a member of the community formed by the legal system. As the law’s fundamental normative idea, human dignity requires the law to be so ordered that the actions of others, and the influence that such actions might have on the opinions of others, be consistent with each person’s standing as a being of intrinsic worth. Aspects of the idea of human dignity are the legal equality of all persons, each person’s entitlement to self-​determination in relation to others (including being uncoerced in thought, speech, and conscience), each person’s independence from domination by others, and each person’s responsibility for one’s actions as well the corresponding immunity regarding consequences that one has not caused. In addition, when these and other aspects of

122  Ernest J. Weinrib human dignity are understood as operating within the community formed by the civil condition, human dignity is contextualized within a particular society with its particular history and with the particular material and social conditions at work in it. Human dignity can, accordingly, be regarded as the juridical idea that addresses what John Rawls called ‘the social bases of self-​respect’.14 Consequently, the law has the task of working out the legal concepts and doctrines through which the human dignity of each person can co-​exist with the human dignity of all persons within their particular community of right-​holders. This conception of the social character of human dignity is explicit in the German constitutional jurisprudence that has been crucial to the elaboration of horizontality. One sees it in the extract from the ground-​breaking decision on horizontality quoted at the beginning of this chapter. There the Constitutional Court refers to the ‘system of value, centering on the dignity and on the free development of the human personality within social community’.15 More extensively, one sees it in the following strikingly Kantian paragraph, which sums up the understanding of human dignity in German constitutional thought: This [the obligation incumbent on state authority in all its manifestations to respect and to protect human dignity] is based on the conception of man as a spiritual-​ethical being endowed with self-​determination and self-​development. The Basic Law understands this freedom not as that of an isolated and sovereign individual, but rather as that of an individual who is based in and bound to the community . . . In view of this, this freedom cannot be ‘unlimited in principle.’ The individual must put up with those limitations of his freedom of action that the legislator deems generally reasonable in the administration of a given matter for the maintenance and for the requirements of social life together; nonetheless, the autonomy of the individual must remain protected . . . This means that even within the community each individual must in principle be recognized as a juridically equal member with a value of his own. It contradicts human dignity to make the person a mere object in the state. The sentence, ‘Man must always remain an end in himself ’, has unlimited validity in all areas of the law; for the dignity of man as a person, which can never be lost, consists particularly in this, that he remains recognized as a personality responsible for himself.16

In Chapter 4 of this book I drew out some of the implications for private law of the transition from the conceivability of various kinds of rights in the Kantian state of nature to their actualization in the civil condition. In this transition rights gain a 14 J Rawls, Justice as Fairness: A Restatement (Harvard UP 2001)60. 15 Lüth (n 1). This phraseology is standard in the German cases. See eg BVerfGE 4, 7 (1954); BVerfGE 6, 32 (1957); BVerfGE 30, 1 (1970); BVerfGE 30, 173 (1971); BVerfGE 34, 269 (1973); BVerfGE 65, 1 (1983). 16 BVerfGE 45, 187 (Life Imprisonment, 1977) C, II, 1.

Horizontality: Presuppositions and Functions  123 public character that reflects the omnilaterality of each person’s relation to every other person through the state’s legal institutions. The publicness and systematicity of law in the civil condition can affect even the content of those rights as they are brought into line with this omnilaterality. To repeat what was observed there, the public character of private law refers not to instrumental or utilitarian considerations that aggregate or trade off persons’ interests, but solely to the actualization through private law of the equal freedom of all within the omnilateral community of interacting right-​holders. The phenomenon of horizontality is a conspicuous example of the public character of private law in the civil condition. Horizontality brings constitutional law’s notion of human dignity and of the particular rights that specify it to bear on legal relationships between private parties. The most fundamental ideas governing the exercise of public authority are thereby brought into the consideration of private law controversies. That this does not leave the posited private law unaltered is undeniable. What, however, at the deeper level of legal ideas is the precise significance of the changes that horizontality brings about? With this question in mind, I enquire in the following section into the normative presuppositions of having private law refer to constitutional norms by considering how constitutional rights can possibly be relevant to private law at all. In the subsequent three sections I set out the three different functions that horizontality might serve. These functions are geared not to the standard distinction between direct and indirect effect, but to different conceptions of the judicial obligation to apply constitutional norms to private law. I shall suggest that the most robust of these functions—​what I call the ‘dignity function’—​is the most coherent of the three, and also the most extensive in its application. After describing this function in the final section of this chapter, I consider its scope and operation in the next one.

2.  The Presuppositions of Horizontality In treating constitutional rights as relevant to private law, the phenomenon of horizontality operates on the basis of presuppositions about rights and about the nature of the legal order. The presupposition about rights is that, despite the different contexts in which they function, the rights of constitutional law and the rights of private law have a normative commonality that allows the former to influence or even determine the latter. The presupposition about the legal order is that, despite the differentiation between public law and private law, the entire legal order forms a juridical unity, the parts of which are systemically related to one another. These two presuppositions of horizontality are intertwined through the role that rights play in the legal order: the normative commonality shared by constitutional rights and private law rights animates the legal system as a unified whole.

124  Ernest J. Weinrib These presuppositions about commonality and unity have long informed the relationship between fundamental rights and private law. From the outset of modern constitutionalism, the idea that the constitution and private law were the components of a unity was a prevailing view, eclipsed only subsequently. In a recently published book, Dieter Grimm has drawn attention to what he calls ‘the original unity of the constitution and private law’17 in the French revolutionary debates about the project of codifying private law. The freedom and equality proclaimed in the Declaration of the Rights of Man and the Citizen were regarded not merely as limitations on governmental action but also as guiding principles for the post-​feudal reformulation of private law.18 This is evident from the legislation mandating a review and reform of the civil law in order to produce a general code of laws that would be ‘simple, clear, and appropriate to the constitution’.19 As a result of the ensuing demolition of the traditional institutions of rank and hierarchy, what emerged as the decisive factor in the reworking of property and contract law was the supremacy of the individual will. Grimm describes the significance of this in these terms: Whereas previously legal restrictions were supreme over the will, now in contrast the will became the source of legal restrictions. Under this premise private law was transformed from a system of objective rules of conduct into an ensemble of subjective rights. It was at its core dispositive. It needed simply to establish the conditions of a valid expression of the will and the consequences of authorizations of the will, and also to draw up for typical social relations the patterns of rules that arose when the individual will expressed itself incompletely or imprecisely.20

Thus, consequent upon the Declaration’s ringing announcement that ‘men are born free and equal in rights’, the primacy of individual will was introduced as the constitutionally appropriate foundation for private law relationships. The constitution was regarded as the basis of the entire legal system unified through the organizing idea of individual freedom. Similarly (and in substantially parallel terms), one can discern the ideas of commonality and unity in Hans Carl Nipperdey’s path-​breaking exposition of horizontality.21 Nipperdey’s starting point was the centrality of human dignity within Germany’s Basic Law. Human dignity, he wrote, lies at the very core of the idea 17 See D Grimm, Verfassung und Privatrecht in 19. Jahrhundert (Mohr Siebeck 2017)39. 18 ibid 4. In Michel Troper’s succinct formulation, ‘the Constitution recognizes rights that are indistinctly vertical and horizontal’; M Troper, ‘Who Needs a Third Party Effect Doctrine? The Case of France’’ in A Sajó and R Uitz (eds), The Constitution in Private Relations: Expanding Constitutionalism (Eleven International Publishing 2005) 115, 123. 19 Lois des 16 et 24 Août 1790 sur l’Organisation Judiciaire, II, 19, available at mafr.fr/​en/​article. Grimm (n 17) 4. 20 Grimm (n 17) 61. 21 Nipperdey (n 9).

Horizontality: Presuppositions and Functions  125 of law, ranking as its highest value and imbuing it with moral force. Human dignity is realized through law when the legal system ‘secures a sphere for the person in which he may act as an independent being who is morally responsible to himself, in which he is neither subject to the domination of another nor rendered the mere means to collective goals, but is a free and individually responsible person’.22 Expounding the idea that freedom so construed is the highest good that worldly justice can bestow on human beings, Nipperdey referred to Savigny’s statement that ‘Right serves morality not by performing its bidding, but by securing the free development of the power indwelling in each individual will’.23 Nipperdey then suggested that the German constitution accords with Savigny, in that it follows up its affirmation that human dignity is inviolable with the declaration that every person has the right to the free development of his personality. Thus, Nipperdey presents human dignity and the right to free development of the personality as foundational not only for private law (Savigny’s original concern)24 but for the legal order as a whole. In his understanding, the role of constitutional law and its basic rights is to regulate the entirety of the legal order, including its subjective private rights, as a unity.25 In so doing, law functions as a cohesive system, premised on the wholeness and indivisibility of freedom itself.26 Accordingly, the constitution and the rights of private law are bound together through the commonality of their concern for human dignity within a legal order conceived as a normative unity that aims at freedom’s realization. The Kantian resonances of Nipperdey’s presentation are unmistakable. First, dignity is characterized in terms of an independence that is incompatible with the individual’s being dominated by others or being reduced to a means for collective ends. Moreover, Nipperdey adopts Savigny’s largely Kantian depiction of Right as a normative domain separate from ethics that has the development of individual freedom as its distinctive purpose.27 How, then, from the Kantian perspective on which Nipperdey’s account is premised, are the ideas of commonality and unity—​ and thus the connection between private law and constitutional rights—​to be understood?28 22 ibid 18. 23 ibid. See FC von Savigny, System of the Modern Roman Law (W Holloway tr, 1867) 270. 24 Savigny (n 23) 269. 25 Nipperdey (n 9) 24. 26 ibid 33. 27 On the Kantian aspects of Savigny’s thought see S Peari, The Foundation of Choice of Law: Choice and Equality (OUP 2018)70–​79. 28 Nipperdey’s ideas about human dignity and the unity of rights buttressed his conclusion that basic rights are directly binding on private law transactions. This conclusion assumed an idea of unity in which the same normative factors are operative across the entire range of legal phenomena within a homogenous legal order. In contrast, the direct application of constitutional rights is not compelled within a Kantian account, where the unity in question is a unity of conceptual sequence; see ch 5, s 1. As that chapter observed, for a system of rights juridical unity consists not in the immediate and uniform application of one justification to every kind of legal arrangement, but in a sequence of presupposed and complementary concepts through which different kinds of legal arrangement are related through the

126  Ernest J. Weinrib As the contemporary designation for what Kant termed the innate right, human dignity plays a role in the operation of the legal order that is identical to that of the innate right in function and in content. They are identical in function because each of them—​human dignity as a doctrinal notion, and the innate right as a theoretical one—​signifies the normative value that most fundamentally underlies the legal system as a whole. For instance, the German constitution from the outset proclaims human dignity to be the inviolate object of respect and protection in all exercises of official power whether by the legislature, by the executive, or by the judiciary. Similarly, Kant’s innate right pervades the entire legal system, including both private law and public law. This is because the innate right is the inalienable subjective right that corresponds to the postulate at the root of all manifestations of rightful legality, that the action of any person be consistent with the freedom of everyone else. Within a system of right, persons carry their innate right with them into every legal transaction and every legal relationship; being innate, this right can be neither lost in one’s interactions with another, nor disregarded by the state as it acts in the name of all. Human dignity and the innate right are also identical in content, in that the innate right evinces the very features that Nipperdey ascribes to human dignity: individual independence and responsibility, freedom from domination by others, and the ineligibility of persons for use as mere means to collective goals. For Kant, the innate right is a reflection of the right of humanity in one’s own person, that is, of the moral power of acting for one’s own purposes, which is generic to human beings as self-​determining agents.29 That right asserts every person’s worth as a human being in relation to others, and therefore rules out a person’s being treated merely as a means.30 Both individual independence and individual responsibility are aspects of the innate right. The individual independence consists of the freedom from the constraint and domination of another’s will; the individual responsibility refers to the possibility of being held accountable only for what one has done and not for who one is. Being endowed with the innate right signifies that as a purposive being interacting with other purposive beings one can neither have one’s humanity violated nor can one violate the humanity of others. The individual person thereby ranks both as a separate and distinct source of claims and as a unit of responsibility.31 The distinctive character of the innate right is that it is not established by an act of acquisition, but belongs to each person by virtue of being born. (One’s right in one’s physical integrity is a paradigmatic example.) The right is a permanent singleness of their justificatory grounding. The present chapter underlines the role in that justificatory grounding of Kant’s notion of the innate right. 29 See ch 2, s 4. 30 I Kant, The Metaphysics of Morals (L. Denis ed, M Gregor tr, CUP 2017) 32 [6:236]. On the right of humanity in one’s own person see ch 2, s 4. 31 I take this formulation from P Benson, Justice in Transactions (Belknap Press 2019)346, 464.

Horizontality: Presuppositions and Functions  127 and inalienable feature of each person’s juridical position in relation to others. It is common to private law and constitutional law in two ways: first, in the innate right’s anticipation of constitutional protections, and second, in constitutional law’s retention and elaboration of what the innate right requires. The first kind of commonality is that, even in the conceptual space of the state of nature, when no civil condition yet exists and persons are conceived as related to each other only through the categories of private law, the innate right contains versions of the entitlements evident in modern constitutions.32 One of these, familiar in the constitutional context as the presumption of innocence, is that a person must be regarded as being beyond reproach, because one is not a wrongdoer until one’s action has affected the rights of others.33 A second is a form of equality, in which each person is innately independent from being bound to others without their being reciprocally bound. From this equality comes the standing of all persons as juridical equals; in the terminology of Roman law, persons are sui iuris rather than alieni iuris. This has a negative and a positive consequence. The negative consequence is that no-​one is entitled to dictate to others how they should live, or to receive preferential consideration in any dispute over rights. The positive consequence is that everyone is subject to legal categories—​of ownership and contract, for example—​that are structured in such a way as to reflect and express the innate equality of the interacting parties. A third entitlement, present in the freedom of thought and speech of modern constitutions, is the freedom to communicate one’s thoughts and to say whatever one wishes, regardless of whether one is being truthful or sincere, so long as one does not thereby infringe the rights of others. Thinking and speaking are integral to the exercise of one’s purposiveness in relation to others; they are comprehended within a person’s innate right because they form the cognitive and communicative basis of the right to humanity in one’s own person. These entitlements, Kant insists, are not separate rights but the various specifications of the single innate right that all persons have by virtue of their humanity. The innate right thus requires that others respect not only to the physical embodiment of a person’s humanity but also the entitlements that belong to a person as an independent and self-​determining being in interaction with others. In Kant’s account these entitlements anticipate in the state of nature the right to the presumption of innocence, to equality, and to freedom of thought and speech that are standard features of contemporary constitutionalism. The second kind of commonality results from the persistence of innate right into the civil condition, where it gains a new significance. In the civil condition a person’s innate right is not confined to the role that it played in the state of nature or in the working out of the private law categories that were then conceivable. Because 32 Kant (n 30) 34 [6:237–​38]. 33 For a recent discussion of this in a Kantian vein see H Stewart, ‘The Right to be Presumed Innocent’ (2014) 8 Crim L & Phil 407.

128  Ernest J. Weinrib freedom under law must be undiminished as compared to the lawless freedom of the state of nature,34 the innate right continues to undergird all legal relationships, including the omnilateral ones that owe their existence to the civil condition. In the preceding chapter I contended that the civil condition, although necessary for the actualization of reciprocal freedom, can produce new forms of dependency among persons, which the state is required to ameliorate through programmes of distributive justice. Now my point is that the civil condition creates a second unfortunate possibility: the state, which must have coercive power over everyone in the exercise of its public authority, might itself act inconsistently with the innate right of those subject to it. For instance (to use Kant’s example), the state might treat a convicted person who merits even the most severe punishment as an abomination; such treatment would be inconsistent with the criminal’s humanity, which, being innate, is not forfeited even through the commission of a crime.35 Accordingly, the state is subject to a general imperative to be attentive to the innate right of those whom it rules, that is (in the terms of the German constitution),36 to regard human dignity as inviolable and to make its respect and protection the duty of all state authority. Modern constitutions strive to fulfil this imperative by enshrining a catalogue of the rights, freedoms and immunities that specify the incidents of the innate right, especially those that are most germane to the particular society’s history and condition, and by elaborating a jurisprudence in which respect for human dignity is a paramount interpretive principle. A person’s innate right is thus as fundamental to constitutional law as it is to private law. Accordingly, the normative commonality of private law rights and of constitutional rights lies in the innate right that grounds them both. Therein also lies the juridical unity of the legal order as a whole. Kant’s assertion that ‘there is only one innate right’,37 stated with reference to the variety of entitlements to which the innate right gives rise, also applies to the significance of the innate right for different domains of law. From the Kantian perspective, all legal relations, whether between private parties or between the state and its subjects, have to be consistent with each person’s innate right. The entirety of the legal order is conceptually constructed on the foundation provided by the innate right, so that the reciprocal freedom of all animates both private and public law. Thus, a person’s innate right is the subjective right imminent in the law’s objective order of value, with the various entitlements of private law and of constitutional law being specifications of it or derivations from it. Given its fundamental and pervasive significance, the innate right of all persons is the idea that underlies both of horizontality’s presuppositions—​the juridical unity of the legal order and the common character of rights in the different



34

See ch 5, s 2. Kant (n 30) 116 [6:333]. 36 GG I, 1. 37 Kant (n 30) 34 [6:237]. 35

Horizontality: Presuppositions and Functions  129 domains of private law and constitutional law. In a leading Israeli judgment on horizontality, Justice Aharon Barak observed that ‘the obligation of an individual to respect his neighbour’s rights stems from the same source as the government’s obligation—​the basic social conceptions and the basic legal principles upon which the legal system is built’.38 Innate right is Kant’s formulation of the theoretical idea at the heart of those basic social conceptions and basic legal principles. Even when it is not explicitly a component of the grounds of liability, human dignity (and its Kantian analogue, innate right) is pervasively present in private law. Consider the following non-​exhaustive miscellany of instances from the common law. The obligation imposed by another’s innate right includes not only the non-​violation of physical integrity but also the avoidance of dignitary wrongs. Certain grounds of liability, such as defamation39 and the intentional infliction of mental suffering,40 are seen by some as wrongs of this sort. Moreover, in certain contexts, the analysis of the loss is understood not as a physical injury to a person or to property but as an adverse effect on dignity, that is, as an interference with the plaintiff ’s moral power to act for his or her own purposes rather than being subject to the purposes of another. As Justice Cardozo famously said: ‘Every human being of adult years and sound mind has a right to determine what shall be done with his own body.’41 From this standpoint, even beneficial conduct (for example, administering a blood transfusion to a severely injured person who has a religious objection to the procedure) can count as a wrong when it is inconsistent with the patient’s self-​determination.42 Moreover, battery is actionable even though the offensive contact caused no physical harm and even though the victim was unaware of the contact at the time it occurred; the wrongfulness of the conduct then consists solely in the indignity of the victim’s being used for another’s purposes. Similarly, in the case of wrongful birth, when a health provider’s negligence has resulted in a pregnancy that should have been avoided, some courts have held that the award of damages should reflect not the costs generated by the child’s existence but the wrongful interference with the mother’s reproductive autonomy.43 In the field 38 Hevra Kadisha (Burial Society) of the Jerusalem Community v Kastenbaum (1992) PD 46 (2) 464, para 23 (translation available at https://​law.ute​xas.edu/​intern​atio​nal/​fore​ign-​law-​trans​lati​ons/​isra​eli/​ case.php?id=​1391). 39 Hill v Church of Scientology [1995] 2 SCR 1130, 1175: ‘Although it is not specifically mentioned in the Charter, the good reputation of the individual represents and reflects the innate dignity of the individual, a concept which underlies all the Charter rights.’ A Ripstein, Private Wrongs (Harvard UP 2016)develops the view that the tort of defamation protects each person’s entitlement that his or her standing in the eyes of others be determined only by one’s own deeds and not by other persons. This treats defamation as the aspect of innate right that Kant termed ‘being beyond reproach’. Kant (n 30) 34 [[6:238], 83 [6:295]. 40 DG Reaume, ‘Indignities: Making a Place for Dignity in Modern Legal Thought’ (2002) 28 Queen’s LJ 61. 41 Schloendorff v Society of New York Hospital 211 NY 125 (1914). 42 Malette v Shulman 67 DLR (4th) 321 (Ont CA 1990). 43 Rees v Darlington Memorial Hospital NHS Trust [2004] 1 AC 309 (HL) critically analysed in N Priaulx, ‘That’s One Heck of an Unruly Horse: Riding Roughshod over Reproductive Autonomy in Wrongful Conception’ (2004) 12 Fem Leg St 317.

130  Ernest J. Weinrib of remedies, aggravated damages are awarded for breaching another’s right in a manner so reprehensible that it infringes on the dignity of the plaintiff as a right-​ holder. In addition, considerations of innate right are woven into the fabric of the various acquired rights, because presupposed in every acquired right is its consistency with the self-​determining freedom of the parties who interact in the process of acquisition. Several heads of public policy manifest the same normative impulse. Although public policy usually refers to a ‘prohibition . . . imposed in the interest of the safety of the state, or the economic or social well-​being of the state and its people as a whole’,44 some of its doctrines are invoked to bar arrangements that render an individual unable meaningfully to exercise some aspect of his or her innate right. Such arrangements are inimical to the ‘well-​being of the state and its people as a whole’ because they fail to accord with the minimal conditions of reciprocal freedom necessary for the normative viability of the legal order.45 I here give two examples. Consider first a contract that places one party under servile obligations to another. Not only is slavery as a legal condition barred, but so is a contract that effectively deprives another of freedom of action. The classic case deals with the invalidation of mortgage covenants that tied the borrower to his employment and to his dwelling, that assigned his present and future earnings to the creditor, and that forbade him from borrowing money or pledging his credit on the security even of any of his unencumbered belongings.46 Such arrangements were inconsistent with the entitlement to be sui juris as a matter of one’s innate right, because they transform a person into the equivalent of a medieval serf.47 A second example is the invalidation of a contract under which a newspaper that held itself out as giving honest advice about land purchases, promised not to publish anything about a particular land dealer.48 This contractual term, which required the newspaper to abstain from reporting even fraudulent activity that fell within the ordinary ambit of its coverage, was against public policy; as the court noted,49 the restriction on publishing might amount to ‘actually and actively mislead[ing]’ the newspaper’s public into supposing that a particular land scheme promoted by this dealer was not fraudulent, even if the newspaper knew that it was. The newspaper could not contractually bind itself not to speak about certain land transactions, as that would be incompatible with the duty that the newspaper had assumed to its readership. In Kantian terms, freedom to speak or not to speak is 44 Re Millar Estate [1938] 1 SCR 1, para 15 (Duff CJC). 45 ‘By the well-​being of the state must not be understood as the welfare of its citizens and their happiness . . . [but] that condition in which its constitution conforms most fully to the principles of right.’ Kant (n 30) 103 [6:318]. 46 Horwood v Millar’s Timber and Trading Co [1916] I KB 305; similarly, Arthur v Society of the Sacred Heart of Jesus 9 OLR 474 (Ont CA 1905). 47 Horwood (n 46) at 311 (Lord Cozens-​Hardy MR). 48 Neville v Dominion of Canada News Company [1915] 3 KB 556 (CA). 49 ibid 568.

Horizontality: Presuppositions and Functions  131 an aspect of the innate right, but only insofar as it does not have the tendency diminish what rightfully belongs to another.50 The presuppositions of unity and commonality imbue the phenomenon of horizontality with both an innovative and a conservative dimension. The innovative dimension is that horizontality enlarges the resources for elaborating the entitlements of innate right that are accessible to private law. In principle, the full range of constitutional rights that are specifications of human dignity is admissible to the legal reasoning about a dispute between private litigants. To take examples mentioned in this chapter, a restriction on speech becomes relevant not only where contractually compelled silence has a tendency to promote fraud (as in the newspaper case), but also where it prevents the organizing of a boycott against a racist film-​maker (as in the case with which this chapter began). The conservative dimension is that the framework of innate right within which this innovation takes place is not alien to private law. This, after all, is what the presupposition of normative continuity entails. A constant apprehension especially among private lawyers is that horizontality subordinates private law to constitutional law, and thereby also subordinates individual freedom and the legal processes on which it depends to the misplaced demands of a state’s public law.51 This may well be the case at the level of positive law doctrine, but may be less radical when considered from the more abstract standpoint of the private law’s underlying normative foundation in each person’s innate right.

3.  Three Functions of Horizontality: An Outline I now turn to the functions of horizontality within the legal system: what is horizontality supposed to do? By ‘function’ in this context I do not refer to a mode of achieving some independently desirable benefit to which horizontality is instrumental. Rather, my enquiry is into the distinctive kind of contribution that horizontality might make to the exposition and concretization of rightful relations in a civil condition. The different possible functions of horizontality are the different ways in which horizontality might be considered immanent in a system of rights. The delineation of horizontality’s functions, so understood, is the object of the rest of this chapter.

50 Kant (n 30) 34 [6:238]. 51 See especially the critique of horizontality in U Diederichsen, ‘Das Bundesverfassungsgericht als oberstes Zivilgericht: Ein Lehrstück der juristischen Methodenlehre’ (1998) 198 ACP 171. An illuminating study of the contrast between subordination and complementarity is O Cherednychenko, Fundamental Rights, Contract Law and the Protection of the Weaker Party: A Comparative Analysis of the Constitutionalization of Contract Law, with Emphasis on Risky Financial Transactions (European Law Publishers 2007).

132  Ernest J. Weinrib One can distinguish three such functions: first, to render legal norms more determinate; secondly, to shape the development of private law; and, thirdly, to give expression within private law to the constitutional construction of the equal human dignity of the interacting parties. To summarize each in turn: First, under the determinacy function constitutional rights figure as sources of legal reasoning that, like other sources of legal reasoning, impart specificity to more abstractly formulated requirements for liability. Secondly, under the development function constitutional rights provide values that courts ought to take into account in developing private law. The development function differs from the determinacy function in that it places judges under a duty—​unparalleled for other sources of legal reasoning—​that is specific to the constitutional status of the rights. Thirdly, under the dignity function, constitutional rights give rise to a set of principles about human dignity to which private law should give a maximally concordant effect. Like the development function, the dignity function places the courts under a duty to orient the private law towards the actualization of constitutional values. However, these two functions are importantly different. Whereas the development function envisages the development of private law within existing doctrinal contours, the dignity function allows for the expansion of private law beyond those contours. And whereas the development function elaborates ideas about interaction that have already been articulated within the doctrinal repertory of private law, the dignity function makes the legal terms on which private parties interact expressive of constitutionally based principles of human dignity even in doctrinally novel situations. These three functions vary in the robustness of the role that they each assign to constitutional rights. The first addresses the law’s substance by deploying constitutional rights to assist in the disclosure of what a particular doctrine is. The second addresses the law’s causality by highlighting the significance of constitutional rights for the direction in which private law doctrine should evolve. The third addresses the law’s functioning as a systemic community of norms under which interacting parties are reciprocally related to one another as bearers of equal human dignity. Each function in this list incorporates but goes beyond the previous one. Although they all draw on the ideas of human dignity that underlie the legal order as a whole, each stands in a different relation to the private law as a posited body of norms. The first function focuses on a particular posited doctrine, the second on a systemic obligation for developing the posited norms, and the third on the constitution as a mandatory source of ideas about dignity even if the posited private law has hitherto not explicitly recognized them. The three functions thus provide a progressively greater alignment of private law and constitutional principles.52 52 Compare the slightly different listing in Dürig (n 10) 177–​81 of the ‘drei Intensitätsgrade’ for elaborating private law concepts and general clauses by reference to constitutional values: the mere clarification (bloße Verdeutlichung) of otherwise plastic concepts, the value-​accentuated interpretation

Horizontality: Presuppositions and Functions  133

4.  The Determinacy Function The determinacy function arises out of the role of courts within the civil condition as the legal institutions that determine the meaning of rights in the context of specific interactions. Although the organizing concepts of private law are conceivable as implications of reciprocal freedom in the state of nature, they can become actual only through authoritative legal institutions—​the legislature as codifier and the judiciary as adjudicator—​that work out, as a matter of public reason, the norms that apply to particular transactions. Does the right to physical integrity, for instance, include protection against an apprehended contact or merely an actual one (that is, is there a tort of assault)? Does this right include security against mental suffering? These institutions determine the content of the rights to which interacting persons are entitled. Otherwise, rights cannot operate as juridical expressions of the reciprocal freedom that everyone is to enjoy. To determine the content of private law rights and other private law norms, courts function as institutions of public reason that view the litigants as participants in a social relationship within a world of shared social meanings. The working out of the juridical significance of the parties’ conduct is historically variable and relative to societal contexts and understandings. Courts draw on the public culture (which of course includes their own prior determinations) in order to give concrete meaning to the legal categories expressive of the parties’ reciprocal freedom as normatively equal participants in a particular transaction. As the evolution of liability for mental suffering shows, the considerations regarded as relevant to this process can change over time as new forms of knowledge and new appreciations of the social implications of different behaviours arise. In this way the specific content of the rights is derived, not through philosophical speculation, but through reference to beliefs, values, and modes of reasoning that have public plausibility.53 Even when operating in contexts not indigenous to private law, the authoritative institutions of the civil condition can affect the judicial project of determining the content of private law’s categories. The relevance of statutes to private law illustrates this point. Statutes that promote safety and penalize those who endanger others regulate the same behaviour that private law addresses. These statutes may not only be enforced in their own terms but may also, where appropriate, influence the setting of private law standards. In Canada, for instance, such statutes are regarded not as formulating a per se standard or as grounding a distinct tort of statutory breach, but as providing evidence for determining whether the defendant should be considered negligent under the general principles of tort law.54 Accordingly, (wertgeschärften Auslegung) of private law’s protective norms, and the filling in of what from the constitutional standpoint are gaps (Lückenausfüllung) in the received system of private law protections.

53

54

See especially J Rawls, ‘The Idea of Public Reason Revisited’ (1997) 64 U of Chi L Rev 765, 766–​73. The Queen in the Right of Canada v Saskatchewan Wheat Pool (1983) 143 DLR (3d) 9 (SCC).

134  Ernest J. Weinrib even when not directly applicable to the facts being litigated, statutory norms can be regarded ‘as crystallizing a relevant fact situation which, because of its authoritative source, the Court [is] entitled to consider in determining, on common law principles, whether a duty of care should be raised’.55 In such a case the statute assists in specifying the content of private law categories in contexts beyond its scope as a legislative command. Under horizontality’s determinacy function, constitutional protections of human rights play a similar role. The constitution, as society’s authoritative repository of legally supreme and publicly accessible values, provides a pre-​eminent source on which public reason can draw to give concrete meaning to private law categories. Indeed, from the standpoint of public reason, the constitution crystallizes in a uniquely conspicuous way the values to be recognized in the process of ascribing specific meanings to the rights and correlative obligations of private law. This is so even if the purpose of the constitutional document is to restrain government, rather than to regulate the interaction of private persons. An illustration of horizontality in this sense is the Canadian case of Canada Trust v Ontario Human Rights Commission,56 which invoked constitutional values to strike down the racially discriminatory terms of a charitable trust. The trust had established scholarships whose eligibility criteria expressly reflected notions of racial superiority. These were invalidated as contravening ‘current principles of public policy under which all races and religions are to be treated on a footing of equality and accorded equal regard and equal respect’.57 The Court noted that the criteria were ‘patently at variance with the democratic principles governing our pluralistic society in which equality rights are constitutionally guaranteed and in which the multicultural heritage of Canadians is to be preserved and enhanced’.58 This reference to the equality and multiculturalism provisions of the Canadian Charter of Rights and Freedoms was particularly apt, because charitable trusts must provide a public benefit—​with which (as the Charter provisions indicated) racist criteria of eligibility are incompatible. Under the rubric of public policy the Charter was used to concretize the notion of public benefit that lies at the root of charitable trusts, because ‘[i]‌t is this public nature of charitable trusts which attracts the requirement that they conform to the public policy against discrimination’.59 Here, constitutional values were relevant to a private law matter even though the Canadian Charter explicitly applies only to legislative and governmental acts. The Canada Trust case exemplifies the determinacy function rather than the development function of horizontality. The Court made use of the Charter provisions in order to determine the meaning of public benefit in the context of a charitable

55

Jordan House v Menow (1973) 38 DLR (3d) 105, 109 (SCC) (Justice Laskin). (1990) 69 DLR (4th) 321 (Ont CA). 57 ibid 334. 58 ibid. 59 ibid 353 (Tarnopolsky J concurring). 56

Horizontality: Presuppositions and Functions  135 trust, but it did not refer to a specific obligation to develop the law in conformity with the Charter values. The Charter (as well as other human rights instruments mentioned in the judgment) was simply material that was ready to hand to give content to a general norm. The role of the Charter in determining that no public benefit existed was no different from the role in Canadian law of safety legislation in providing standards for the law of negligence. Although neither the constitution nor safety statutes directly govern private law controversies, they nonetheless help to specify the meaning of private law categories and standards.

5.  The Development Function The difference between the determinacy function and the development function is that the latter, but not the former, requires courts to develop private law in accordance with constitutional rights or their underlying values. The determinacy function treats constitutional rights and values as one source of legal reasoning among many. In performing their obligation to exercise public reason appropriately, courts must consider all relevant sources of legal reasoning. This obligation is general rather than specific to constitutional rights and values. In contrast, under the development function the courts are obligated specifically to develop the law in accordance with the constitution or with other human rights instruments. This duty may be self-​imposed, as when the Supreme Court of Canada announced that, although constitutional rights do not apply to private law, ‘the judiciary ought to apply and develop the principles of common law in a manner consistent with the fundamental values enshrined in our Constitution’.60 Or it may be legislatively or constitutionally imposed as an obligation applicable to the exercise of all public authority, as when the Human Rights Act in the United Kingdom makes it ‘unlawful for a public authority’ (specifically defined to include courts and tribunals) ‘to act in a way that is incompatible with a convention right’.61 The distinguishing feature of the development function, therefore, is that it marks out constitutional rights or the values that animate them as the particular objects of a judicial obligation. In this respect constitutional rights and values are no longer analogous to other sources of law such as legislated safety provisions. A statutory safety standard relevant to the breach of a common law duty of care cannot be said to create a judicial obligation to develop the common law consistently with that particular statute or even with the entire ensemble of statutes. The statutory provision is merely incidental to the development of the common law, providing evidence of the standard in a particular case. Its availability is a happy fortuity that a plaintiff can exploit and that a court can find helpful. It does not

60 61

RWDSU v Dolphin Delivery Ltd [1986] 2 SCR 573, para 39. Human Rights Act, s 6(1).

136  Ernest J. Weinrib present a norm to be incorporated for its own sake. However, when the principles of the common law must be developed in a manner consistent with constitutional values, those values acquire an affirmative and non-​incidental significance for private law. Judges are obliged to regard the constitutional rights as a magnet towards which the various shavings of private law doctrine are drawn. Under the development function these shavings already have a recognized status in legal doctrine. Development in accordance with constitutional values takes place from specific doctrines that are current in private law. The invocation of constitutional values looks both forward and backward: forward to the law’s alignment with constitutional values, and backward to particular doctrines of private law. What private law develops from is as relevant as what it develops towards. The Canadian treatment of horizontality illustrates this dynamic. In addition to specifically stating that the Canadian Charter of Rights and Freedoms applies only to legislatures and governments, the Constitution of Canada also declares itself the county’s supreme law and renders inoperative ‘any law’ (in the French version, ‘tout autre règle de droit’) that is inconsistent with its provisions.62 The Supreme Court of Canada has interpreted these apparently contrasting sections as bifurcating the significance of the Charter for private law. On the one hand, private persons do not owe constitutional duties to one another, nor can private causes of action be founded on the Charter. On the other hand, judicial rulings ought nonetheless to develop the common law consistently with Charter values. This developmental obligation arises at the point at which a court’s inherent jurisdiction to extend or modify the common law in the light ‘of prevailing social conditions and values’ intersects with the Charter as the ‘restatement of the fundamental values that guide and shape our democratic society and our legal system’.63 On this interpretation, the entirety of the legal system, including its private law, is to be ‘guided and shaped’ by the values inherent in the Constitution as supreme law. The regulative role of the Constitution for the legal system as a whole obliges courts to exercise their inherent jurisdiction in order to develop the common law in accordance with Charter values. To constrain this obligation within proper adjudicative limits, judicial revisions to the common law in horizontality cases must be only ‘incremental’—​the Court’s characterization of appropriate judicial creativity in non-​horizontality cases as well64—​thereby leaving major changes to the legislature. This incremental process is said to render the common law ‘consistent with Charter values without upsetting the proper balance between judicial and legislative action’.65



62

Constitution Act 1982, s 32(1) and s 52(1). Hill v Church of Scientology of Toronto (n 39) paras 91–​92. 64 Watkins v Olafson [1989] 2 SCR 750. 65 R v Salituro [1991] SCR 654, para 49. 63

Horizontality: Presuppositions and Functions  137 Within this structure of analysis much depends on whether the contemplated change is indeed incremental. This characterization is easier to sustain by modifying existing causes of action than by establishing new ones. The most notable example of this is the recognition of responsible communication on matters of public interest as a defence to defamation.66 This is a significant doctrinal development within an existing cause of action. The Supreme Court of Canada is explicit that, because it concerns the substance and not the occasion of the publication, this is a new defence rather than an extension of the traditional defence of qualified privilege.67 This focus on substance is animated by principles that the Court had earlier identified as underlying the Charter’s guarantee of freedom of expression, namely, that this freedom is essential to the proper functioning of democratic government and that it is a precondition of the search for truth. These principles support the common law’s protection for published statements on matters of public interest even if these statements cannot be proven to be true. Presumably, the incremental nature of this change follows from its providing an additional defence within an established tort that already has a suite of judicially created defences. In one instance, however, the invocation of Charter values in accordance with the development function has led to a new common law cause of action.68 A bank employee who used her position to make unauthorized examinations of the plaintiff ’s personal bank account was held liable for intruding into another’s personal affairs.69 This was the first recognition in Canada at the appellate level of a new privacy tort of ‘intrusion upon another’s seclusion’. The judgment set out the converging variety of legal considerations that had previously brought Canadian courts at least to the threshold of the recognition of this tort. First, Charter jurisprudence had articulated the special way in which privacy underlies specific Charter rights. Secondly, legislation in provinces that had enacted a statutory tort of invasion of privacy had done so in sweeping terms, leaving courts in those provinces with approximately the same task as courts in provinces without such legislation. Accordingly, the judicial recognition of this tort would not upset the balance between legislative and adjudicative action. Thirdly, aspects of privacy had long been recognized as an underlying value in various traditional causes of action such as defamation, breach of confidence, breach of copyright, and nuisance and other property rights. Fourthly, a long history of academic writing had supported recognition of a right of action for breach of privacy. In view of the role of Charter values in the inherent jurisdiction of courts to adjust the common law to prevailing social conditions, these considerations led the Court to affirm the common law’s capacity to take what the Court regarded as an incremental step in response to the 66 Grant v Torstar Corp [2009] 3 SCR 640. 67 ibid paras 92–​95. 68 On the Canadian approach, this new cause of action is a creation not of the Charter but of the common law informed by Charter values. 69 Jones v Tsige 346 DLR (4th) 34 (Ont CA 2012).

138  Ernest J. Weinrib novel threat to privacy posed by the routine collection of personal information in a readily accessible digital form. One finds similar attention to incremental change in the judicial treatment of privacy in the United Kingdom. Under the Human Rights Act, it is unlawful for courts to act incompatibly with the rights guaranteed under the European Convention of Human Rights, which includes a right to respect for one’s private and family life. Although the Convention deals with rights against the state, this Convention right has been held to apply horizontally as well, on the ground that no reason exists to have less protection against a private individual than one would have against the state.70 Nonetheless, it has been held not only that no general tort of invasion of privacy exists at common law,71 but also that the Human Rights Act does not create new causes of action between private parties.72 Instead, the Human Rights Act has spurred the development of a tort protecting only the specific aspect of privacy that is concerned with personal information, because the pedigree of this tort can be traced back to the well-​established idea of breach of confidence. This tort originated in pre-​existing relationships in which information confidential in nature was imparted in circumstances that signified a duty of confidence with respect to it. It subsequently developed to cover confidential information, however obtained, even in the absence of a prior confidential relationship, and then, under the stimulus of the Human Rights Act, to protect against the disclosure of private information or images whether confidential or not.73 The recognition of a tort concerning private information is merely the latest incremental stage in the evolution of the action for breach of confidence. Thus, although courts are obliged to act compatibly with Convention rights, this statutory directive applies not at large but rather to a relevant cause of action. The statutory duty is premised on an already existing component of the common law that is available for further adjudicative development by reference to Convention rights. In the case of the disclosure of private information the relevant cause of action is breach of confidence, into which the rights protected by the Convention are absorbed.74 A merit of the developmental function’s incrementalism is that it presents the enquiry into the role of human rights in private law as a familiar adjudicative activity. The stage-​by-​stage evolution of a private law doctrine such as breach of confidence is ‘typical of the capacity of the common law to adapt itself to the needs of contemporary life’.75 The burden on a court of appreciating the social consequences of its decision is relatively manageable when change is gradual and limited. Moreover,

70

Campbell v MGN Ltd. [2004] 2 AC 457, para 50 (HL) (Lord Hoffmann). Wainwright v Home Office [2004] 2 AC 53, para 18 (HL). 72 Campbell v MGN Ltd [2004] 2 AC 457, para 125 (HL) (Baroness Hale). 73 ibid paras 43–​49 (Lord Hoffmann). 74 ibid para 132 (Baroness Hale). 75 ibid para 45 (Lord Hoffmann). 71

Horizontality: Presuppositions and Functions  139 such change is less likely to be perceived as trenching on the domain and responsibility of the legislature.76 Incremental change thus conforms to Oliver Wendell Holmes’s famous aphorism that, although courts inevitably legislate, ‘they can only do it interstitially; they are confined from molar to molecular motions’.77 Nonetheless, the law that emerges at the end of this incremental process differs significantly from the law that existed at its inception. The development from breach of confidence to infringement of privacy moves from protection against having one’s trust abused by the disclosure of information held in confidence, to protection against the disclosure of personal information generally. This transition entails significant changes in the scope of liability, in its structure, and in its basis. The change in scope is that privacy is concerned with personal information, whereas breach of confidence was concerned with any information, whether personal or not (valuable commercial information, for instance) that was given in confidence. The structural change is that a general limit on the disclosure of personal information has to be aligned with the right of others, especially of the media, to free expression. In contrast, the disclosure of confidential information as originally conceived necessitated no such alignment. In the paradigmatic case of confidential information, for instance, the public interest in revealing what Prince Albert’s etchings might have disclosed about him was not an issue despite his eminent public station.78 The change in the basis of liability is that liability rests no longer on a duty of good faith with respect to confidential information whether personal or not, but on the identification of personal information ‘as something worth protecting as an aspect of human autonomy and dignity’.79 This aspect consists in ‘the right to control the dissemination of information about one’s private life and the right to the esteem and respect of other people’.80 Presumably underlying these sentiments is the thought that the right to control one’s own personal information is integral to the capacity to construct the terms of one’s interaction with others, including the capacity to calibrate disclosure to particular levels of intimacy.81 This right confirms that a person is more than an object of scrutiny or a source of information or a means to the satisfaction of others’ curiosity.82 Instead, self-​determination about the disclosure of personal information allows the presentation of oneself to the world as a person, that is, as an agent who interacts with others as master of one's own private sphere and not as someone at 76 These are the reasons for incrementalism mentioned in Watkins v Olafson (n 64). For emphasis on the importance of incrementalism in the context of horizontality under the Human Rights Act see G Phillipson and Alexander Williams, ‘Horizontal Effect and the Constitutional Constraint’ (2011) 76 MLR 878. 77 Southern Pacific v Jensen (1917) 244 US 205, 221. 78 Prince Albert v Strange (1849) 41 ER 1171. 79 Campbell v MGN Ltd (n 72) para 50. 80 ibid para 51. 81 C Fried, ‘Privacy’ (1968) 77 Yale LJ 475. 82 SI Benn, ‘Privacy, Freedom, and Respect for Persons’’ in JR Pennock and J Cooper (eds), Privacy: Nomos XII (Harvard UP 1971) 1.

140  Ernest J. Weinrib the mercy of others’ exposure. Thus, the right to privacy is an expression (in our society, at least) of what Kant, in explaining the innate right, called the right of humanity in our own person: it protects our capacity to assert ourselves as ends in relation to others.83 The development of the right to privacy indicates that, even within the development function, horizontality can substantially transform the positive private law. This is because the forward push of the right-​based developmental principle is more powerful than the backward pull of the antecedent doctrine. This is as it should be. For if, as I contended in the first two chapters of this book, private law actualizes reciprocal freedom by making determinate the abstractions appropriate to a system of rights, then those abstractions have to be regulative of those determinations. The determinations draw their breath from the abstractions even as they actualize them in the positive law. This, in turn, suggests that one should perhaps not overstate the importance of incrementalism. It is, of course, the case that skilful judges produce admired and innovative judgments by reweaving the threads of pre-​existing legal material into a new fabric.84 And it is also the case that the availability and use of such material reinforces the legitimacy of judgments as adjudicative rather than as legislative exercises. However, the difference between judicial activity and legislative activity does not depend on the proximity of a judicial decision to pre-​existing case law; if it did, some of the great cases in which modern private law broke with the past would turn out to have been wrongly decided.85 Instead, so far as private law is concerned, the legitimacy of adjudication ultimately depends on the fairness and coherence of the connection that the judge’s reasoning forges between the normative position of the particular plaintiff and that of the particular defendant.86 As a matter of justice—​that is, of the corrective kind of justice pertinent to private law—​fair and coherent reasoning about the parties’ relationship trumps incremental reasoning about the law’s development. Accordingly, although the incremental character of judicial decision is useful and desirable, it is not necessary for the legitimacy of adjudication, nor does it represent ‘a fundamental constitutional constraint on the developmental activities of judges’.87 Be that as it may, within the development function human dignity occupies an ambiguous position. On the one hand, as the value underlying fundamental human and constitutional rights, human dignity directs the development of private 83 Kant (n 30) 32 [6:236]. 84 For an appreciation of Justice Robert Sharpe’s judgement in Jones v Tsige (n 69) see TDC Bennett, ‘Privacy, Corrective Justice, and Incrementalism: Legal Imagination and the Recognition of a Privacy Tort in Ontario’ (2013) 59 McGill LJ 49. 85 In the oral argument of Dobson v Dobson [1999] 2 SCR 753, one member of the Supreme Court of Canada speculated that if Donoghue v Stephenson, [1932] AC 562 (HL) was argued before that court today, the incrementalism criterion would cause the case to be decided for the defendant. 86 EJ Weinrib, The Idea of Private Law (Harvard UP 1995, OUP 2012)208–​209. 87 Phillipson and Williams (n 76) 903.

Horizontality: Presuppositions and Functions  141 law by requiring it to be consistent with the values represented by those rights. On the other hand, the role of human dignity is contingent on the merely incremental nature of the change to the pre-​existing law that would be effected by developing it in accordance with constitutional values. Accordingly, the development function treats human dignity as both regulative of and incidental to the evolution of private law. In this respect, the development function of horizontality manifests a tension between the human dignity towards which, and the positive law from which, private law develops.

6.  The Dignity Function This tension is absent from the dignity function, to which I now turn. Under the dignity function, a court is obligated to adjudicate in accordance with constitutional values regardless of whether this produces a merely incremental development of the law. These values and the rights in which they are embodied are the specifications of the single overarching idea—​the ‘mother right’ as Aharon Barak has termed it88—​of human dignity. The dignity function requires courts to treat that idea as paramount in the legal relationships between private persons, just as the constitution makes it paramount in the relationship between the state and those it governs. The development function and the dignity function vary in the way that courts connect the private law as antecedently posited to the constitutional norms that might influence it. The development function anchors the court’s enquiry in the antecedent private law, to which it tethers the constitutional norms to the extent that incremental development allows; the dignity function, in contrast, assigns primacy to the constitutional norms, in the light of which the posited private law may be revised and reshaped. The judicial obligation under the dignity function emerges from the stringent and comprehensive conception—​delineated in Kant’s account of the civil condition—​of the role of all state institutions. Within a system of rights the state is obliged to exercise its official power in a way that secures the freedom of everyone who is subject to its authority. Operating through its official institutions as they act within their competence and authority, the state integrates everyone’s rights into an effective and publicly authoritative system that actualizes the freedom of all. This is the meaning of the Kantian notion that persons emerge from the state of nature into the civil condition with their freedom undiminished, despite their subjection to the state’s panoply of coercive powers. The undiminished character of each person’s freedom requires the state to bring itself into an ever-​closer approximation of the condition in which that freedom is most adequately realized.



88

A Barak, Human Dignity: The Constitutional Value and the Constitutional Right (CUP 2015) 157.

142  Ernest J. Weinrib In the Kantian view, the state fulfils its role as the guarantor of the freedom of all by treating freedom as ‘the principle and indeed the condition of any exercise of coercion’.89 For Kant, the idea of a system of rights is not merely a way of describing a distinctive kind of legal system; it also refers to the normative standard that ought to animate all exercises of public authority within any legal system. For a system of rights organized on the Kantian model, freedom is the guiding idea governing all state activity, which must work towards its ever fuller and more adequate actualization over time. For the relations between persons, this freedom is realized through the fair and coherent operation of the various private law categories of right that, for any given transaction, set the boundary between one person’s freedom and another’s. For relations between a person and the state, this freedom is realized through the operation of public law norms, both constitutional and non-​constitutional, that reflect the pre-​eminent status of human dignity. Human dignity, as the contemporary formulation of innate right with respect to the omnilateral relationships of the civil condition, is the most abstract and general representation of persons’ independence in all their legal relationships, whether with one another or with the state itself. The state, conceived as the totality that unites its members for the realization of their rights, is the normative entity that gives legal effect to the co-​existence of everyone’s reciprocal freedom. In striving for the progressive actualization of freedom, the state is engaged in a dynamic and historically contingent exercise. Among the aspects of this contingency are the following: First, the steps that are to be taken towards a more adequate realization of everyone’s freedom reflect the deficiencies with respect to rights that happen to be present in a given legal system at a given time and place. These deficiencies, in turn, may be the manifestations of a particular historical pathway in which the fundamental and systemic violation of rights may call for bold and comprehensive amelioration. Or they may reflect the more sporadic but still serious residues of the absence of reciprocal freedom during a society’s gradual evolution towards a more mature system of rights. Secondly, the public culture of a given legal system may over time develop new determinations of human dignity, such as a right to privacy, or new institutional responses to rights violations, such as the judicial review of allegedly unconstitutional legislation. Thus, the conceptual and institutional resources for developing a system of rights vary over time and across jurisdictions. Thirdly, the task of actualizing freedom involves a continuous and infinitely protracted effort rather than the achievement of a definitive and predetermined utopia, because the progressive realization of freedom itself creates new legal possibilities for freedom’s further realization.90 Thus, the state’s actualization 89 Kant (n 30) 122 [6:340]. 90 ‘[W]‌hatever might be the highest degree of perfection at which humanity must stop, and however great a gulf must remain between the idea [of maximum reciprocal freedom] and its execution, no-​ one can or should try to determine this, just because it is freedom that can go beyond every proposed boundary.’ I Kant, Critique of Pure Reason (PW Guyer and AW Wood trs, CUP 1988) 397 [A317/​B374].

Horizontality: Presuppositions and Functions  143 of freedom is a dynamic process, antecedently fixed neither in its starting point nor in its execution nor in its endpoint. The judicial obligation to infuse constitutional rights or values into private law is an instance of the general requirement, incumbent on the state through all its institutions, to actualize freedom. Just as legislative and executive activity must conform to the fundamental rights set out in the constitution, so the judiciary must take account of those rights in carrying out its adjudicative role. The assumption underlying horizontality is that the constitutionally enumerated rights are markers of the freedom that the state must actualize for reasons that are not exhausted by the threats that the legislature and the executive might pose to them or by the inadequacies that the legislature and executive might exhibit in realizing them. Under the dignity function, the constitutional status of human dignity and of the basic rights that specify it imposes on the court, as an authoritative body charged with the actualization of freedom through state power, the obligation to treat the constitutional norms as definitive for private law relationships as well. German constitutional jurisprudence provides the clearest contemporary illustration both of this conception of the judicial role and of the dignity function of horizontality. So far as the judicial role is concerned, the jurisprudence has proceeded along two lines of thought.91 First, since its earliest cases on horizontality,92 the German Constitutional Court has repeatedly affirmed that basic rights are not merely ‘value-​neutral’ defensive bulwarks against state action but also objective norms that apply to all domains of law, including civil law.93 Accordingly, in exercising its adjudicative authority, a court has to treat those rights not only as subjective entitlements of the citizen as against the state, but also as objective principles to which the state is obliged to adhere in adjudicating the claim of one person against another. What the constitution originally formulated as a set of rights is interpreted as a set of values that courts are obligated to actualize. Secondly, the Basic Law proclaims human dignity to be inviolable, and places all state authority under the duty ‘to respect and protect it’.94 Moreover, it declares the basic rights that it enumerates to be binding as directly applicable law on all branches of government, including the judiciary.95 The Constitutional Court has interpreted these provisions as requiring the state through its institutions not only to ‘respect’ constitutional rights by refraining from governmental actions that violate them, but also to ‘protect’ these rights even when threatened by non-​governmental actors.96 This all-​embracing protective function of the state applies in the first instance to the 91 On the controversy about the relation between these two lines of thought see Cherednychenko (n 51) 103–​13. 92 Lüth (n 1). 93 See eg BVerfGE 81, 242 (Handelsvertreter 1990); 33 BVerfGE 303 (Numerus Clausus 1972). 94 GG I, 1. 95 GG I, 3. 96 D Grimm, ‘The Protective Function of the State’’ in G Nolte (ed), European and US Constitutionalism (CUP 2005) 137.

144  Ernest J. Weinrib legislature, which is obliged to enact laws—​civil and criminal codes, for example—​ that protect one person’s rights from violation by another. It also applies to the judiciary in the event that legislative action is absent, insufficient, or indeterminate. Thus, the judiciary is obligated in such cases to consider constitutional rights as pertinent to private law controversies. These two lines of thought require the judiciary, like the other institutions that exercise official power, to play its part in contributing to the fullest realization of the freedom of all. This includes making its private law judgments conform to constitutional norms. Corresponding to this conception of the judicial role in Germany (although developed independently of the notion of the state’s protective function) is the non-​ incremental character of horizontality under the dignity function. The elaboration in German law of liability for invasions of privacy is a well-​known and conspicuous example. Briefly put, the story of this development goes like this: In the drafting and enactment of the German Civil Code a deliberate decision was made, over the objection of prominent jurists, that except when consequential on physical injury or restraint, no claim to monetary damages for intangible injuries would be available. Nonetheless, in the aftermath of the Second World War, the German civil courts recognized the existence of a general private law right of personality on the basis of the new constitutional rights to human dignity and the free development of the personality.97 Moreover, having recognized this right, they subsequently held that the right would be ineffective unless the injured party could claim monetary satisfaction for its breach. This monetary award was thought to be particularly appropriate for defendants who had commercially profited from the infringement of the plaintiff ’s personality right, for example, by using the image of a gentleman riding a horse to advertise a sexual enhancement pharmaceutical,98 or by referring to a professor of canon law as vouching for the scientific efficacy of a medicinal herb.99 The Constitutional Court confirmed these developments in its Princess Soraya judgment,100 refusing to overturn a monetary award against a magazine for publishing made-​up interviews supposedly given by the former wife of the Shah of Iran. The magazine claimed that, although the constitutional value accorded to human dignity supported the recognition of a general right of personality, the remedy of an award of damages for intangible injuries usurped legislative authority, in violation of the constitutional principle of separation of powers; it thereby unconstitutionally restricted the magazine’s free expression. This claim the Court rejected. A more dramatic example of the non-​incremental character of horizontality’s dignity function is hard to imagine. Private law in Germany is governed by the



97

BGHZ 13, 334 (1954). BGHZ 26, 349 (Herrenreiter 1958). 99 BGHZ 35, 363 ( Ginsengwurzel 1961). 100 BVerfGE 34, 269 (1973). 98

Horizontality: Presuppositions and Functions  145 precise formulations of an elaborate, respected and legislatively authorized code, but that code had from the beginning explicitly set its face against general monetary awards for intangible injury. Thus, what the Civil Code excluded—​deliberately, as the Constitutional Court acknowledged—​the Court allowed. Even decades later one distinguished critic, who sympathized with the result as a matter of policy but was repelled by the blatant disregard of express provisions of the Civil Code, remarked that the Court gave constitutional sanction to ‘the most spectacular example of a legal development that was contra legem’.101 From the Court’s standpoint, the world of Princess Soraya was much different from the world of the Code’s legislators many decades earlier. The Nazi horror had attenuated the scope of free personality to a previously inconceivable degree. Also previously inconceivable were the possibilities for breaking into the individual’s personal sphere that were presented by technological advance, by new forms of publicity and advertising, by the increased emphasis on information, and by the role of mass media. Moreover, the constitution now enshrined a value system of fundamental rights centering on human dignity and the free development of the personality within the social community. Under these constitutional values the individual alone had the right to decide whether and how to make the events of her private life available to the public. Although freedom of the press was also constitutionally protected and had its place in the constitutional balancing, a court could consider in a specific case (such as in connection with the publication of an invented celebrity interview) whether the press was contributing to the formation of public opinion or was instead providing its readership with superficial entertainment. Nor did a court have to await the passage of amending legislation that might never come. Given that the Basic Law asserted that the judiciary was bound by ‘law and justice’ (Gesetz und Recht),102 a judge need not maintain formal adherence to the posited law at the price of significant loss of justice in a particular case. Once a general private law right to personality was recognized as an effect of the constitution, a lacuna unforeseeable at the time of the Civil Code was revealed regarding the remedy that would bring private law into line with ‘a changed sense of right and the values of a new constitution’.103 In allowing the damages remedy, the Constitutional Court acted in accordance with what has since become a standard feature of contemporary constitutionalism, that the constitution and its provisions are to be treated purposively.104 The constitution’s guarantee of human dignity led to the recognition of a private law right to informational self-​determination; the availability of a monetary award was necessary if human dignity and the private law right that it spawned were to be



101

Diederichsen (n 51) 193. GG 20, 3. 103 BVerfGE 34, 269, C, IV, 3. 104 Barak (n 88) 69. 102

146  Ernest J. Weinrib effective. The purpose of having human dignity as a constitutional value would not be adequately achieved if magazines could benefit by using fictitious interviews to portray a person’s life to the public without being subject to monetary awards for violating that person’s dignity. Given its constitutional status, a person’s human dignity could not be at the mercy of the history of codification. What mattered was not so much what the past had achieved as what human dignity now required. Moreover, the point of this purposive approach is to realize human dignity within the social community to which the constitutional norm applies. As observed above,105 the social community is central to the idea of human dignity. When social community is part of its meaning, human dignity is understood as governing the real world of the individuals whom the constitution protects, and as being attuned to its specific conditions and to the social and material forces that affect it.106 One aspect of this is the self-​referential idea that the embedding of human dignity in the constitution has transformed the world to which it applies: the constitutional imperative to see that world through the lens of human dignity itself reflects and consolidates the ‘changed sense of right’107 that is to inform the analysis of legal relations. Another aspect is that the law conceives of human dignity in a way that repudiates the abuses of it (such as invasions of privacy) that the community has historically perpetuated or experienced. Another aspect is that the constitutionalization of human dignity within social community prompts a realistic assessment of the circumstances of a transaction in accordance with the parties’ reciprocal freedom and with their membership in what the constitution declares to be a Sozialstaat.108 Yet another aspect is that in its application human dignity is sensitive to, and asserts itself over, the material conditions of technology and mass communication that threaten to undermine it. And yet another aspect is that the elucidation of human dignity is a dynamic process that takes account of and keeps pace with social change, so that the jurisprudence of human dignity is never frozen.

105 See nn 15 and 16 above. 106 D Grimm, ‘The Basic Law at 60: Identity and Change’ (2010) 11 German LJ 33, 44. 107 See n 103 above. 108 In the Bürgschaft case, BVerfGE 89. 214 (1993), the relatively impecunious twenty-​one year old daughter of a real estate developer guaranteed a bank loan made by her father. The Federal Court of Justice upheld the contract on the ground that the bank owed the daughter no duty to inform; she was a person of full age who, even if she lacked business experience, would know the consequences of signing a guarantee. The Constitutional Court held that this decision violated her right under GG 2, 1 (guaranteeing the free development of the personality) and GG 20, 1 (declaring Germany a social state). The guarantee of private autonomy under GG 2, 1, which protects contract as medium of self-​determination, extends to both parties, so that both should be able freely to decide to conclude the contract. In this case, the daughter, in an economically inferior position and subject to the pressure of her father, entered into an unusually burdensome contract from which she derived no benefit. On the ensuing reconsideration by the civil court, the guarantee was struck down: BGH NJW 1994, 1341. For an extended discussion of the case see Cherednychenko (n 51) 234–​47. From the standpoint of English law, the German Constitutional Court played the role of a court of equity; see generally S Waddams, Sanctity of Contracts in a Secular Age: Equity, Fairness and Enrichment (CUP 2019).

Horizontality: Presuppositions and Functions  147 In the Princess Soraya case the Constitutional Court expressed these interconnected ideas of purposiveness, dynamism and the relevance of social community by stating that: The explication of a legal norm cannot always continue in the long run with the meaning attributed to it at its origin. One must take into consideration what reasonable function it could have at the time of its application. The norm constantly stands in the context of the social relations and the socio-​political ideas on which it is supposed to act; its content can and must change with them as needed. This is especially true when, in the interval between the origin and the application of a law, living conditions and legal ideas have changed as profoundly as in this century. A judge cannot evade a possible conflict between the norm and the substantive concepts of justice of a transformed society by referring to the unchanged wording of the law; he is compelled to treat the legal norms more freely if he is not to fail in his task of speaking ‘right’.109

Despite its strikingly innovative approach, the Princess Soraya case stays within the mold of private law. It is recognizably a torts case, in which the defendant is assessed damages for an injury to the plaintiff ’s right. Indeed, the point of the judgment was to regularize the right to personality: it gave the normal range of tort remedies to a private law right that was constructed out of constitutional material but based on an idea that already had a limited presence in the civil law. In language reminiscent of (but going beyond) Aristotle’s description of equity, the Constitutional Court itself described its judgment variously as a ‘corrective of the written law’ and as ‘not opposing but fulfilling and continuing the written law’.110 So far as the written law is concerned, the innovative dimension was not imported into the law from the outside, but ‘has its source within the constitutionalized legal order as a meaningful whole’.111 Given that the Civil Code already allowed damages for intangible harm consequential on physical injury, the civil courts had ‘merely further extended by systemically immanent means the basic ideas of the legal order as shaped by the constitution’.112 109 BVerfGE 34, 269, C, IV, 2. The words ‘speaking right’ (‘Recht zu sprechen’) are a play on the German term for case law (Rechtsprechung). 110 ibid IV, 1; IV, 3. Compare Aristotle Nicomachean Ethics V, 14, 1137b, 11, 26. The Court is going beyond Aristotle in this sense: Aristotle was dealing with correcting a deficiency inherent in the inevitable generality of the language of the positive law. This corrective is required because of particular unexpected circumstances for which the legislated language was not apt. Aristotle then appeals to what the legislator would have intended for these circumstances. In contrast, in Princess Soraya, the legislator’s intent was indisputable and the complaint fell squarely within that intent. What made the law’s language inapt were the contemporary social circumstances, the intervening history, and especially the advent of a new constitutional order that responded to that history. The Court is appealing not to what the legislator would have intended, but to what the law must be if it is to be consonant with the principles now present in the legal order as a whole. 111 ibid IV, 1. 112 ibid IV, 3.

148  Ernest J. Weinrib The judgment thus conforms to both of the presuppositions of horizontality set out earlier in this chapter. First, it treats the legal system as a unity that systemically relates the constitution and private law. Second, it locates the commonality of private law and the constitution in the idea of human dignity, which can then influence private law because it is immanent in the legal order as a whole. Thus, the innovative and the conservative dimensions of horizontality combine to create a conception of liability in which human dignity is not limited by the antecedent positive law, but rather infuses it with its distinctive and fundamental value.

7

Horizontality: Scope and Operation 1.  The Distinction between Scope and Operation Of the three functions of horizontality discussed in the preceding chapter, the dignity function gives the strongest role to human dignity in the law’s construction of the reciprocal freedom of interacting private parties. It differs from the development function, which restricts the effect of the constitutional markers of human dignity to merely incremental changes in the private law. From the standpoint of the dignity function, restricting horizontality to an incremental effect is arbitrary. Human dignity (and its Kantian analogue, the innate right) is the law’s fundamental normative idea; in its application to private law, it should be allowed its full scope, unencumbered by the historical contingencies of the law’s past development. But what is its full scope? Given the inappropriateness under the dignity function of restricting horizontality to its incremental effect on private law, is horizontality subject to any other kinds of restriction? And for private law controversies properly within the scope of horizontality, how does horizontality operate? In particular, the operation of horizontality often involves the reconciliation of constitutional rights through a process of ‘balancing’. What is this process, and what is its relationship to the constitutional principle that an infringement of a constitutional right can be justified if it is a suitable, necessary and proportional way of achieving a legitimate or pressing public purpose? These are the issues to which I now turn. They go to the root of horizontality by setting the terms on which private law interacts with constitutional law. The central distinction in this chapter is between the scope and the operation of horizontality. Determining the scope of horizontality requires delineating the conceptual boundaries beyond which constitutional rights cannot be applicable to private law. The operation of horizontality, and particularly the process of balancing the rights invoked by the two parties, takes place only with respect to a controversy between the parties that is within horizontality’s scope. Accordingly, the invocation of horizontality can be defeated in two different ways: by showing either that the controversy is not within horizontality’s scope, or that the operation of horizontality in the circumstances of a particular case does not require ascribing priority to the constitutional right that has been invoked. Put another way, balancing may be crucial to the operation of horizontality, but it plays no role in determining the scope of horizontality.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0007

150  Ernest J. Weinrib The distinction between the scope and the operation of horizontality echoes the distinction between scope and limits that is familiar from constitutional discourse about specific rights. The scope of a right refers to the domain to which the protection afforded by a right applies; one asks, for instance, whether a particular practice or belief falls within the scope of freedom of religion,1 or whether a specific set of utterances falls within the scope of freedom of expression.2 Only once the impugned matter is regarded as falling within the scope of the right do the limits on the exercise of that right become pertinent. These limits do not repudiate the applicability of the right; rather, they refer to the notion of proportionality in order to qualify the right’s exercise due to the effect of other specific rights or of the entire system of rights. Limits on rights go to the way in which rights operate within the constitutional order. Proportionality plays no role in defining the right’s scope, but is decisive for determining whether the act that limits the exercise of the right is constitutionally justified. This chapter deploys a similarly structured distinction between scope and operation not in relation to specific rights but more generally in relation to the phenomenon of horizontality as a whole. The scope of horizontality is, accordingly, a threshold issue. The question it addresses is whether, in a given situation, a constitutional right is implicated in the private law controversy at all. This question is conceptually prior to the issue of horizontality’s operation, that is, of how private law should respond to the presence of an applicable constitutional right. Despite the extensive literature on horizontality, however, the issue of horizontality’s scope has received little systematic attention.3 The present chapter aims to address this academic lacuna. Because horizontality marks the interface between private law and constitutional law, the scope of horizontality becomes visible at the seam between these two varieties of legal ordering. More specifically, the boundary that delineates the point beyond which constitutional rights are inapplicable to private law emerges either where the constitutional dimension ceases to be compatible with the structure of the private law relationship, or where the incorporation of rights into private law ceases to be consistent with their constitutional role. The first of these possibilities (incompatibility with the structure of the private law relationship) occurs when the constitution specifies rights that are distributive in nature, as in the case of social and economic rights. Constitutional rights that are distributive in nature impose correlative obligations on the state to exercise its legislative and governmental power for their immediate or progressive realization. Taken on their

1 Ktunaxa Nation v British Columbia (Forests, Lands and Natural Resource Operations) [2017] SCR 386. 2 R v Keegstra [1990] 3 SCR 697. 3 Notable in the horizontality context are the remarks about the Abgrenzungsmodell (the delimitation model) in F Maultzsch, ‘Die Konstitutionalizierung des Privatrecht als Entwicklungsprozess: Vergleichende Betrachtungen zum deutschen und amerikanishen Recht’ (2012) JZ 1040, 1046–​50.

Horizontality: Scope and Operation  151 own they do not link the right-​holder to a specific private party in accordance with corrective justice; they therefore seem to be beyond the scope of horizontal application. Despite this general inapplicability, one may ask whether under particular circumstances this kind of constitutional right nonetheless becomes relevant to a private law relationship. These matters are discussed in section 2. The second of these possibilities (inconsistency with the right’s constitutional role) occurs when the supposed exercise of a constitutional right fails to fulfil the right’s constitutional purpose or negates the rights of others. These seemingly oracular contentions are explicated below in sections 3 and 4. Section 5 then moves from the scope to the operation of horizontality. The focus is on the balancing that is standardly taken to characterize how horizontality functions. In particular, the section asks how this process of balancing works, and how it differs from the version of balancing that plays a prominent role in the justifications for limiting rights in the public law context. As in the preceding chapter, my focus is not on the working of horizontality in any specific regime of positive law, but on thematic aspects engaged by the very phenomenon of horizontality. In particular, this chapter describes how the scope and operation of horizontality make for a coherent conception of its union of private law and constitutional law. Of course, as in any Kantian approach to law, regimes of positive law provide highly articulated illustrations—​'excellent guides’ as Kant called them4—​of how the idea of rightful relations might be and has been actualized. The theoretical objective, however, is not to explain the details of a particular legal system, but to exhibit the underlying ideas implicit in a system of rights. The legal doctrine in given legal systems might express these ideas with varying degrees of adequacy.

2.  Social and Economic Rights A common, although not ubiquitous, feature of contemporary constitutions (or of their jurisprudence) is the inclusion of social and economic rights. These rights create an obligation on the state to act for the realization of each person’s entitlement to such matters as food, water, housing, education, and health care.5 The constitutional requirement that the state ensure (or work towards ensuring) the availability of the subject matter of these rights reflects the recognition that human dignity—​and the reciprocal freedom to which human dignity is essential—​can otherwise not be secured. Viewed in Kantian terms, the subject matter of the social and economic rights are aspects of a person’s innate right; they are necessary to truly being one’s own master; that is, not subject to another’s arbitrary will.6

4

I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 26 [6:230]. KG Young, Constituting Economic and Social Rights (OUP 2012). 6 See ch 5, s 2. 5

152  Ernest J. Weinrib The idea that the social and economic rights enumerated in the constitution reflect human dignity creates an apparent dilemma for horizontality. On the one hand, under the dignity function private law should have the same regard for these constitutionalized aspects of dignity as it has for others. If the justification for making constitutional rights applicable to private law is that the law should require interacting parties to respect the constitutional expressions of each other’s dignity, then another’s social and economic rights should be no less pertinent to the parties’ relationship than privacy or freedom of speech. On the other hand, the horizontal application of social and economic rights is problematic. Social and economic rights have a feature that, although present in other rights,7 is especially salient in them: they subject the state not merely to a negative obligation of non-​interference, but also to a positive obligation to work for their actualization. From the standpoint of private law, the placing of private persons under a parallel positive obligation would be inconsistent with the fundamental private law distinction between non-​feasance and misfeasance,8 for it would lead to liability for failing to benefit another. Moreover, whereas a state necessarily has the leeway to provide for the satisfaction of these rights in different ways and to different extents,9 the private law counterpart to the state’s obligation can serve as a basis of liability only if it is unequivocal. Similarly, for some of these rights the constitution may explicitly permit their merely progressive realization within the state’s available resources;10 in contrast, private law makes defendants fully and immediately subject to the obligations that are correlative to its rights. Thus, treating social or economic rights as horizontally applicable might, oddly, allow these rights to operate more stringently for private law than for the state action that is their primary objective, while at the same time undermining the standard private law denial of liability for non-​feasance. All these problems arise because the constitutional mandating of legislation that advances distributive justice is categorically different from, and ill-​suited to, the achievement of corrective justice through private law adjudication. The most revelatory approach to this dilemma is found in the South African jurisprudence. The South African constitution combines extensive guarantees of socio-​economic rights with the explicit authorization of horizontal application. These two features reflect the transformative aspirations of a constitution committed to replacing past oppression and inequality with a system of rights based on human dignity. The motivation for these features was that the basic law of the new South Africa needed to correct the injustices of apartheid both by constitutionally 7 S Fredman, Comparative Human Rights Law (OUP 2018) 67. 8 P Benson, ‘Misfeasance as an Organizing Normative Idea in Private Law’ (2010) 60 UTLJ 731. 9 See ch 5, s 3. 10 For example, the Constitution of South Africa, ss 26 and 27 require the state to ‘take reasonable legislative and other measures, within its available resources, to achieve the progressive realization’ of the right to have access to adequate housing, healthcare, food, water, and social security.

Horizontality: Scope and Operation  153 affirming a wide range of social and economic rights and by rehabilitating the legal relationships between individuals through the horizontal application of constitutional rights. So far as the named socio-​economic rights are concerned,11 the constitution enumerates rights to housing, health care, food, water, social security, and education. The state must progressively realize these rights within its available resources through reasonable legislative measures.12 The constitution also lists without qualification the right to a basic education13 and the rights of children to nutrition, shelter, basic health care, and basic social services.14 Moreover, the constitution of South Africa includes every basis for horizontal application suggested in the jurisprudence of contemporary liberal democracies.15 It stipulates that the state must ‘respect, protect, promote and fulfil’ the enumerated rights;16 that the Bill of Rights is binding on the judiciary and on all organs of state, as well as on the legislature and the executive;17 that when developing the common law, courts must promote the spirit, purport and object of the Bill of Rights;18 and, most importantly, that a provision of the Bill of Rights binds ‘a natural and juristic person if, and to the extent that, it is applicable, taking into account the nature of the right and the nature of any duty imposed by the right’.19 Through these provisions, some of which were introduced as late as the transition from the interim to the final version of the constitution, the framers deliberately and unequivocally entrenched the direct applicability of constitutional rights to private law relationships. In particular, the framers’ revisions amounted to rejecting the development function of horizontality found in Canada and espoused under the interim constitution,20 in favour of the more robust dignity function. When one considers these two features of the South African constitution—​the socio-​economic rights and horizontality—​together, what is the relationship between them? The Constitutional Court’s leading case on the horizontal application of social rights is Governing Body of the Juma Musjid Primary School v Essay NO.21 This case deals with the right of children to a basic education. As the Court noted, the constitutional text makes this right immediately realizable: the right is not one that needs to be realized only progressively within the state’s available resources 11 For a meticulous and thoughtful treatment see S Liebenberg, Socio-​Economic Rights: Adjudication Under a Transformative Constitution (Juta 2010). 12 Constitution of South Africa, ss 26, 27, 29(b). 13 ibid s 29(a). 14 ibid s 28. 15 See N Friedman, ‘The South African Common Law and the Constitution: Revisiting Horizontality’ [2014] SAJHR 63 on how these provisions should be read together. 16 Constitution of South Africa, s 7(2). 17 ibid s 8(1). 18 ibid s 39(2). 19 ibid s 8(2). 20 Du Plessis v De Klerk 1996 (3) SA 850. 21 [2011] ZACC 13.

154  Ernest J. Weinrib through reasonable legislative measures.22 The urgency about this right reflects the recognition that in the modern world basic education is indispensable for interaction among persons on the footing of their intrinsic human dignity. The juridical significance of the basic education of children becomes evident when human dignity is viewed as the contemporary designation for what Kant termed a person’s innate right. It will be recalled that the innate right, defined as independence from constraint by another’s choice, is the right that belongs to every human being not through an act of acquisition but by virtue of being born.23 The exercise of this right consists in the exercise of a power to act for one’s own purposes, subject only to the co-​existence of this exercise with the similar exercise of that power by others. Therein lies the equal reciprocal freedom on which the Kantian conception of law insists. When children are born into the world, their exercise of the power to act for their own purposes consistently with the innate right of everyone else is not immediately available to them but has to be developed. Only through this process of development can infants progress from their initial state of complete dependence on others and complete absorption in their own needs to the condition in which they can assert their entitlement freely to set and pursue their own ends and can observe their duty not to interfere with the equal freedom of others. Their development of this power is itself an entitlement under their innate right; otherwise, their innate right would remain incapable of proper exercise through no fault of their own. Correlative to this entitlement is a duty incumbent on the parents who brought the child into the world. These parents would wrong the child if, having produced on their own initiative and without the child’s consent a being endowed innately with this right, they then failed to take the steps necessary to allow the child to exercise the right that was consequent on their procreative action. Providing an education to one’s children is, accordingly, not the bestowal of a benefit on them as a matter of grace. Rather, in the traditional terminology of private law, failure on the part of parents to follow through on the duty that their own conduct has created would be a misfeasance violative of the child’s right, no different in its structure from failing to press the brakes when driving a vehicle.24 In other words, even in the state of nature the child has a right to be educated into the proper way of participating in interaction with fellow right-​holders. From the standpoint of a person’s innate right the child is entitled to be developed ‘pragmatically and morally’:25 the pragmatic development aims at cultivating the ability to make one’s way in life in accordance with one’s own purposes; and the moral development aims at inculcating respect for the rights of others. Both the pragmatic and the moral



22

ibid para 37. See ch 2, s 4. 24 Kelly v Metropolitan Ry [1895] 1 QB 944. 25 Kant (n 4) 71 [6:281]. 23

Horizontality: Scope and Operation  155 development of the child are necessary if persons are to interact on terms of their reciprocal independence, without either constraining others or being constrained by them. Because for Kant the civil condition realizes and makes effective the norms of the state of nature, this conception of the relation between education and rightful interaction continues into the civil condition. For a person’s life as a whole, education has several functions (including education towards physical well-​being and education towards morality), but the function that reflects the juridical significance of the civil condition is education towards reciprocal freedom. This requires the formation of ‘a freely acting human being who can support itself and be a member of society;’ such a being has been moulded into a citizen by having learned ‘not only how to direct civil society for his purposes, but also how to fit with civil society’.26 In its provisions on the right to an education, the South African constitution in several ways adapts this Kantian idea to the specific civil condition that the constitution governs. First, the constitution implicitly recognizes that under the conditions of modern life, the right of every person to an education cannot be effectively satisfied if its correlative obligation is incumbent on parents alone. The complexity of the contemporary world into which children are to be educated, the diversity of abilities and resources among students and among parents, and the sheer size of the student population requires the state to offer and certify a range of educational experiences and services if education is to be universal in accordance with each student’s right. Moreover, the state’s role is normatively and not only practically necessary, because education that prepares students to interact with their fellow citizens ‘pragmatically and morally’ is integral to the evolution of the reciprocal freedom for which the state exists. Secondly, the South African constitution, cognizant of the burden that the universal right to education imposes on the state’s resources, distinguishes between the right to a basic education, which the state must immediately realize, and the right to further education, which the state must make progressively accessible and available through reasonable measures. In this way, the constitution gives priority to the most formative, essential and pervasive kind of education. Thirdly, under the South African constitution the right to a basic education is not confined to children, but also expressly extends to adult basic education. This is readily understandable: because the constitutional right to a basic education obliges the state rather than the parent to realize it, the focus is not on the status of childhood but on the reciprocal freedom to which basic education contributes. On the theoretical plane, the difference between the entitlement to education in the state of nature and in the civil condition brings us back to the dilemma that 26 I Kant, ‘Lectures on Pedagogy’ in G Zöller and RB Louden (eds), The Cambridge Edition of the Works of Immanuel Kant: Anthropology, History, and Education (CUP 2007) 448 [9:455].

156  Ernest J. Weinrib social and economic rights pose for horizontality. As already observed, the idea that in the state of nature the parent is the obligor of the child’s right to be educated allows the right’s infringement to be understood as a private law misfeasance that consists in not following through on the normative implication of the procreative act. This notion of misfeasance no longer obtains once the state becomes the right’s obligor in the civil condition. The state is obligated to honour the right not because it must follow through on some previous act but simply because (in the Kantian conception, at least) the state, as the repository of public right, is under an obligation to exercise its public authority to actualize freedom.27 From the standpoint of private law, the provision of education by the state is a benefit that occurs within a context of non-​feasance, from which no private law obligation can emerge. From the standpoint of public law, however, education is the object of a right that the state is duty bound to honour out of respect for the human dignity of the right-​ holder. The challenge for horizontality in applying to private law the constitutional right to a basic education is to reconcile these two standpoints. The Juma Musjid case provides a response to this challenge. The Court had to decide whether the trustees of a mosque could evict a primary school that was located on mosque premises. The school was established as a government-​aided public school that would be administered subject to a written agreement between the trustees and the state educational authority. However, the efforts to conclude this agreement failed. Nor did the state authority reimburse the trust, as it alleged was expected, for its rental and maintenance expenses. When the trust attempted to evict the school before the end of the school term, the parents and guardians of the children enrolled at the school objected that the eviction would be an infringement of the children’s constitutional right to a basic education. Under the constitution the children’s right to a basic education was binding on the trust as a juristic person, even though the trust was not considered to be performing a public function. In explicating the trust’s duty under the constitution, the Court made two points. First, the primary positive obligation with respect to the children’s right to a basic education rested on the state authorities, not on the trust. This was an obligation that the state had failed to fulfil. The trust, in contrast, was under no positive obligation either to provide a basic education or to make its property available for use as a school. But second, this conclusion did not mean that the trust was under no constitutional duty at all to the children with respect to their right to a basic education. Because that was a right that the children were presently enjoying on the trust’s property, the trust was under a constitutional obligation not to impair its exercise. This was a negative obligation only. With regard to the purpose of making constitutional rights binding on every natural and juristic person, the Court said:



27

See ch 6, s 6.

Horizontality: Scope and Operation  157 [T]‌he purpose of section 8(2) of the Constitution is not to obstruct private autonomy or to impose on a private party the duties of the state. It is rather to require private parties not to interfere with or diminish the enjoyment of a right . . . [T]he primary obligation in respect of the learners’ right to a basic education is that of the state. The Trust’s obligation is secondary and . . . arises only from its willingness to allow the property to be used as a public school . . . It did not give up its rights of ownership of the property. At most, the Trust’s constitutional obligation, once it allowed the school to be conducted on its property, was to minimise the potential impairment of the learners’ right to a basic education.28

The result was that, although the trustees had acted reasonably in seeking to evict the school in the light of the state’s delinquencies regarding the rental arrangements, no eviction order could be issued until the state, in accordance with its positive obligation to fulfil the children’s right to a basic education, secured alternative placement for them. Subsequently, the eviction order was granted when the state education authorities arranged satisfactory placement for the following term. Thus, the dilemma that socio-​economic rights pose for horizontality is resolved by differentiating between the significance of the right for distributive justice on the one hand and for corrective justice on the other. The entitlement to basic education, considered on its own as an aspect of distributive justice that the Constitution guarantees, ties the beneficiaries of the right only to the state and not to a private party. This fully accords with the notion that is standard for private law, that no duty arises in situations of non-​feasance, that is, when one has not through one’s conduct participated in infringing a right. That notion itself reflects a fundamental dimension of a person’s innate right, that one is responsible only for what one has done and not for who one is. Conversely, the state stands in a different relationship to the right to a basic education than does a private person, because the state is bound to bring itself into ever greater conformity with its animating idea of reciprocal freedom. Through its enumeration of the various rights, the Constitution becomes the authoritative legal instrument that formulates how that animating idea is to be realized in a specific civil condition. Accordingly, in Juma Musjid the positive obligation correlative to the right to a basic education falls on the state and not on the trust. The situation changes, however, once the right-​holder is enjoying the right in question. The right then cannot legitimately be impaired by a private party, because it pertains to the relationship between the parties as a matter of corrective justice. If the trust exercised its property right—​a right also guaranteed to it under the South African constitution29—​to expel the school and its students, it would be an act of misfeasance that infringed the students’ constitutional right to education.

28 29

[2011] ZACC 13, paras 58, 62. Constitution of South Africa, s 25.

158  Ernest J. Weinrib Misfeasance has been defined as ‘any act or omission by one that impairs, damages, or otherwise affects something that comes under another’s exclusive rights vis-​à-​ vis the first’.30 The phenomenon of horizontality, especially in the strong form presented by the South African constitution that makes constitutional rights directly binding on natural and juristic persons, brings constitutional rights into the ensemble of exclusive rights that one person has vis-​à-​vis another. The Constitutional Court could, therefore, not allow the trust to impair a right that the children were already enjoying. It may seem odd that in the Juma Musjid case the constitutional right has the chameleon-​like quality of shifting its significance from distributive to corrective justice, and of shifting the conduct that affects it from non-​feasance to misfeasance. How is it possible for the trust to be under no obligation to establish a school but nonetheless to be obligated not to expel the school once it has been established? From the standpoint of private law, however, this poses no special difficulty. The same pattern of thought is exemplified in well-​established doctrine quite distant from the horizontality context. In the well-​known tort case of Depue v Flatau,31 for instance, the plaintiff became sick while visiting the defendant’s house for business purposes. Even though it was the height of winter in rural Minnesota, the plaintiff was refused permission to stay overnight at the defendant’s house. Instead, he was carried out to his cutter, had the reins wrapped around him, and had his team started in the direction of his home. He suffered injury when he fell from the cutter about a mile from the defendant’s house and nearly froze to death during the night. In holding the defendant liable, the court dismissed the argument that the defendant had committed no wrong by refusing to accommodate the plaintiff. Although one is under no obligation to provide assistance to a person in distress—​ ‘suffering humanity has no legal complaint against those who pass by on the other side’32—​the defendant here had inflicted an injury by exposing the plaintiff to the Minnesota winter. Because of the plaintiff ’s condition, it should have been obvious that the action that the defendant took would cause him injury. Under these circumstances, the defendant was under a duty ‘to exercise care commensurate with the situation in which he finds himself, and with which he is confronted, to avoid such danger’.33 Despite the obvious factual differences between the Depue and Juma Musjid cases, in principle they are identical. Neither the defendant in Depue nor the trust in Juma Musjid was under an obligation to take the plaintiff or the children into their space. Moreover, once the plaintiff or the children had been allowed into the space the owner could disinvite or expel them at will; that is simply a consequence



30

P Benson, Justice in Transactions: A Theory of Contract Law (Belknap Press 2019) 7. 100 Minn 299 (SC 1907). 32 ibid 303. 33 ibid. 31

Horizontality: Scope and Operation  159 of the owner’s absolute dominion over what is owned. But they could not be disinvited or expelled in a way that would impair the existing rights that they were enjoying (the right to bodily security in Depue, and the right to a basic education in Juma Musjid). Although neither the defendant in Depue nor the trust in Juma Musjid were under an antecedent obligation to assist the plaintiff or the children in the exercise of their rights, they could not injure those rights through their ownership of the location where the rights were being exercised; doing so would cross the line from non-​feasance to misfeasance. Although in Juma Musjid the Court refused to authorize an immediate eviction, it pointed out that the trust’s permitting a school on its property was not a surrender of its right of ownership.34 Rather, the Constitution counterposes to the traditional ownership rights a new set of constitutional rights. As a result, courts must now ‘balance and reconcile the opposed claims in as just a manner as possible, taking account of all the interests involved and the specific factors relevant in each particular case’.35 A significant consideration was that the eviction was postponed in view of the imminent end of the school year; the Court insisted that the state take steps to accommodate the students’ schooling for the new year.36 Presumably the balancing was implicitly in terms of proportionality, with the postponement being suitable and necessary for maintaining the children’s right to basic education, and with its short duration exhibiting proportionality between the counterposed rights of the owners and the students. Again, although the effect of constitutional rights on the owner’s legal position is new, the process of reasoning is not; the duty of the owner in the Depue case to ‘exercise care commensurate with the situation’ by allowing the plaintiff to stay overnight can also be read as proportionally balancing the prospect of serious physical harm to the plaintiff against the transient incursion into the defendant’s right as owner.37 Thus, the treatment of the right to a basic education in the Juma Musjid case illustrates both of the themes of this chapter. The first theme concerns the scope of horizontality. The application of the constitutional right is not within horizontality’s scope if it is inconsistent with the structure of private law relationships. Insofar as they impose positive obligations on the state in order to advance distributive justice through the exercise of its public authority, social and economic rights do not connect the beneficiaries of the right to any particular private party and, therefore, are not amenable to consideration within private law. Accordingly, the distributive dimension of social and economic rights (along with the attendant irrelevance to it of the notion of non-​feasance) is beyond the scope of horizontal application. Balancing has no role to play in such determinations of horizontality’s scope. 34 See n 28 above. 35 [2011] ZACC 13, para 70, citing Sachs J in Port Elizabeth Municipality v Various Occupiers [2004] ZACC 7. 36 ibid paras 74–​78. 37 Compare ch 4, s 4c.

160  Ernest J. Weinrib The second theme concerns the operation of horizontality. As the litigation over the school eviction shows, the existing enjoyment of a socio-​economic right can link private parties and thus bring the right within horizontality’s scope. As a consequence, a private party may come under an obligation as a matter of corrective justice not to impair the exercise by others of their constitutional rights. In the Juma Musjid case, the fact that the children, in the exercise of their right to a basic education, are being schooled on the trust’s property establishes the normative connection between the children and the trust from which the obligation flows. Then the issue arises of the operation of horizontality: how is this obligation not to interfere with the children’s right to a basic education to be reconciled with the other entitlements (for example, the ownership of property) present in the relationship between the parties? The balancing of the trust’s ownership against the limited duration of the school’s continuing presence on the property brings the disparate components of the parties’ relationship into alignment with one another.

3.  The Illusory Exercise of Constitutional Rights So far in this chapter I have been delineating the scope of horizontality from the side of private law. I now move to the side of constitutional law. My contention in this section is that some purported exercises of a constitutional right are beyond the scope of horizontality because, despite appearances, they are illusory rather than actual. When no constitutional right is really being exercised, the issue of horizontal application does not arise. The notion of an illusory exercise of a right follows from the difference between what a right is about and what falls within the right’s scope. Eminent German constitutionalists have distinguished between the right’s subject area and the right’s protective content.38 The former points in a general and abstract way to the ‘area of material and life’ (Sach-​und Lebensbereich) to which the fundamental right refers, for example, religion or expression. For that area the right enshrines a principle of freedom that renders certain conduct potentially eligible for constitutional protection. Whether the specific conduct in issue merits this protect, however, is a separate issue. The identification of a subject matter is a descriptive exercise that indicates whether the right might be engaged at all. It does not in itself perform the normative function of specifying the nature and extent of the constitutional guarantee. That function requires working out the right’s protective content (Gewährleistungsgehalt), with respect to its particular ‘area of material and life’. The scope of the right marks the boundary of the right’s protective content. 38 E-​ W Böckenförde, ‘Schutzbereich, Eingriff, Verfassungsimmanente Schranken: Zur Kritik gegenwärtiger Grundrechtsdogmatik’ (2003) 42 Staat 165, 174; W Hoffmann-​ Riem, ‘Grundrechtsanwendung unter Rationalitätsanspruch’ (2004) 43 Staat 203, 215.

Horizontality: Scope and Operation  161 The illusory exercise of a constitutional right occurs when conduct that falls within the right’s subject area is presumed without normative analysis to warrant constitutional protection. For example, one might deem conduct identified as religion or as expression to be thereby entitled to constitutional protection, subject only to the further possibility of justifiable limitation in accordance with the strictures of necessity, suitability and proportionality. The illusion arises from treating the identification of the subject area as dispositive without elucidating the right’s protective content. Because in a regime of rights, constitutions are purposively oriented to the freedom of all, the nature and the extent of constitutional protection depend on the right’s purpose in giving expression to the specific aspect of freedom that the right designates. Whether, for instance, religion or expression is protected in a given controversy requires examination of whether the specific adverse effect on religion or expression derogates from the freedom for the sake of which religious or expressive rights are given constitutional status. If it does not, the claim that this effect is inconsistent with the exercise of one’s right is illusory, because it is based merely on the right’s subject area rather than on its protective principle. Such a conclusion does not involve justifiably limiting the right. Rather, no right is really being exercised, so that no right is being infringed, whether justifiably or not. So far as horizontality is concerned, an illusory exercise of a constitutional right has no more relevance for private law controversies than it should have for constitutional ones. In this sense, illusory exercises of constitutional rights are beyond the scope of horizontal application. The exercise of a right can be illusory in two ways. In one of these, the right is considered on its own; in the other, the right is considered as a component of a system of rights. The following two German cases illustrate each of these in turn. In the Blinkfüer case,39 two publishing houses in West Germany notified magazine and newspaper dealers that, in the wake of the building of the Berlin wall, they should boycott newspapers that printed radio and television programming from the Eastern zone. Such programming, they claimed, was a vehicle of East German propaganda. The publishing houses reinforced this call for a boycott by threatening to reconsider their business relationships with dealers who carried those newspapers. So economically dominant were the publishing houses that loss of their business would have been ruinous for the dealers. The plaintiff, who produced a small left-​wing magazine that carried programming information from East Germany, sued the publishing houses on the ground that their call for a boycott violated the civil code provision against inflicting harm contrary to public morals. The publishing houses pleaded that they were merely exercising their



39

25 BVerfGE 256 (1969).

162  Ernest J. Weinrib constitutional right to free expression, as in the path-​breaking case on horizontality mentioned at the beginning of the previous chapter.40 In finding for the plaintiffs, the Constitutional Court held that in this instance the call for a boycott was not constitutionally protected. The purpose of the right of free expression and of freedom of the press was to facilitate the free formation of opinion through intellectual disagreements on questions of public concern. In accordance with this purpose, it would have been constitutionally permissible for the publishers to appeal to readers to boycott publications like the plaintiff ’s. That, however, is not what the defendants did. Instead, they acted beyond the purposes of the constitutional right to free expression and a free press by directing the call for a boycott to dealers who were economically dependent on them. Given its coercive nature, this was an inappropriate means of implementing their call for a boycott. And given that the plaintiff here could also claim the right of press freedom for itself, the defendants’ action was not so much an affirmation of press freedom as a suppression of it. In this case, the Court held that the defendants’ conduct did not fall within the right that they asserted. The defendants’ call for a boycott—​a communication about an issue of public importance—​certainly fell within ‘the area of material and life’ to which the right to free expression referred. But consideration of the purpose of the right showed that the defendants’ conduct was not protected by it, because they threatened those who would not heed their call with a potentially disastrous consequence. The economic vulnerability of the dealers meant that they could not freely choose their side in what should have been, as the Court put it, ‘an intellectual clash of opinion’. Examination of the right from the standpoint of its purpose revealed that the defendants’ supposed exercise of that right was illusory. Whereas this example focuses on the right itself, the next example focuses (or so I shall propose) on the system of rights of which the asserted right is a part. In the case of the Sprayer of Zurich,41 the Swiss artist Harald Naegeli, dissatisfied with the drabness of Zurich’s buildings, spray-​painted graffiti on their walls. When prosecuted for this damage to property, he fled to Germany, where the legal question became whether the doctrine of double criminality allowed him to be extradited. It was contended on his behalf that under German law he would be protected by the constitution’s guarantee of artistic freedom, a guarantee so strong that the German constitution, in reaction to the Nazi destruction of artistic freedom, subjects it to no explicit limitation. The Constitutional Court rejected this argument, stating that although the guarantee of artistic freedom lacked express limitation, ‘from the outset its scope does not extend to the unauthorized use or impairment of another’s property for purposes of artistic development’.42 The Constitutional Court did not

40

7 BVerfGE 198 (Lüth 1958). BVerfG (Vorprüfungsausschuß) v 19. 3 (1984), NJW 1984, 1293. 42 ibid 1294. 41

Horizontality: Scope and Operation  163 consider whether the criminal provision protecting property rights was a justified infringement of the right to artistic expression. Instead, it affirmed that the constitutional guarantee of artistic freedom did not entitle the artist to flout the property right of others for the sake of his artistic activity.43 To put the issue in terms of the distinction introduced earlier in this section, the Court’s view seemed to be that although as a descriptive matter the production of the graffiti fell within the subject area of the right to artistic freedom, it was as a normative matter excluded ab initio from the right’s protective content. There was no infringement that needed to be justified, because in this situation, where the use of another’s property was beyond the scope of Naegeli’s right to artistic freedom, that right was not really in play at all. It may seem strange that the right to artistic freedom does not protect the artist’s manifestation of his dissatisfaction with his city’s desolate architecture. The case has proved to be more popular academically than judicially. Indeed, the Constitutional Court subsequently voiced doubt about whether the Sprayer of Zurich case was properly handled.44 But the case’s underlying rationale is worth deciphering for its bearing on the scope of horizontality. The assumption behind the Sprayer of Zurich judgment is that freedom of artistic expression involves protection of the right to self-​expression with one’s own means or with the means of others who give permission. In this case, however, the means used by the artist were neither his own nor used with their owners’ authorization. The graffiti as figures merited protection, but usurpation of the property of others did not. To be sure, the figures could fulfil their artistic function only if they were exhibited on the very buildings whose drabness inspired the need for them. That, however, did not give the artist the right unilaterally to conscript the property of others to his artistic purposes. Even if the figures on their own qualified for constitutional protection, the artist could not use someone else’s buildings as his canvas.45 The larger conceptual background for these observations on the case is the contrast between limits and negations of rights. Modern liberal constitutionalism recognizes that rights are not absolute. They can justifiably be limited for the sake of other rights or of the entire system of rights. It is in the nature of such justified limits that they maintain their continuity with what is being limited. In the public law setting, this continuity is reflected in an articulated ensemble of considerations: that the right is the norm, and the limit is the exception that has to be justified; that the party asserting the limit has the onus of proving the justifying elements; that the limit is ineffective unless prescribed by law; that the limit forwards a cogent public purpose within the regime of rights as a whole; that the limit

43 B Rusteberg, Der grundrechtliche Gewährleistungsgehalt: ein veränderte Perspektive auf die Grundrechtsdogmatik durch einer präzise Schutzbereichsbestimmung (Mohr Siebeck 2009) 99. 44 119 BVerfGE 1 ( Esra 2007). 45 I am grateful to Gerard Westland for this way of putting it.

164  Ernest J. Weinrib is rationally connected to its justifying public purpose; that the impairment of the right it limits is as minimal as possible; and that the limiting measure does not impose a disproportionate burden on the right that it limits. All these requirements maintain the continuity between right and limit by systematically honouring the right and affirming its superordinate position even as it is being limited. Indeed, it is only by conforming to these requirements that a limit counts as a limit at all. In other words, a justified limit does not efface the right that it limits. In this respect, justified limits on rights are categorically different from negations of rights. A negation leaves the right with no residue of acknowledgment. It treats the right as a nullity, empties it of its content, and calls into question its very existence as a right.46 Whereas limits can be justified by showing their continuity with the right being limited or their compatibility with the system of rights as a whole, negations are impossible to justify, because treating a right as non-​existent cannot be a feature of a system of rights. Thus, unlike limits, negations are impermissible. This is the case even if a right is negated for the sake of exercising another right. Rights are not isolated normative entities; they are components of a system in which they all contribute to securing everyone’s freedom. Negating one right for the sake of another is inconsistent with every right’s place in this system. Thus, negation of a right for the sake of another right undermines not only the right that it negates but also the right that it affirms, because negation denies the systematic character of all rights as they interrelate in the realization of freedom. Instead of recognizing that all the rights in the system of rights are to be enjoyed by everyone equally, a person who denies another’s right in the act of asserting one’s own is engaged in a contradiction that disguises self-​preference as rightful conduct. Although when viewed in isolation a right might appear to be exercised, that exercise is illusory from the systemic point of view. Constitutional protection is, therefore, in principle unavailable to it. In the Sprayer of Zurich case, the artist negated the property rights of the owners of the buildings on which he sprayed his graffiti. This negation is not a reference to his intentions but an implication of his actions. He treated those buildings simply 46 I take this formulation from the explanation by Koen Lenaerts, President of the Court of Justice of the European Union, of what it means to compromise the essence of a fundamental right in contravention of Article 52(1) of the Charter of Fundamental Rights of the European Union. See K Lenaerts, ‘Limits on Limitations: The Essence of Fundamental Rights in the EU’ (2019) 20 German LJ 779. The notions of negation of a right and of interference with the essence of a right are similar but slightly different. The similarity is that in both of them the infringement of the right is such that no room is left for balancing its adverse effects against its beneficial ones. The difference is that the essence of the right refers to a feature present within the right, whereas negation refers to the implication of a person’s action with respect to that right. Interference with the essence goes to the minimal condition of its being a right; negation of the right goes to what can be implied from the way another treats the right. One might say that the essence of the right is an ontological issue, whereas the negation of a right is a transactional issue. The consequence of this distinction is that negation may not be exposed to some of the criticisms of essence. For an example of these criticisms see G van der Schyff, ‘Cutting to the Core of Conflicting Rights: The Question of Inalienable Cores in Comparative Perspective’ in E Brems (ed), Conflicts between Fundamental Rights (Intersentia 2008) 131.

Horizontality: Scope and Operation  165 as lying open to his artistic purposes, thereby asserting that their external appearance was for him to decide and not for their owners. To be sure, the figures in themselves were the product of the artist’s artistic freedom. Here, however, they cannot be considered solely as figures, as they were painted onto other persons’ buildings in an implicit denial of their rights as owners. On this explanation of the case, the idea that the artistic production negated the property rights of others accounts for the Court’s opinion that the artist’s right of artistic expression stops short of their property. In negating the rights of others, the artist’s activity is not itself the exercise of a constitutional right. It merely gives the illusion of being one.47 The Sprayer of Zurich case was not itself about the application of constitutional norms to private law. However, the same issue could have arisen in a tort action for trespass to property. If my observations about that case are sound, then the idea of negation provides another way of delineating the scope of horizontality from the constitutional side. In this second way, even if the exercise of a right is not illusory considered on its own, its effect on the right of another may make it illusory when considered from the standpoint of the system of rights in which every right participates. The idea of negation has its own boundary. Because negation involves treating the other party’s right as a nullity, it cannot readily be found in a continuing legal relationship. Such a relationship creates a nexus of so many points of legal contact between the parties that an action affecting one aspect of that relationship cannot plausibly be considered a negation of all the entitlements inherent in the relationship as a whole. In this respect, the notion of negation in the horizontality context runs parallel to the related idea of respecting the essence of the right in human rights law. In the latter context, ‘[a]‌measure that limits the exercise of certain aspects of a fundamental right, leaving others untouched, or that only applies in a specific set of circumstances regarding the individual conduct of the person concerned, is not such as to compromise the essence of that fundamental right’.48 Similarly, in the horizontality context an act by one party that limits or proposes 47 For an example of the contrast between a limit and a negation in constitutional law see the Canadian case of B (R) v Children’s Aid Society of Metropolitan Toronto [1995] 1 SCR 315. The issue was the constitutionality of the state’s interference in the religiously based decision of parents not to authorize a blood transfusion for their newly born daughter. The Court upheld the constitutionality of the state’s action, but narrowly split on the appropriate reasoning. Five judges held that the parent’s decision was an exercise of their right to freedom of religion, because that freedom included the right to rear their child in accordance with their religious beliefs; the state’s intervention was, however, a justified limit on that freedom. Four judges held that the prospective impact on the child meant that the parent’s decision was not within the scope of their freedom of religion to begin with. In their view the parents’ decision impinged on the infant’s freedom of conscience, which arguably included the right to live long enough to make one’s own choice about one’s religious beliefs. In their words (at page 437): ‘ “Freedom of religion” should not encompass activity that so categorically negates the “freedom of conscience” of another.’ As this opinion points out, the contest in this case was effectively not between the state and an aggrieved individual, but between the child and her parents as individuals. This makes its relevance to horizontality all the more direct. 48 Lenaerts (n 46) 785.

166  Ernest J. Weinrib to limit the exercise of only a certain aspect of the entitlements that make up the parties’ relationship while leaving the others untouched, or that applies only in a specific set of circumstances within a more extended relationship does not negate the right of the other party. For example, acts (or proposed acts) that, for the sake of one party’s constitutional right, alter the ongoing contractual arrangements for the occupation of living space limit, but do not negate, the other party’s rights. The fact that the act impinges on one aspect of the parties’ relationship while leaving all the other aspects intact means that its claim to constitutional protection is in principle available. Whether the act is permissible in the final analysis is then a matter of horizontality’s operation rather than its scope; this depends on the proportional impact of the competing claims in the specific case. On the one hand, one considers factors that impart special urgency to the assertion of the right: the centrality of the home to one’s personal and family existence, the home’s role as the focus of a person’s ordinary mode of living, and the indispensability of the home as a location for the exercise of the constitutional right in question. On the other hand, one considers the duration, expense, and effect of the act on the rightful interests of the other party. Take the German Parabolantenne case.49 A Turkish guest worker wished to install an antenna in his rented apartment in order to receive programming from his homeland. When the landlord refused permission on the ground that an additional antenna would optically deform his building, the tenant sued, invoking his right under the German constitution ‘to inform himself without hindrance from generally accessible sources’.50 The Constitutional Court held for the tenant. Decisive was the consideration that at the time there was no other way for the tenant to secure this programming. The Court also recognized, however, that the landlord had the power to determine the placement of the antenna so as to minimise visual disturbance, to insist on professional installation, to be indemnified for all expenses and liabilities, and to have the tenant bear the cost of removal on the termination of the lease. Although the landlord’s consent was required, it could not be unreasonably withheld. Under the principle of good faith, which was applicable to tenancy law, the landlord could not deny facilities that, while only negligibly impairing his own right, allowed the tenant to enjoy his constitutional right to information. The outcome thus hinged on balancing the antenna’s role in realizing the tenant’s right to information against its adverse effect on the landlord’s exercise of its proprietary

49 90 BVergGE 27 (1992); see especially the incisive analysis of C-​W Canaris, Grundrechte und Privatrecht; eine Zwischenbilanz (Walter de Gruyter 1999) 60–​62. A similar case from Quebec is Syndicat Northcrest v Amselem [2004] 2 SCR 551 (right of observant Jewish resident to build a succah on the balcony of condominium in contravention of signed declaration of co-​ownership). A case in which the tenant’s claim was unsuccessful despite being within the scope of horizontality is 7 BVerfGE 230 (Wahlplakat 1958). 50 Article 5(1).

Horizontality: Scope and Operation  167 right. In other words, the issue went to horizontality’s operation not its scope. And presumably the reason for this was that the antenna affected only one aspect of the continuing relationship between the parties, so that the proposed action only limited but did not negate the landlord’s right. The idea of negation arises from within constitutional analysis; it is not a consequence of property as such. One might be tempted to think that the significance of the Sprayer of Zurich case is this: because the issue of horizontality’s scope involves delineating the boundaries of horizontality and because property law also delineates boundaries, the scope of horizontality can be defined by the scope of property.51 This conclusion is suspect, because all private law rights delineate boundaries in some sense; property is not unique in this respect. In contrast, the idea of negation does not involve identifying the scope of horizontality with the scope of property. Rather, given the pervasiveness in contemporary constitutional discourse of the idea of limits on rights, the question arises: limits as opposed to what? What is the nature of an infringement that does not merely limit a right? Negation is the response to this question. The idea of negation emerges out of constitutional law through its contrast with the idea of limit. My suggestion is that this contrast is indispensable to delineating the boundaries of horizontality. Whereas limits on rights are within horizontality’s scope and, therefore, are taken into account in horizontality’s balancing operation, negations lie beyond horizontality’s scope. Accordingly, the consequence of thinking in terms of negation is twofold. First, as cases like Parabolantenne indicate, the application of constitutional rights to property relations may not necessarily raise the issue of horizontality’s scope. Secondly, horizontality’s scope may be in issue even when property rights are not. It is to this latter possibility that I next turn.

4.  Snyder v Phelps The facts of the case of Snyder v Phelps,52 decided by the Supreme Court of the United States, illustrate the relevance of negation in a non-​proprietary setting. The defendants were members of a religious group that publicized its anti-​Catholic and anti-​gay views by picketing in the vicinity of the funerals of soldiers killed in war. Some of their placards celebrated the deaths of American military personnel as divine punishment for the country’s sins. In Snyder, the group stationed itself on an adjacent street after having notified the police (and the media) of their protest. They fully complied with police guidance. The goal of their protest was not to disrupt the funeral, but to attract media attention to their view that God hated the United States for its toleration of homosexuality. Broadcast and news outlets converged to

51 52

This seems to be the view of Maultzsch (n 3). 131 S Ct 1207 (2011).

168  Ernest J. Weinrib cover an event notable for the incongruity between the protesters’ message and the event at which they were protesting. The father of the marine being buried sued the picketers for intentionally inflicting emotional distress. Although the Supreme Court noted the anguish that the defendants’ conduct added to the father’s ‘already incalculable grief ’,53 it dismissed the father’s claim, holding that the defendants’ speech in a public place on a matter of public concern was entitled to the constitutional protection accorded to freedom of speech. The decision reflects the pre-​eminence of the right to free speech within the constitutional order of the United States, a pre-​eminence born of ‘a profound national commitment that debate on public issues should be uninhibited, robust and wide-​ open’.54 This pre-​eminence renders the exercise of free speech the almost unique occasion for the horizontal application of constitutional rights in the United States.55 Although constitutional rights in the United States are understood to govern state power alone, the Supreme Court has held that a court decision in a civil lawsuit is itself an exercise of state power that might invalidly restrict a defendant’s constitutional freedom of speech.56 Although the Supreme Court has a well-​developed doctrine of free speech, it does not have a considered doctrine of horizontality. Neither Snyder nor the line of free speech cases that preceded it advert to the difficult issues raised by the nexus of constitutional law and private law. Moreover, the doctrine of free speech is regarded as jurisprudentially self-​sufficient, rather than as a component of a comprehensive and integrated system of rights. The question that then arises for the Supreme Court is not how the defendants’ right to free speech fits within a system of rights that also protects plaintiff ’s right not to be subjected to intentionally inflicted emotional distress, but whether the defendant’s statements qualify as constitutionally protected speech in view of the public significance of the issues to which they referred and the public location in which they were uttered. This latter question is directed towards the speaker and the values forwarded by free speech; it does not engage what the plaintiff has suffered from the speech. Accordingly, in Snyder once the Supreme Court determined that the defendants’ speech was in a public place on a matter of public concern, it held that the speakers were immune to tort liability despite the effect of the speech on the plaintiff. The facts in Snyder would generate a different analysis if subjected to the conception of horizontality’s scope put forward in this chapter. While acknowledging that I am wandering far beyond the doctrinal boundaries of constitutional

53 ibid 1218. 54 ibid 1215. 55 On the impact of constitutional rights on private actors in the United States see S Gardbaum, ‘The “Horizontal Effect” of Constitutional Rights’ (2003) 102 Mich L Rev 307. 56 NY Times v Sullivan 376 US 254, 265 (1964). For criticism of the Court’s assimilation of judicial decision and legislative action see NB Oman and JM Solomon, ‘The Supreme Court’s Theory of Private Law’ (2013) 62 Duke LJ 1109.

Horizontality: Scope and Operation  169 jurisprudence in the United States, I suggest that the idea of negation is pertinent in two ways. The first point is that the devastating effect of the picketing on the plaintiff ’s emotional tranquillity was not a merely co-​incidental consequence of defendants’ demonstration. The very point of the defendants’ action was to picket at a time and place where this effect was substantially certain. On the one hand, the funeral was a solemn occasion for grief and commemoration on the death of a son in the service of his country. On the other hand, the demonstrators proclaimed this death to be God’s punishment of a sinful nation. The juxtaposition of these two gave the picketing its rhetorical power, generating the media coverage that the defendants craved. The defendants treated the ceremony marking the tragic death of the plaintiff ’s son as simply available for exploitation in their quest for publicity, associating themselves with it in order to create a spectacle that aggravated the plaintiff ’s suffering and interfered with his ability to engage in personal and communal mourning.57 The defendants thereby signalled that in the pursuit of their goals the plaintiff ’s right to be free from the intentional infliction of emotional distress counted for nothing. They supposedly exercised their constitutional right through the negation of the plaintiff ’s private law right.58 The second (and more complicated) point focuses not on the emotional effect of the picketing on the father but on the defendants’ violation of the dignity both of the dead soldier and of his family. That the dead have dignity (or a cultural equivalent thereof) has been a commonplace of Western thought ever since Achilles committed the enormity of tying the body of the slain Hector to his chariot and dragging it around the tomb of Patroclus. The law, however, usually recognizes this dignity by highlighting the effect on the living rather than the treatment of the dead; the recognition of dignity is present but only in an oblique form, because presumably a reason for the distress of the living is the disrespect shown to the dead. The plaintiff ’s claim in Snyder exemplifies the longstanding assumption that, while the dead have no legally cognizable interest of their own, emotional distress can ground liability for the invasion of an interest in the memory or the body of the deceased.59 Similarly, with defamation. Unlike the common law, some civil law jurisdictions have recognized that the dead can be defamed, but since Roman times the basis for liability has been the injury to the reputations of heirs or relatives.60 57 M Lane Bruner and S Balter-​Reitz, ‘Snyder v Phelps: The U.S. Supreme Court’s Spectacular Erasure of the Tragic Spectacle’ (2013) 16 Rhet & Pub Aff 651, 658. 58 Exactly on point is Steven Heyman’s observation that the facts in Snyder should not lead to a balancing of rights, because ‘an asserted right can derive no value from its negation of another right’; SJ Heyman, ‘To Drink the Cup of Fury: Funeral Picketing, Public Discourse, and the First Amendment’ (2012) 45 Conn L Rev 101, 157. 59 FV Harper and MC McNeely, ‘A Re-​examination of the Basis for Liability for Emotional Distress’ [1938] Wis L Rev 426, 439–​43; B Zipursky, ‘Snyder v Phelps, Outrageousness, and the Open Texture of Tort Law’ (2011) 60 De Paul L Rev 473, 507–​11. 60 See eg Chiniquy v Bégin 7 DLR 65 (Que. SC 1912); for Roman law see D 47, 10, 1. 4 (Ulpian).

170  Ernest J. Weinrib Because human dignity is the guiding idea of the dignity function, the phenomenon of horizontality provides a platform for modifying this view of the status of the deceased. One’s human dignity does not come to an end when one dies. Recall that, as formulated in German constitutional jurisprudence, human dignity refers to the conception of a person endowed with self-​determination who exercises freedom ‘not as that of an isolated and sovereign individual, but as an individual who is based in and bound to the community’.61 On this conception, human dignity is a relational term that comprehends a person’s standing as a self-​determining being within his or her community. Integral to that standing is how one is regarded by others in view of one’s actions over a complete life. Understood in this way, human dignity does not terminate on death but endures in the recollection and esteem that others have of a person who is no longer alive. Consequently, posthumous defamation or the interference with the process through which the recollection is formed and the esteem is expressed is a wrong against the deceased and against those closest to the deceased, quite apart from the aggravation of their emotional distress. Within the horizontality context the dignity of the dead can be recognized directly and not merely obliquely.62 The notion that dignity is relevant after death appears in two leading horizontality cases from different jurisdictions. The first is the famous Mephisto case in Germany.63 That case concerned the publication of a roman à clef by the writer Klaus Mann based on the life of a well-​known and easily recognizable theatre personage (in fact, Mann’s one-​time brother-​in-​law), who had advanced his career by opportunistically currying favour with the Nazi authorities. The author’s stated intention was ‘to expose and analyze the abject type of the treacherous intellectual who prostitutes his talent for the sake of some tawdry fame and transitory wealth’.64 However, the novel included discreditable episodes that had not occurred in the life of the person on whom the book was based. By the time the book was to be published, that person had passed away. An application to restrain publication and distribution was successfully brought by his son and sole heir. The Constitutional Court held that, although freedom of artistic expression is not explicitly limited in the constitutional text, it is nonetheless subject to a limit immanent in the constitution when interpreted in the light of the unity of its underlying value system. The supreme and dominant value in that value system is the human dignity that 61 BVerfGE 45, 187 (1977); see ch 6, n 16. 62 For an engaging treatment outside the context of horizontality see D Herzog, Defaming the Dead (Yale UP 2017). 63 BVerfGE 30, 173 (1971), translated in Decisions of the Bundesverfassungsgericht: Federal Constitutional Court: Federal Republic of Germany, vol 2/​Pt I: Freedom of Speech, 147–​80 (1998). The most exhaustive discussion in English is H Rösler, ‘Dignitarian Posthumous Personality Rights: An Analysis of U.S. and German Constitutional and Tort Law’ (2008) 26 Berkeley J Intl L 153. For a recent treatment of the theme in the European Court see MAC TV v Slovakia (http://​hudoc.echr.coe.int/​ eng?i001-​178​958). 64 ibid 148.

Horizontality: Scope and Operation  171 is proclaimed as inviolate in the initial article of the constitution. The exaggerated portrayal of the applicant’s father was held to be inconsistent with the father’s human dignity. As for the fact that the person defamed was no longer alive, the Court said: It would be incompatible with the constitutionally guaranteed precept of the inviolability of human dignity that underlies all the fundamental rights for it to be possible for a person, having dignity in virtue of being a person, to be lowered or disparaged in this general entitlement to regard even after his death.65

The second case, the decision by the Supreme Court of Israel in Hevra Kadisha of the Jerusalem Community v Kastenbaum,66 is closer to the situation in Snyder, because it deals with the commemoration rather than the defamation of the deceased. The action was brought against a burial society by the husband of a person buried in a cemetery that it administered. The issue was whether, in accordance with the regulations of the burial society, the inscription on the tombstone had to be exclusively in Hebrew letters. The husband had submitted the society’s standard burial form that included the stipulation that arrangements were to conform to the burial society’s regulations. However, because the deceased had spent most of her life in the United States, was known by her English name, and had lived in accordance with the Gregorian calendar, her husband now requested that her name and dates be inscribed on the tombstone in English letters and numerals. The burial society refused. Justice Barak was willing to assume that the burial society was a private body and that there was a contract between the parties. He identified horizontality as the core problem: how should the basic principles of public law—​such as the value of the Hebrew language, human dignity, and tolerance—​be applied within the realm of private law? In holding that the exclusivity of Hebrew on the tombstone was inconsistent with the dignity of the deceased and her loved ones, he described the relationship between human dignity and the function of the tombstone in these terms: Human dignity is not simply the dignity of man in his own lifetime. It is also the dignity of man after his death, and the dignity of his loved ones, who preserve his memory in their hearts. This dignity is expressed, among other ways, in the very placement of a tombstone, in the visits of his loved ones to the cemetery, in memorial days, in public ceremonies, and in the upkeep of the grave. This is the



65

ibid 157. PD 46 (2) 464 (1992), translated at https://​law.ute​xas.edu/​transn​atio​nal/​fore​ign-​law-​trans​lati​ons/​Isra​eli/​case. php?id =​1391. 66

172  Ernest J. Weinrib connection—​sometimes rational, sometimes irrational—​between the living and the dead, which helps mold our humanity and gives expression to our deepest inner yearnings. This is the ‘hand’ that those living extend to the dead. This is the external expression of the inner bond between generations. The recognition of human dignity requires granting man the freedom to determine the inscription on his tombstone in accordance with his desire. The negating of this freedom, and the subsequent requiring of exclusively Hebrew inscriptions, is to infringe, severely and significantly, the basic value of human dignity.67

In this passage Justice Barak explicates the relational character of human dignity in the context of memorializing the dead. Just as the relationship between loved ones continues after the death of one of them, so does their human dignity with respect to this relationship. Their dignity is infringed if the tombstone, as the permanent marker and physical focus of this continuing relationship, is not inscribed in the language specifically appropriate to it and intelligible to the community in which the deceased lived her life. These observations about the continuing human dignity of the deceased are germane to the facts in Snyder. A funeral is not merely a process for sticking a corpse into the earth. Rather, it is a testimony to the relationship between the deceased as an actor and the world on which the deceased acted, an occasion for solemnly considering the meaning of the deceased’s life, for celebrating the deceased’s character, for recognizing the deceased’s accomplishments and aspirations, and for recapitulating the significant relationships that the deceased had with family and community. In Kantian terms, a funeral is a commemoration of the deceased as a being who was endowed with innate right and lived in community with other such beings.68 The claims of human dignity in Snyder are at least as strong as those in Hevra Kadisha. Both cases feature human dignity as expressed through the commemoration of the deceased’s life and connections with others. They differ only in that a funeral is an intense and declamatory event, whereas an inscription is a permanent

67 ibid para 10. In Hebrew, the word ‘hand’ has the secondary meaning of ‘memorial’. 68 In a difficult and provocative passage concerning ‘a phenomenon as strange as it is undeniable’ Kant (n 4) 83–​84 [6:295–​96] explains defamation of the dead by arguing that when a person is regarded solely in terms of his humanity (as is the case with defamation), one abstracts from the physical conditions of time, including the fact that the defamed person has ceased to exist. The passage is suggestive in its statement that the violation of the deceased’s right does not depend on what the victim experiences. Kant regards the victim of a posthumous defamation as being wrongly injured, but he expressly repudiates the notion that his view presupposes that the deceased continues to exist as a disembodied spirit who somehow feels the injury. However, the passage is not quite apposite to issues of liability. Kant is here concerned not with a legal proceeding, but with one’s entitlement to publicly rebut a posthumous calumny and so bring the calumniator into disrepute in the eyes of public opinion. His point is that the public dishonouring of the calumniator is the proper retribution for the dishonour wrongly brought about by the original calumny. For a commentary on the puzzles in this passage see T Mertens, ‘Bona Fama Defuncti in Kant’s Rechtslehre: Some Perspectives’ (2019) 24 Kantian Rev 413.

Horizontality: Scope and Operation  173 and concise memorial. From the standpoint of horizontality’s scope, the significant distinction between the two cases is that Snyder, but not Hevra Kadisha, involved the negation of human dignity. In Hevra Kadisha, the burial society’s insistence on an exclusively Hebrew inscription infringed on human dignity but did not negate it. The exclusively Hebrew inscription would have contained the relevant lapidary information, but in a form that did not adequately correspond to the way that the deceased and her family lived and communicated. In Snyder, in contrast, the defendants exploited the funeral of a slain marine to publicize their own hateful beliefs. The funeral and the mourners were relegated to the role of involuntary props in a media stunt of the defendants’ creation. Through their signs (‘Thank God for Dead Soldiers’, ‘Semper Fi Fags’, ‘You’re Going to Hell’, and so on) they denigrated the marine’s sacrifice and mocked the mourners’ grief, using that grief—​and thus the mourners themselves—​solely as a means to their end. They thereby treated as nullities the rights inherent in the dignity of the mourners and of the deceased. Their conduct was a negation of rights with respect not only to emotional distress but also to human dignity. In a legal system in which the phenomenon of horizontality is circumscribed within its proper scope, these negations of the rights of others should in principle have precluded the defendants’ successful appeal to their exercise of a constitutional right.

5.  Balancing How, then, does horizontality operate for a controversy within its scope? The distinctive feature of horizontality’s operation is the balancing of the divergent effects of different rights. Because modern constitutions enumerate the various rights that act as specifications of human dignity, the possibility arises as a matter of both public law and private law that in a given situation these rights may pull in different directions. Balancing resolves this kind of tension. On the public law side, in the contest between the state and the individual only the individual has an infringeable right. When the state invokes a countervailing constitutional right, it is to show that the encroachment on the individual’s right is justified. This exercise in justifying the infringement of one right by appeal to another right is the second stage in a two-​stage structure of analysis. The first stage considers whether a right was infringed, the second whether the forwarding of a countervailing right justifies that infringement. This second stage features a complex inquiry into the objective of the impugned measure, its consistency with the rule of law, its suitability and necessity for achieving its objective, and the proportionality of its effect on the right being forwarded as compared to its effect on the right being infringed. The point of this two-​stage analysis is to determine whether the infringement of one right for the sake of another leaves the system of rights as a whole better off from the standpoint of its own right-​based values.

174  Ernest J. Weinrib A parallel issue arises—​but to a different effect—​in the horizontality context when both parties to a dispute point to rights that the private law or the constitution recognizes (or to values that the constitution expresses).69 The paradigmatic example is the publication of details of another’s life that sets one person’s right to free expression against another person’s right to privacy. Such situations do not trigger a two-​stage inquiry into infringement and justification. Instead, as befits adjudication about interaction between transactional equals, a court must reconcile the divergent effects of the rights in order to determine whether the defendant’s conduct was rightful to begin with. A second stage dealing with justification is unwarranted, because the countervailing rights go to the initial composition of the parties’ normative relationship. From the standpoint of corrective justice, the balancing that characterizes the operation of horizontality is unusual. In the traditional contexts of liability, corrective justice eschews balancing, because balancing is usually undertaken to achieve a goal specifiable independently of law, such as the maximization of efficiency or of utility. Corrective justice focuses more narrowly on identifying the plaintiff ’s entitlement—​in Hohfeldian terms, a claim-​right that carries with it a correlative duty—​and on specifying the kind of action on the defendant’s part that is inconsistent with that entitlement (for example, exposing the plaintiff ’s bodily integrity to unreasonable risk in negligence law,70 or not performing an action to which the plaintiff is contractually entitled). Talk of balancing would, accordingly, be misplaced. A single homogeneous right pervades the parties’ entire relationship, constituting both the juridical manifestation of the plaintiff ’s freedom with respect to the defendant and the determinant of the defendant’s permissible action with respect to the plaintiff. Horizontality sets up a new dynamic. A rights-​protecting constitution partitions human dignity into a variety of enumerated rights and liberties. Although they all together make up the ensemble of rights for which human dignity provides the unifying abstraction, each of them on its own has a distinctive content. This gives rise to possible conflicts in their operation in particular circumstances. Thus, when horizontality is in play, the parties’ relationship is made up not of a single homogeneous right through which the two parties are related, but of a multiplicity of distinctive rights and freedoms, to one or the other of which each of the parties can 69 By this last clause, I refer to a jurisdiction such as Canada, where the fundamental values enshrined in the constitution, rather than the constitutional rights themselves, apply to private litigation. For example, privacy or reputation are not specifically mentioned in the constitution, but they are relevant to the development of the common law in accordance with constitutional values because they reflect ‘the innate dignity of the individual, a concept which underlies all Charter rights’; see Hill v Church of Scientology [1995] 2 SCR 1130, paras 120, 121. 70 Some might regard the Learned Hand formula for negligence (United States v Carroll Towing 159 F 2d 169, 173), in which the probability and gravity of the injury is compared to the burden of precautions as a form of balancing. Corrective justice rejects the Learned Hand formula; EJ Weinrib, The Idea of Private Law (Harvard UP 1995, OUP 2012) 147–​52.

Horizontality: Scope and Operation  175 appeal.71 The different contents of these rights and freedoms requires a method for bringing them into mutual alignment when in particular contexts they pull in different directions. For example, when horizontality juxtaposes privacy and freedom of expression, how can the two rights co-​exist in the specific circumstances of the case? The answer is: through the process of balancing. Although the term ‘balancing’ is so firmly ensconced in the legal literature that its use can now hardly be avoided, the image it summons up is infelicitous. One might think that different rights are being placed on the different pans of a scales, with a view to measuring and comparing the difference in their response to constitution’s gravitational pull. This is misleading. What is being compared is not the rights themselves, but the effect of those rights on the reciprocal freedom of the parties in the particular circumstances of a case. The operation of horizontality involves not the quantitative measurement of weights (however metaphorically conceived), but a practical judgment comparing the countervailing effects of different rights in a specific context. When understood as the determinations of human dignity within a system of reciprocal freedom, rights share a common standpoint from which their effects relative to each other can be assessed though exercises of legal reasoning and legal judgment. Conclusions are reached not by metaphorically gauging the pull of a physical force, but by casuistically assessing the interplay of constitutional rights in the circumstances at hand. The reference to balancing does nothing to illuminate this process of reasoning and judgment.72 More apt than ‘balancing’ are terms like ‘practical concordance’ and ‘reconciliation’.73 These terms make it clear that the point is to relate the operation of each countervailing right to that of the other in such a way as to implement both of them to the extent possible. The concordance is ‘practical’ in that it is geared not to the rights as abstractions but to their reciprocal effects in the particular circumstances of the case. The distinction between practical concordance and the more standard conception of balancing has been summed up as follows: Practical concordance differs from normal balancing, in that it should not . . . be exhausted in a facile balancing of interests or even in an abstract weighing of values. Rather, the balancing to be carried out is subject to a dual duty to optimise: both of the legal interests must be kept within limits so that both achieve

71 For an exemplary discussion concerning copyright and human rights see A Drassinower, What’s Wrong with Copying (Harvard UP 2015) 186–​26. 72 For a more charitable appreciation of balancing while recognizing its metaphorical character see A Barak, Proportionality: Constitutional Rights and their Limitations (CUP 2012) 164–​76. 73 These are the terms suggested by judges on the highest courts in Germany and Canada, respectively; see T Marauhn and N Ruppel, ‘Balancing Conflicting Human Rights: Konrad Hesse’s Notion of ‘Praktische Konkordanz’ and the German Federal Constitutional Court’ in Brems (n 46) 273; F Iacobucci, ‘“Reconciling Rights”: The Supreme Court of Canada’s Approach to Competing Charter Rights’ (2012) 20 SCLR 137. A similar idea is the ‘horizontal balancing’ to which Barak refers; see Barak (n 72) 171.

176  Ernest J. Weinrib optimal effectiveness. In [normal] balancing, one legal interest is typically brought into relationship with another in such a way as to determine the degree to which the former is limited by the latter. In this respect, one can speak of an analytic perspective that is one-​sidedly oriented. One thereby asks whether the limitation of the fundamental right at the centre of the analysis is appropriate, i.e. whether the limitation observes a tolerable measure. Practical concordance, on the other hand, pursues a different and reciprocal approach, which does not start from a one-​sided preconception, but from latently fluctuating boundaries, the optimal course of which is dictated by the circumstances of the individual case. The analysis is therefore based on a bipolar perspective . . . Despite the reciprocal delimiting of the two constitutional interests, as much as possible of the original protective content of each is preserved.74

Practical concordance thus involves mutually adjusting the operation of each right in a particular context because of the presence of the other. If ‘balancing’ matters, it is only because it is the verb that refers to this effort to establish a practical concordance of rights, and not because the scalar image in itself has any explanatory power. When engaging different constitutional rights, practical concordance proceeds from four fundamental presuppositions about the constitutional order.75 First, the rights enumerated in the constitution are not individually absolute; they operate in relation to one another as mutually qualifying. In the example posed above, when the effect of freedom of expression pulls against the effect of the right to privacy, the solution is not to apply either of these rights to its full extent and to ignore the other. Rather, each right operates in the light of the other. Secondly, constitutional rights are not ranged in a hierarchy in which one is subordinated to another. The tension between countervailing rights cannot be resolved by postulating, for instance, that freedom of expression, as such, takes priority over the right of privacy, as such (or vice versa). All constitutional rights have an equal status as the coordinate juridical manifestations of human dignity, and thus all have an equal claim to be respected. Thirdly, constitutional rights comprise the constituents of a unified whole that gives expression to the legal pre-​eminence of human dignity in one’s interactions with another. This participation of rights in a unitary system of value is merely the positive correlate to their non-​absolute and non-​hierarchical character: rights that are mutually qualifying and co-​ordinate can be conceived as coherently related only through the more inclusive unity that they constitute. Fourthly, all rights are practical in that they produce effects in particular situations. Accordingly, their application calls for an appreciation of the specific 74 M Schladebach, ‘Praktische Konkordanz als Verfassungsrechtliches Kollisionsprinzip: Eine Verteidigung’ (2014) 53 Staat 263, 271. 75 Compare Iacobucci (n 73) 138–​40.

Horizontality: Scope and Operation  177 contexts in which they operate and of the specific effects that they might generate. In the words of the German Constitutional Court, the judicial decision must take account of ‘the intensity and extent of the impact of various possible interpretations and applications on the affected legal interests of both parties’.76 In the operation of horizontality, the equality of constitutional rights corresponds to the transactional equality of the parties in private law. In a situation calling for the achievement of a practical concordance of rights, each of the parties asserts a different right. A court can treat the parties as equals only if it regards their different rights as equally applicable. The bipolarity of practical concordance matches the bipolarity of the parties’ relationship. The point of practical concordance is to minimize the degree to which the exercise of each person’s right impairs the exercise of the right of the other. To this end, practical concordance brings together the reciprocal impact of the exercise of two countervailing rights, the respective purposes of the rights in play, and the proportionality of the operation of these rights on each other. These three features, listed in ascending degrees of generality, each reflect aspects of a rights-​based legal order embodying reciprocal freedom. To begin with, the reference to reciprocal impact makes salient the specific effects that rights, for all their abstract character, have in the particularity of human interaction. For practical concordance the particular context in which horizontality operates in a given case and the particular effects of the countervailing rights within that context are paramount. Practical concordance works at the granular level of the specific interaction between the parties, comparing the impacts on the enjoyment of each right-​holder’s right that would follow from the vindication of the right of the other right-​holder. This assessment of impacts requires the exercise of a judgment about their relative severity and the relative closeness of their connections to the right being exercised. In the contest between freedom of expression and privacy, for instance,77 how much more severely does a newspaper affect a person’s privacy by publishing a photograph rather than by merely describing an event? And from the standpoint of freedom of expression, is the publication of a photograph needed for validating the truth of the story or for allowing the newspaper the margin of editorial judgment necessary for its functioning? Or, to take another instance,78 does it matter that publishing the name of a mother facing criminal charges for murdering her child might have a significant but indirect effect on the mental health of the mother’s other child? Next, such effects are related to the right through the purpose that the right serves. In a rights-​based system, the protection afforded by a right reflects the

76 1 BvR 1585/​13 (Sampling 2016) para 85. 77 Campbell v MGN Ltd. [2004] 2 AC 457. I consider the ‘ultimate balancing test’ in the United Kingdom (In re S (FC) (a child) [2005] 1 AC 593, para 17) to be a version of practical concordance. 78 In re S (FC) (a child) [2005] 1 AC 593.

178  Ernest J. Weinrib purpose underlying the right.79 Hence, the more central the impugned conduct is to the right’s purpose (that is, to the value that the right expresses), the more significant is the right’s impairment if the conduct is not allowed. Although a court has to treat a right as ‘in itself desirable’ and therefore not to be impaired without good reason, ‘[i]‌t must also consider those features which enhance its importance in the particular case’.80 Thus, when freedom of speech is pitted against privacy, '[t] here are undoubtedly different types of speech, just as there are different types of private information, some of which are more deserving of protection in a democratic society than others’.81 Political speech or artistic expression, for instance, rank as more central to the purpose of freedom of expression than the revelation of intimate details of a celebrity’s private life. Accordingly, a court looks at ‘the comparative importance of the actual rights being claimed in the individual case’,82 not in order to compare the right to freedom of expression to the right to privacy—​ex hypothesi, these are of equal importance83—​but within those rights to compare the importance to each of them of the types of entitlement (the ‘actual rights’) that the right-​holders are asserting. Finally, the various impacts of these rights with their respective purposes are related to each other through a principle of proportionality that functions as ‘the mechanism by which the court reaches its conclusion on countervailing or qualified rights’.84 Under this principle neither right can be given a broader operation than is required to achieve the most adequate concordance of both. Conduct that impairs another person’s right for the sake of forwarding one’s own is proportionate when such conduct is a suitable means of forwarding one’s right, when it does not impair the other right more than is necessary, and when, in view of the purposes of each of the countervailing rights, the adverse effect on the right being impaired is less significant than favourable effect on the right being forwarded. The inquiry into proportionality addresses the question of ‘the extent to which it is necessary to qualify one right in order to protect the underlying value which is protected by the other. And the extent of the qualification must be proportionate to the need’.85 Essential to this assessment is that it starts from the equality of the two countervailing rights, and then ascribes proper significance to both of them. Although it restricts the operation of one of the rights in the particular circumstances of the case, the resulting accommodation of rights yields the optimal construction of reciprocal freedom from the standpoint of the system of rights as a whole.

79 J Weinrib, Dimensions of Dignity: The Theory and Practice of Modern Constitutional Law (CUP 2016) 221. 80 In re S (a child) [2003] EWCA Civ 963, para 55 (Hale LJ). 81 Campbell (n 77) para 148 (Baroness Hale). 82 ibid para 141. 83 ibid para 140. 84 Douglas v Hello! Ltd [2001] QB 967 1005. 85 Campbell (n 77) para 55 (Lord Hoffmann), emphasis in original.

Horizontality: Scope and Operation  179 Proportionality so understood is similar to the idea of proportionality that applies on the public law side to legislative and governmental action, but with an illuminating exception. In the public law context, the objective of the rights-​ encroaching measure is the ‘fixed point of reference’86 for the enquiry into proportionality. The government has the burden of proving that the end that the state has set for itself in the specific legislation is sufficiently compelling to warrant the encroachment on a guaranteed right. The assessments of rational connection and minimal impairment are relative to that objective: is the impugned measure a suitable means of accomplishing the measure’s supposed purpose, and can that purpose be accomplished by a less serious infringement of the right? Similarly, the comparison of the adverse effects of encroaching on the right and the beneficial effects to the system of rights hinges on the advantages gained from attaining measure’s objective. Thus, the entire inquiry revolves around the identification, assessment and achievement of the objective ascribed to the impugned measure. In contrast, elucidation of the specific objective of the parties’ conduct plays no role in the horizontality context. The compelling state interest is contained in the right itself, which would not have legal or constitutional status unless it was for the public benefit. Moreover, the purpose that is relevant is not the purpose of either litigant, but the purpose of each right, because that purpose is the measure of the significance of the effects to be compared. In controversies involving horizontality, the parties confront each other solely as right-​holders, and not as seekers of particular ends. Whereas persons acting in an official governmental capacity are obligated to pursue public purposes, private parties act (as is entirely legitimate) for their own self-​determined purposes. It makes no sense for the law to enquire into the suitability and necessity of the means that they are adopting for the ends that they have set for themselves. Within the bounds of reciprocal freedom, they can adopt any ends they want and choose to pursue them through any means they want. This is just a consequence of the distinction between public law and private law: public law requires government authorities to act for public purposes, whereas private law is indifferent to the purposes for which persons exercise their rights, provided that they do not infringe on the rights of others. The difference between these two kinds of proportionality reflects nothing less than the categorical difference between public law and private law. By treating constitutional rights as a source of pervasively applicable principles, horizontality does not leave private law as it was, but neither does it erase the line between private law

86 E-​W Böckenförde, Constitutional and Political Theory: Selected Writings (OUP 2016) 254. See also the incisive comments in H Collins, ‘On the (In)compatibility of Human Rights Discourse and Private Law’ in H Micklitz, Constitutionalization of European Private Law (OUP 2014) 26, 49–​51.

180  Ernest J. Weinrib and public law. Rather, when one puts this chapter’s observations on horizontality’s operation together with its earlier observations on the exclusion of distributive justice from horizontality’s scope, it is fair to say (on the analogy of the distinction between private international law and public international law) that horizontality’s constitutionally grounded specification of human dignity belongs not to constitutional law simpliciter, but to a special branch of it: private constitutional law.

8

The Rule of Law 1.  Public Law in Disguise? Thematic for this book has been the idea that private law is distinct from public law. Accordingly, the book began in Chapter 1 by elucidating the structure of the private law relationship in terms of corrective justice, and then moved in Chapters 2 and 3 to the rights that form the content of that relationship. Chapter 4 set out the role of legal institutions in specifying and securing those rights and in transforming them from bilateral to omnilateral markers of reciprocal freedom. Chapters 6 and 7 contended that even constitutional rights (often regarded as the supreme manifestation of public law), when applied to private law in the most robust version of horizontality, absorb the structure of the private law relationship into their scope and operation. This anchoring of private law in corrective justice does not deny the significance of distributive justice within the legal system considered in its entirety. Just as private law is not the only kind of law, so corrective justice is not the only kind of justice. Rather, as argued in Chapter 5, distributive justice is located within a conceptual sequence that presupposes and complements—​but also leaves intact—​the distinctive role of private law in actualizing corrective justice. Given the irreducible pluralism of these different kinds of justice, the juridical unity of the legal system consists in their integration within that conceptual sequence. On this view, distributive justice plays no role in private law specifically, but is nonetheless an obligatory end of the state, if the legal relationships that the state authorizes are to be modalities of reciprocal freedom. A non-​distributive conception of private law is thereby situated within a non-​libertarian conception of the state. The polar opposite of this sharp distinction between private law and public law is the view, rooted in American legal realism, that private law (or any branch of it, such as tort law) is really ‘public law in disguise’.1 This view graphically combines a claim about the substance of private law with a claim about its discourse. The claim about substance is that private law litigation at its heart does not focus (as the corrective justice approach maintains) on the normative relationship between the plaintiff and the defendant. Rather, ‘ “We the People” are a party to every lawsuit, and it is our interest that weighs most heavily in its determination’.2

1

L Green, ‘Tort Law Public Law in Disguise’ (1959) 38 Texas LR 1.

2 ibid.

Reciprocal Freedom. Ernest J. Weinrib, Oxford University Press. © Ernest Weinrib 2022. DOI: 10.1093/​oso/​9780198754183.003.0008

182  Ernest J. Weinrib Lawsuits are viewed as exercises in public law because courts ‘cannot and should not escape taking into consideration . . . the interests of “We the People” at large or of important groups whose interests are identified with the interests of the social order’.3 For example, in the famous English decision of Rylands v Fletcher,4 in which a mine owner sued the builder of an adjacent reservoir for damage caused by escaping water, the plaintiff ’s success is to be ascribed to the long-​standing importance of the mining industry to the English economy: the defendant was held liable because ‘[t]‌o hurt mining was to hurt England’.5 On this view, the judicial process is in truth a political process in which certain actors—​especially lawyers, judges, and jurors—​are invested with the power to translate those interests, already embedded in them as members of society, into legal decisions.6 The claim about discourse asserts a disjunction between the language of court judgments and the factors really at work in determining liability. Court decisions (so the argument goes) are motivated not by the characteristic reasoning of private law, but by unacknowledged policies that reflect the interests of ‘We the People’. Court decisions, accordingly, are not the consequence of the reasons that are publicly displayed as their basis. The relation between the real and the professed grounds of decision is one of disguise, with courts ‘masking’7 what they are really doing in the language of ‘dry and lifeless’ legal doctrine.8 The consequence is a ‘waste in judicial effort where policies are swallowed up in doctrine’.9 Although the common law, perhaps more than any other legal system, emphasizes the centrality of publicly declared reasons to determinations of liability, the claim about discourse presents the legal reasoning in private law cases as opaque and misleading. Instead of being the directly relevant, publicly certified, and readily accessible explanations of the bases for a person’s rights and obligations, the professed reasons turn out to be mechanisms of deception. The thesis that private law is public law in disguise thus seems to set private law at odds with the rule of law. This observation about the role of disguise brings me to the theme of this chapter: the significance of the rule of law for private law and more generally. While the rule of law has long figured in discussions of public law, its relationship to private law has only recently become the object of focused academic enquiry.10 The present chapter contributes to this literature by first presenting a general account of the rule of law, and then by turning to the connection between private law and the rule of law.



3

ibid 2. [1868] LR 3 HL 330. 5 Green (n 1) 3. 6 ibid 2–​3. 7 ibid 8. 8 ibid 10. 9 ibid. 10 L Austin and D Klimchuk, Private Law and the Rule of Law (OUP 2014). 4

The Rule of Law  183 The concept of the rule of law leads a double life as a principle of positive law and as a juridical idea. In the former role, it figures in the positive law to an extent and with an effect that may vary from jurisdiction to jurisdiction.11 In the latter role, the rule of law states an ideal (or set of ideals) to which the law should conform.12 In view, however, of the existence of many ideals of which this might be said, what has to be added to make the idea a juridical one is that the rule of law stands in a particularly intimate relationship with the law that other ideals cannot claim. By failing to conform to the rule of law, an official act is unfaithful to deeper or more general values that are inherent in the project of legality. The rule of law thus offers a critical standpoint on law from within the law itself.13 This standpoint, as elaborated throughout this book, views legal institutions and legal concepts as actualizing what is normatively presupposed in the legal system as a coherent framework for reciprocal freedom. In a well-​ordered legal system the positive realizes the juridical to a greater or lesser extent, while the juridical informs the positive, drawing the law to a version of itself that is truer to its own normative presuppositions. It is from the standpoint of the juridical so conceived that the present chapter proceeds.

2.  Two Functions of the Rule of Law The rule of law ties the avoidance of arbitrary rule to principles that belong to the very idea of legal ordering. By holding laws (and their application) to what is inherent in their own legality, the rule of law affirms that law is not completely pliable to the will of those who control its apparatus for exercising power. Hence, the rule of law stands in clichéd contrast to the rule of men. The rule of law is often presented as a miscellany of elements. In contemporary treatments, the starting point is Lon Fuller’s masterly explication of the eight desiderata of what he called ‘the law’s internal morality’: a law-​maker fails to make law if the norms created lack generality, are not publicly promulgated, are retroactive, incomprehensible, contradictory, impossible to perform, unstable over time, or not congruent with official action.14 To these desiderata some propose to add procedural elements (eg a right to a fair hearing, to independent counsel, and to reasons 11 In Canada, for instance, the rule of law, which is now recognized in the preamble to the Canadian Charter of Rights and Freedoms, has been called ‘a fundamental postulate of our constitutional structure’ (Roncarelli v Duplessis [1959] SCR 121) and ‘a principle of profound constitutional and political significance’ (Reference Re Secession of Quebec [1998] SCR 217), providing a shield from arbitrary state action and a stable, predictable, and ordered framework for the conduct of individual affairs. It has been used to strike down administrative action and to prevent a wholesale absence of law, but not to invalidate legislation on the basis of its content. (British Columbia v Imperial Tobacco Canada Ltd [2005] 2 SCR 473). 12 J Raz, ‘The Law’s Own Virtue’ (2019) 39 OJLS 1. 13 TRS Allen, The Sovereignty of Law: Freedom, Constitution, and Common Law (OUP 213) 88 (distinguishing the rule of law as an internal value of the law from external ideals). 14 LL Fuller, The Morality of Law (rev edn, Yale University Press 1965)33–​94.

184  Ernest J. Weinrib for adjudicative and administrative decisions), as well as (more controversially) democracy, political rights, individual rights, and rights to welfare.15 This miscellany of elements is then standardly divided into those pertaining to competing ‘formal’ and ‘substantive’ versions of the rule of law, or situated on a spectrum from ‘thin’ to ‘thick’ calibrated to the number of elements included.16 Determining which elements properly belong to the rule of law remains a matter of significant dispute. In the face of this apparent ‘hodgepodge of unrelated legal aspirations’,17 the next sections of this chapter focus on the considerations that sort out and organize our understanding of the rule of law as a juridical concept: the functions of the rule of law, the demands that law places on those who are subject to it, and the tasks for which law is deployed. My contention is that the rule of law takes as its subject matter the constitutive aspects of a state that institutionally separates its legislative, executive, and adjudicative tasks. To begin with, the rule of law has two conjoined functions. In exercising its power in the form of law, a state that adheres to the rule of law operates both through law and in accordance with law.18 It operates through law by using laws to articulate its demands on its subjects. It operates in accordance with law by submitting itself to the laws that it articulates. Law is thus a medium both for exercising and for restraining state power: the state does not merely enact law that binds but also is itself bound by the law that it enacts. This concurrence of exercise and restraint—​the idea that the state power is inherently self-​limiting—​is what makes the rule of law the antithesis of arbitrary rule. Fuller’s desiderata are normative concomitants of the first of these functions, the state’s exercise of public power through law. Fuller’s desiderata set out what is presupposed in the enterprise of submitting human conduct to the governance of legal rules. The desiderata cluster around three intertwined facets of that enterprise. The first facet is that law governs human conduct through general norms rather than through a congeries of ad hoc determinations applicable to each specific episode of human behaviour considered on its own. This requirement of generality rests on the truism that to subject human conduct to the control of rules, there must first be rules.19 However, in addition to framing the rule-​bound content of law’s internal morality, the requirement of generality also contributes to the 15 A Bedner, ‘The Promise of a Thick View’ in C May and A Winchester (eds), Handbook on the Rule of Law (Edward Elgar Publishing 2018) 34. 16 BZ Tamanaha, On the Rule of Law: History, Politics, Theory (CUP 2004) 91; J Beatson, The Rule of Law and the Separation of Powers (Hart Publishing 2021)17–​21. 17 GF Gaus, ‘Public Reason and the Rule of Law’ (1994) 36 Nomos 328, 330. 18 D Grimm, Constitutionalism: Past, Present and Future (OUP 2016) 345 (describing the idea of the Rechtsstaat, in which the state exercises its power ‘nach Regeln und . . . durch Regeln’; Grimm, ‘Stufen der Rechtsstaatlichkeit’ (2009) 12 JZ 596). Similarly, in discussing the rule of law the Supreme Court of Canada has characterized the state as ‘an entity bound by and acting through law’; R v Nova Scotia Pharmaceutical Society [1992] 2 SCR 606, 641. 19 Fuller (n 14) 51.

The Rule of Law  185 law’s practical intelligibility by facilitating the discernment of the patterns of behaviour that are subjected to the law’s governance. The second facet takes up the relation between this practical intelligibility of rules and their aptitude for guiding conduct. The desiderata of public promulgation, prospectivity, comprehensibility, non-​contradictoriness, possibility of performance, and stability over time are the conditions for the possibility of the state’s exercising its power in a way that could influence conduct. The third facet, represented in the requirement of congruence between the rules as announced and their actual administration, ensures that compliance with the rules is not pointless. The absence of such congruence renders futile the very adherence to law that the enterprise of governing human conduct by rules seeks to elicit. In Kantian terms, the state has the authority to lay down rules that obligate its subjects to coercively enforceable obedience. This authority is necessary for the reciprocal freedom of everyone. In its absence, the determination, interpretation, and enforcement of rights would be unilateral acts of private power rather than omnilateral acts of public right. However, the state disrespects the freedom of its subjects if it requires them to comply with laws that were secret or retroactive or contradictory or incomprehensible or impossible to fulfil or constantly changing. State enforcement of practically unfulfillable laws cannot induce conduct in accordance with their terms. Thus, Fuller’s desiderata specify the minimal conditions for the coercive enforcement of laws in a manner consistent with the freedom of all. Accordingly, if the practical impossibility or pointlessness of complying with the state’s prescriptions precludes the obedience due to them as law, they cannot function as authoritative public norms. Equally, officials who posit such prescriptions cannot be acting for the public purpose that official acts presuppose, because no intelligible or feasible public purpose emerges from norms that cannot realistically influence the action of those who are supposed to adhere to them. What are invoked as laws would be mere pretexts for state intimidation. As Michael Oakeshott observed, Fuller’s desiderata are ‘inherent in the notion of law’, because they are ‘conditions which distinguish a legal order and in default of which whatever purports to be a legal order is not what it purports to be’.20 Whereas the first of the functions of the rule of law (state action through law) applies generally to a state of any kind by virtue of its authority to make binding law, the second function (state action in accordance with law) requires a state to restrain itself by being organized in a particular way. Kant calls this kind of state a ‘republic’.21

20 M Oakeshott, ‘The Rule of Law’ in M Oakeshott, On History and Other Essays (Blackwell 1983) 119, 140. 21 I Kant, ‘Toward Perpetual Peace’ in I Kant, Practical Philosophy (M Gregor trans, CUP 1996) 324 [8:352].

186  Ernest J. Weinrib In the Kantian account of the different forms of government, a republic is the form most adequate for the realization of reciprocal freedom. This is because a republic ‘makes freedom the principle and indeed the condition for any exercise of coercion’.22 Because politics is just the application of the idea of reciprocal freedom to the conditions of experience,23 the principle that is supremely regulative for the state’s law and politics—​‘the final end of all public right’24—​is the realization of reciprocal freedom. This requires the establishment to the extent possible of the legal norms and institutions that enable each person’s freedom to co-​exist with that of every other person. Kant’s work presents an integrated and expansive outline of such a state, sketching both its internal organization and the kinds of activity in which it would engage. The organizational precondition of a republic’s activity—​and thus its defining principle as a form of government—​is the separation of legislative, executive, and judicial authority. This distinguishes a republic from its opposite, a regime in which these kinds of authority are conflated, which Kant and other thinkers of his time termed a ‘despotism’.25 The exercise of each of these three varieties of authority consists in the performance of a specialized normative task. Legislative authority commonly lays down norms that are abstract and impersonal, governing situations and persons through general descriptions without in their terms specifying any particular situation or identifying any particular person.26 The legislative effect on particular situations and persons is determined not individually but by reference to the general properties that the legislation formulates. In contrast, executive authority has the task of administering and enforcing the legislation in its application to concrete cases and specific persons, thereby dealing with the particulars that fall under the general norm that the legislation articulates. Finally, judicial authority has the task of impartially and disinterestedly determining in accordance with evidence and legal argument the rights and obligations of parties to a dispute about the law. These three tasks of articulating a general rule, subsuming particulars under a general rule, and determining the legal effect of such subsumption on a specific person are categorically distinct from one another and yet are jointly necessary to the enterprise of political governance. Absent the exercise of executive authority, the state, like a paralysed person unable to accomplish what he willed,27 could 22 I Kant, The Metaphysics of Morals (L Denis ed, M Gregor tr, CUP 2017) 122 [6: 340] (emphasis in original). 23 I Kant, ‘On a Supposed Right to Lie from Philanthropy’ in I Kant, Practical Philosophy (M Gregor tr and ed, CUP 1996) 614 [8:429]. 24 Kant (n 22) 122 [6:341]. 25 ibid; cf J-​J Rousseau, The Social Contract, III, 1, 9 in J-​J Rousseau, The Social Contract and Other Later Political Writing (V Gourevitch ed and tr, 2nd edn, CUP 2019) 85. 26 This statement is not true of Private Acts and Local and Personal Acts of Parliament. These are not relevant to the present discussion. 27 Rousseau’s image in The Social Contract, III, 1, 2 in Rousseau (n 25) 84.

The Rule of Law  187 promulgate general laws that would be without effect on the persons subject to them. Absent the exercise of legislative authority, the state would have nothing to carry out. Absent the exercise of judicial authority, the state would be unable to resolve disputes about what the law required. The state’s functioning requires the availability and concurrence of all three. In Kantian terms, the three are subordinate and coordinate, mutually limiting and mutually complementary.28 They are subordinate in that each designates an activity that cannot be performed by either of the others and to which the others have to defer. And they are coordinate in that each complements the others so that together they constitute the entire range of state action. A republic is the form of government that gives effective institutional recognition to the separateness of these three tasks.29 The standard way in which this is done is to allocate the different tasks to the specified public institutions—​the legislature, the executive, or the judiciary—​that are dedicated to one or the other of them. Each of these institutions has the structure and responsibilities appropriate to its particular task. In modern polities, this institutional separation of powers is never completely achieved. Nonetheless, it can be made sufficiently effective with respect to the operation of each institution’s core competency, so that the exercise of public powers is limited in accordance with the distinctive task assigned to each.30 And even when institutional separation is not achieved—​when, for instance, an administrative agency has the power both to apply rules and to adjudicate the legality of that application—​the distinctness of the two kinds of tasks can nonetheless be respected at the institutional level through adherence to a principle of procedural fairness which requires that the executive and adjudicative tasks be carried out by separate procedures or separate personnel and to be subject to

28 Kant (n 22) 101 [6:316]. Kant’s comments apply to the offices invested with the three varieties of authority rather than to the tasks attached to them. For a similar statement taken from German constitutional jurisprudence see C Bumke and A Voßkuhle, German Constitutional Law: Introduction, Cases, and Principles (OUP 2019) 348. 29 The separation of powers is sometimes criticized on the ground that the distinction between the legislature, the executive, and the judiciary no longer provides a dispersal of power that is suitable to a modern society; see eg MJC Vile, Constitutionalism and the Separation of Powers (2nd edn, Liberty Fund 1998) chs 12, 13. From the Kantian perspective, the separation of powers is not, in the first instance, an issue of political science about dividing power, but an issue of jurisprudence about the legal functions of the state. As Arthur Ripstein puts it: ‘Each of the basic things that states do must be shown to be made consistent with freedom before turning to any question of how various offices might be staffed or kept under control’; A Ripstein, Force and Freedom (HUP 2009) 175. 30 BVerfGE 95, 1, V, B, 2 (Stendal South Bypass 1996): ‘The principle of the separation of powers is never realized in its pure from. The powers interlock and balance in numerous ways. The Basic Law does not demand an absolute separation, but rather the mutual control, limitation and moderation of the powers. However, the division of powers between the three powers, as laid down in the Constitution, must be preserved. No power may be given a preponderance over another power not provided for by the Constitution. No power may be deprived of the competences necessary for the fulfilment of its constitutional tasks.’

188  Ernest J. Weinrib standards of impartial decision-​making and judicial review that are appropriate to the task being performed.31 These three tasks are hierarchically ordered. By enacting legislation, the legislature determines what must be enforced and interpreted by the other institutions, thereby not only defining the ambit of executive and judicial activity but also confirming its own superordinate status. The existence of apparatuses for administration and adjudication, separate from the apparatus of legislation, certifies that what is being administered and adjudicated actually has the public status of legislation rather than being merely the personal command or the informal table-​talk of the ruler. Only the product of a legitimate legislative practice can command the attention of the officials who, separately from the legislators, exercise executive and adjudicative authority. By placing the enforcement and interpretation of any specific piece of legislation beyond the reach of the legislators, the separation of powers facilitates compliance with several of Fuller’s desiderata. First, it reinforces the desideratum of public promulgation, because it leads to an authoritative articulation of the legislative will that the other public institutions apply and interpret. Next, it institutionally entrenches Fuller’s desideratum of legislative generality through its abstract and impersonal formulation of the legislated norm. Conversely, it precludes the paradigmatic apprehension that, if legislative and executive authority were conflated, rulers would exempt themselves and their associates from having to obey the law because they would have the power, in Locke’s words, to ‘suit the Law, both in its making and execution, to their own private advantage’.32 Such exemptions, while formally leaving the law intact, would effectively strip it of its general character, and would enable the legal system to be used to target some and immunize others. Legislation would, in effect, change with particular cases and particular magistrates, so that it neither produces uniform effects over time nor remains aligned with official action. While legislation might retain its outward form as a general and stable expression of the legislative will with which official action must be congruent, it would be manipulated in ways that make a mockery of the reasons for valuing that outward form. The republican notion of the separation of powers is less concerned with such episodic Fullerian failings than with the institutional structure that would avoid them. By organizing its institutions to differentiate legislative from executive and judicial authority, a republican regime preserves the integrity of legislation’s claim to obedience. In overseeing the application of legislation, neither the executive nor the judiciary acts as a legislatures. Subjects obey legislation as the legislature

31 For examples from Canadian administrative law see Québec Inc v Quebec (Régie des permis d’alcool) [1996] 3 SCR 919; Ocean Port Hotel Ltd v British Columbia (General Manager, Liquor Control and Licensing Branch) [2001] 2 SCR 781. 32 J Locke, Two Treatises of Government (P Laslett ed, CUP 1988) 364.

The Rule of Law  189 created it; their obedience is to the legislation exclusively.33 They are obligated to do only what the law requires and are at liberty to do whatever the law does not require. In words that echo the rhetoric long associated with the rule of law, Kant observes that under the republican form of government, the posited law is ‘self-​ruling and depends on no particular person’.34 By this he means that the content of the law is controlled solely by the terms in which the law was enacted, rather than being responsive to the wishes of whoever oversees its operation. In contrast, in a despotism, where powers are not separated, the ruler can indulge in the ‘high-​handed management of the state by laws that he has himself given, inasmuch as he handles the public will as his private will’.35 Only in a republic is the rule of law possible, because only in a republic is the state institutionally organized to operate not only through law but in accordance with law.36 The separation of powers channels the exercise of state power through institutions that exhibit the specific kind of rationality and have the specific organizational structure appropriate to their distinctive tasks, whether of articulating a general norm, or of subsuming a particular situation under a general norm, or of adjudicating a dispute regarding a norm. These institutions are also governed by bodies of law that allocate jurisdictions and set out the procedures appropriate to the task allocated to each institution. State power is thereby not merely diffused among different institutions but is also subjected to the varieties of legal ordering that inform their operation. This is the sense in which the state acts in accordance with law. When these notions of state action through law and state action in accordance with law are put together, the picture that emerges is this: the rule of law deals with the constitutive ideas of a state organized in terms of the separation of legislative, executive, and adjudicative authority. Two such ideas come into play. The first idea, which distinguishes such a state—​and, indeed, any state—​from a condition of lawlessness, is that the state has the authority to impose obligations on (and to create other legal incidents that apply to) its subjects. The idea is constitutive in that, without such authority, there is neither state nor law. The rule of law, as encapsulated in Fuller’s desiderata, incorporates the conditions under which the state’s authority can be effective, so that the rules it generates can govern human conduct. Rules incapable of doing this are defective from a juridical standpoint. A pervasive non-​fulfilment of Fuller’s desiderata shows that the prescriptions or prohibitions in question are the product not of a legal system but of lawless imposition; a more 33 M Troper, ‘The Development of the Notion of Separation of Powers’ (1992) 26 Israel LR 1, 6–​7. 34 Kant (n 22) 122 [6: 341]. I have amended Gregor’s translation of ‘selbstherrschend’ to bring out its reflexive character. 35 Kant (n 21). 36 This indicates why Fuller’s otherwise admirable exposition is inadequate as an account of the rule of law. His parable about the travails of Rex proceeds on the assumption that if only Rex were sufficiently reflective, he could introduce the right kind of rules. There is no focus on the institutional structure that would be necessary to secure the rule of law. Rex is both legislator and judge.

190  Ernest J. Weinrib sporadic non-​fulfilment shows that particular practices do not live up to what legality requires. The second idea—​this one distinguishing such a state from a condition of despotism—​is that the separation of legislative, executive, and judicial authority is made effective through the state’s institutions. This idea too is constitutive, because this separation is what makes the state the kind of state it is, a republic. Only a republic provides the articulated schema of institutional authority that enables a state not only to rule its subjects through law but also itself to submit to law. The conclusion from these ideas, formulated abstractly, is that the rule of law takes the constitutive aspects of a republican polity as its subject matter. Courts and commentators make the rule of law more determinate by specifying the content of these constitutive aspects and by working out what is needed to preserve their integrity and effectiveness. Such elaboration spells out, for instance, what counts for legal purposes as retroactivity or comprehensibility or the avoidance of contradiction, and so on.37 It also spells out the institutional arrangements (for instance, about regular, impartial, and duly authorized administration, or about the integrity and independence of the judiciary, or about access to the courts)38 that are necessary if the various kinds of authority are to be kept separate, function fairly, and maintain the legal order as an institutionally specialized and coordinated whole.

3.  Rule of Law and Rechtsstaat The close connection between the common law idea of the rule of law and the German idea of the Rechtsstaat has frequently been noticed, even to the extent of the former serving to translate the latter. Some scholars have assimilated the two. At one extreme is the suggestion that the rule of law should be understood expansively to be composed of everything contained in the Rechtsstaat, including the supposed requirement of state adherence to a coherent and consistent scheme of justice.39 At the other extreme is the suggestion that the Rechtsstaat should be understood restrictively to include no more than Fuller’s version of the rule of law.40 In this section I want to comment on the Rechtsstaat’s connection with the

37 Bumke and Voßkuhle (n 28) 347–​69. 38 Canadian instances of each of these are: Roncarelli v Duplessis [1959] SCR 121; Reference re Provincial Court Judges [1997] 3 SCR 3; Trial Lawyers Association of British Columbia v British Columbia (Attorney General) [2014] 3 SCR 31. 39 TRS Allan, ‘The Rule of Law’ in D Dyzenhaus and M Thorburn (eds), Philosophical Foundations of Constitutional Law (OUP 2016) 201, 204; Allen (n 13) 88–​132, especially 128–​32. The notion that the Rechtsstaat includes justice has been contested. One thinks of the famous lament of the East German activist Bärbel Bohley after German reunification: ‘Wir wollten Gerechtigkeit und bekamen den Rechtsstaat’ (We wanted justice and got the Rechtsstaat); I von Münch, ‘Rechtsstaat Versus Gerechtigkeit?’ (1994) 33 Der Staat 165. 40 DN MacCormick, ‘Der Rechtsstaat und die rule of law’ (1984) 39 JZ 65.

The Rule of Law  191 rule of law in the light of my contention that the rule of law takes as its subject matter the constitutive aspects of a republican state. As is well known, historically the Rechtsstaat and the rule of law arose independently in response to the different constitutional circumstances of Germany and England.41 In Germany, the Rechtsstaat signifies three related things: a theoretical conception of the relationship between law and the state, an historical construct that had its own specific development,42 and a constitutional principle explicit in positive law.43 The starting point in Germany (and in continental Europe more generally) was the conception of the state as a Polizeistaat, in which an absolute monarch, holding a monopoly of legitimate coercion, benevolently governed for what he in his discretion determined was the welfare of his subjects. The Rechtsstaat originated in reaction to this conception of the state, offering a new conception in which the value of individual freedom modulated and transmuted the state’s authoritarianism. In contemporary Germany the significance of individual freedom became more pronounced after the idea of the Rechtsstaat was intertwined with the conception of the post-​war constitution as the articulation of a system of values based on human dignity.44 The characterization of the Rechtsstaat as ‘material’ (concerned with substantive principles) and no longer merely as ‘formal’ (concerned with structural and procedural issues) was sealed by the infusion of constitutional values into the idea of the Rechtsstaat. In England, in contrast, what was primary was not the conception of the state but the dominant role of the law. Monarchical absolutism (or as Dicey, writing from an English perspective, termed it, ‘royal lawlessness’)45 was excluded because even the king, although not subject to any other person, was held to be subject to the law—​‘Quod rex non debet esse sub homine, sed sub deo et lege’ as Chief Justice Coke, citing Bracton, famously told King James I.46 Instead, parliamentary supremacy developed within a constitutional order based on law as administered by the courts 41 B Sordi, ‘Révolution, Rechtsstaat, and the Rule of Law: Historical Reflections on the Emergence and Development of Administrative Law’ in S Rose-​Ackerman, PL Lindseth, and B Emerson, Comparative Administrative Law (2nd edn, Edward Elgar Publishing 2017) 23–​37; LC Blaau, ‘The Rechtsstaat Idea Compared with the Rule of Law as a Paradigm for Protecting Rights’ (1990) 107 South African LJ 76, 88–​92; HJ Berman, ‘The Rule of Law and the Law-​Based State (Rechtsstaat)’ in DD Barry (ed), Toward the ‘Rule of Law’ in Russia (ME Sharpe 1992) 43; MacCormick (n 40). 42 E-​W Böckenförde, ‘The Origin and Development of the Concept of the Rechtsstaat’ in E-​W Böckenförde, State, Society and Liberty: Studies in Political Theory and Constitutional Law (Berg 1991) 47–​70; G Gozzi, ‘Rechtsstaat and Individual Rights in German Constitutional History’ in P Costa and D Zolo (eds), The Rule of Law: History, Theory, and Criticism (Springer 2007) 237–​59; P Tiedemann, ‘The Rechtsstaat-​Principle in Germany: The Development from the Beginning until Now’ in JR Silkenat, JE Hickey Jr, and PD Barenboim (eds), The Legal Doctrines of the Rule of Law and the Legal State (Rechtsstaat) (Springer 2014) 171. 43 GG, art 28(1). 44 Tiedemann (n 42) 184. 45 AV Dicey, An Introduction to the Study of the Law of the Constitution (8th edn, Macmillan 1915)112. 46 Prohibitions del Roy (1607) 77 ER 132. See generally JP Reid, Rule of Law: The Jurisprudence of Liberty in the Seventeenth and Eighteenth Centuries (Northern Illinois UP 2004).

192  Ernest J. Weinrib in accordance with long-​standing legal practice. This primacy of law was held to imply the three features that Dicey identified with the rule of law: that punishment could be consequent only on a duly found breach of the law, that the law applied equally to everyone, and that courts determine the rights and liberties of private persons in particular cases.47 Without reconceptualizing the state, these elements of the rule of law (as Dicey pointed out) prevented imprisonments without legal cause, the exemptions from law enjoyed by officials, and the need for constitutional revision through express declarations of rights that marked the development of law in continental Europe. Although they have common elements, the Rechtsstaat and the rule of law differ in that the former refers to a kind of state (a state committed to Recht) and the latter refers to a role for law (law rather than something else is what rules). What accounts for the common elements is that the contemporary German Rechtsstaat is an instance—​indeed, one of the modern world’s pre-​eminent instances—​of a republic in the Kantian sense. The congruence of Rechtsstaat and Kantian republic is hardly surprising. Although he did not use the term ‘Rechtsstaat’, Kant’s account of the ‘condition in which the state’s constitution conforms most fully to the principles of right (Recht)’48 was decisive in the formation of the concept.49 In other words, the link suggested in the previous section between the kind of state, on the one hand, and the kind of strictures identified with the rule of law, on the other, is present in the relationship between the Rechtsstaat and the rule of law. The English rule of law requires a Kantian republic, and the German Rechtsstaat provides a conspicuous example of one. To gauge the extent of the overlap between the rule of law and the Rechtsstaat, one should notice that the Kantian conception of a republic has two aspects. One is its defining constitutive aspect: a republic is the form of government in which legislative, executive, and judicial authority are institutionally separated from one another. This is what distinguishes a republic from the alternative form of government, a despotism. The other is the regulative aspect that consists in the state’s ‘making freedom the principle and indeed the condition for any exercise of coercion’.50 The extent to which the state realizes freedom for its citizens does not make it a republic, but rather supplies the standard for assessing its adequacy as a republic.51 A complete and perfect freedom for all may be unachievable, but the state is obligated to bring itself into the closest possible approximation to that condition. 47 Dicey (n 45) 110–​16. 48 Kant (n 22) 103 [6:318]. 49 Böckenförde (n 42) 50; Gozzi (n 42) 255. 50 Kant (n 22). 51 On the distinction between the constitutive and the regulative see J Weinrib, Dimensions of Dignity (CUP 2016) 61. To avoid confusion, let me make my use of this distinction clear: the rule of law operates as a regulative principle, in that it that imposes standards on state action, but the subject matter of the rule of law is the constitutive aspect of the state. It is this subject matter to which I refer when I say that the rule of law is tied to or deals with the constitutive.

The Rule of Law  193 That is a republic’s regulative principle. This regulative principle can be operational in a state’s public arrangements only when its institutional structure precludes the rulers from treating the state as the plaything of their private wills. Despotisms are incapable of being animated by freedom. The regulative principle can gain traction only within a polity constituted as a republic. These constitutive and regulative features of the Kantian republic are present in Germany’s Rechtsstaat. On the constitutive side, the German constitution explicitly mandates that state authority be exercised through the legislature, executive, and judiciary, with the first of these three bound by the constitutional order and the latter two bound by law and Recht.52 On the regulative side, all state authority is under a duty to respect and protect the inviolability of human dignity.53 Not only does the constitution expressly make human dignity its central value, but the Constitutional Court has held that the constitution imposes on the state both a negative duty not to infringe constitutional rights and a positive duty, derived from the rights as objective principles, to take steps to protect them when they are at risk.54 Human dignity—​both on its own and as concretized in the enumerated rights—​is thus the state’s supreme regulative value. As was succinctly observed many years ago: ‘In the Rechtsstaat of the Basic Law, the primacy of law (Recht) obtains in the sense of being bound not only to the law as such, but also to certain contents of the law.’55 This distinction between ‘law as such’ and ‘certain contents of the law’ tracks the distinction between what is constitutive of the structure of legal authority within a Kantian republic and what is regulative of its exercise in the realization of freedom. If the rule of law is tied only to the constitutive aspects of a Kantian republic, it cannot be co-​ extensive with the Rechtsstaat. Given its concern with the constitution’s regulative content, the Rechtsstaat has a wider ambit than the rule of law. What links the two is that, as a polity that adheres to an effective separation of powers, the Rechtsstaat exemplifies the form of government into which the strictures of the rule of law fit. At the root of both the Rechtsstaat and the rule of law is the idea of human dignity. Fuller himself, in developing his notion of the law’s inner morality, commented that every departure from its principles is ‘an affront to man’s dignity as a responsible agent’.56 He then graphically continued: ‘To judge his actions by unpublished or retrospective laws, or to order him to do an act that is impossible, is to 52 GG, art 20(2)2 and (3). See D Currie, ‘Separation of Powers in the Federal Republic of Germany’ (2014) 9 German LJ 2113. 53 GG, art 1. 54 D Grimm, ‘The Protective Function of the State’ in G Nolte (ed), European and US Constitutionalism (CUP 2005) 137. 55 K Hesse, ‘Der Rechtsstaat im Verfassungssystem des Grundgesetzes’ in K Hesse, S Reicke, and U Scheuner (eds), Staatsverfassung und Kirchenordnung: Festgabe für Rudolf Smend (JCB Mohr 1962) 71, 77. This statement has become even more true in the decades since Hesse made it. 56 Fuller (n 14) 162.

194  Ernest J. Weinrib convey to him your indifference to his powers of self-​determination.’57 It may well be that, given the role that human dignity plays in each, one cannot consistently be committed to the rule of law without also being committed to the constitutional rights that provide the Rechtsstaat’s regulative content (although Fuller himself thought otherwise).58 But that is far from saying that this regulative content is part of the rule of law itself. But is the subject matter of the rule of law rightly restricted to the constitutive aspects of a Kantian republic? Can one not stipulate whatever meaning for those words that is most desirable? Surely (one might say) if the rule of law is a good thing, it should be given the broadest possible scope. And that counts in favour of including the regulative within its subject matter. This is difficult from the standpoint both of the Rechtsstaat and of the rule of law. The Rechtsstaat of contemporary Germany is closely associated with the post-​ war constitution, which through its entrenchment of human dignity supplies the regulative impulse for its content. For much of the previous century the consensus had been that the Rechtsstaat was ‘formal’ rather than ‘material’, focused on legislative supremacy and administrative disputes, and dealing with the manner in which the state realized its objectives rather than with the objectives themselves.59 In the absence of a constitutional bill of rights for the United Kingdom, the equating of the English rule of law with the modern Rechtsstaat requires adopting the German idea, while remaining disconnected from the kind of constitutional grounding that gives it its contemporary significance. From the standpoint of the rule of law, the paradox is that the rule of law is less serviceable the more expansively it is construed. One invokes the rule of law in order to assess whether state action satisfies the minimal conditions of legal acceptability. The more broadly one conceives of the rule of law, the more unclear the distinction between these minimal conditions and ‘a shorthand for an ideal state’.60 The point of the rule of law is lost if it does not serve as the platform of a distinctive basis for assessment.61 Stretching the scope of the rule of law negates the purpose of its own existence as a category of legal analysis. In contrast, viewing the rule of law as dealing with the constitutive aspects of a republican polity preserves for it a distinctive and coherent role. Moreover, by understanding the rule of law as having these constitutive aspects as its subject matter, one is spared from having recourse to adjectives that are either uninformative 57 ibid. 58 Weinrib (n 51) 258–​60. 59 Böckenförde (n 42) 53–​60; Gozzi (n 42) 245–​50. On the Rechtsstaat in the administrative context see KF Ledford, ‘Formalizing the Rule of Law in Prussia: The Supreme Administrative Law Court 1876-​ 1914’ (2004) 37 Central Eur Hist 203. 60 Bedner (n 15) 45. 61 Compare J Raz, The Authority of Law: Essays on Law and Philosophy (OUP 2011) 211: ‘[I]‌f the rule of law is the rule of the good law, then to explain its nature is to propound a complete social philosophy. But if so, the term lacks any useful function.’

The Rule of Law  195 (‘thick’ and ‘thin’) or elusive (‘formal’ and ‘material’).62 The reference to the constitutive enables the standard of legality to be set by what makes for a republican state; that is, for a kind of organization that is both a state and a state of a specific sort. The issues that are then addressed deal with the implications of the state’s authority, the practical possibility of complying with the law, and the separation of the three different tasks of state activity. These issues are conceptually prior to, and can be treated apart from, the regulative question of whether the state’s actions live up to a standard of goodness, whatever that might be.

4.  Injustice and Inhumanity Justice, ‘the first virtue of social institutions, as truth is of systems of thought’,63 is an idea that pertains to the regulative rather than the constitutive aspect of legal ordering. If it is the case that the rule of law has only the constitutive aspect as its subject matter, then the rule of law is never about justice or about the avoidance of injustice, however grave. In this section I want to deal with this seemingly unpalatable conclusion. The position that the rule of law is consistent with the greatest iniquity was emphatically rejected by Arthur Chaskalson, the former Chief Justice of South Africa. Reflecting on the legal framework of apartheid, he noted that apartheid was instituted and perpetuated in compliance with Fuller’s desiderata and through the collaboration of legislature, executive, and judiciary: The apartheid government, its officers and agents were accountable in accordance with the laws; the laws were clear, publicized and stable and were upheld by law enforcement officials and judges. What was missing was the substantive component of the rule of law. The process by which the laws were made was not fair (only whites, a minority of the population, had the vote). And the laws themselves were not fair. They institutionalised discrimination, vested broad discretionary powers in the executive, and failed to protect fundamental rights. Without a substantive content, there would be no answer to the criticism, sometimes voiced, that the rule of law is an ‘empty vessel into which any content can be poured’.64

Chaskalson was in effect denying what had been affirmed in a much-​cited article championing a ‘formal’ conception of the rule of law based on the law’s capacity to guide human action; understood in this formal way, the rule of law was said to be 62 On the perplexities of ‘formal’ in this context see J Gardner, Law as a Leap of Faith (OUP 2012) 195. 63 J Rawls, A Theory of Justice (rev edn, HUP 1999) 3. 64 A Chaskalson, Remarks at the World Justice Forum I (July 2018), quoted in M Ellis, ‘Toward a Common Ground Definition of the Rule of Law Incorporating Substantive Principles of Justice’ (2010) 72 U Pitt LR 191, 194–​95.

196  Ernest J. Weinrib compatible with gross violations of human rights, even to the extent of the law’s instituting slavery.65 In Chaskalson’s view, the experience of apartheid indicated that so contentless a conception of the rule of law could not be adequate. Is the view being put forward in this chapter, that the rule of law ignores injustice by dealing with constitutive rather than regulative aspects of the state, exposed to the same objection? To see why it is not, one needs to consider a distinction put forward by the distinguished interpreter of Kant’s legal philosophy, Julius Ebbinghaus. In his post-​war analysis of the juridical character of the Nazi regime, Ebbinghaus distinguished between state infringements of justice and state violations of the law of humanity.66 The former fall within the law and must be obeyed despite being unjust. The latter fall outside the law’s competence; far from requiring obedience or compliance, these are ‘juridically completely meaningless’.67 The conceptual background of this distinction lies in the Kantian idea that the state is necessary in order to give assurance that the creation, enforcement, and interpretation of legal norms are not the expression of anyone’s unilateral will. To observe norms in one’s interaction with others without the assurance of reciprocity on their part is incompatible with one’s innate right, specifically, with the aspect of innate right that Kant calls innate equality.68 This form of equality consists in one’s being bound to others only to the extent that they are in turn reciprocally bound. The rights and correlative obligations of the state of nature have the provisional capacity to bind, but they cannot actually bind in the absence of a mechanism to guarantee that all persons shall comport themselves as equally bound to the norms to which everyone else is bound. In the absence of such a mechanism, the non-​ reciprocated observance of interactional norms exhibits the inequality of a person exposing herself to being used as a means by others, while respecting their rights and treating them as ends in themselves. Put another way, the inconsistency with innate right is that by allowing oneself to be taken advantage of by others, a person fails to assert one’s worth, inherent in one’s innate right, as an independent being in relation to them.69 Hence, the state is the necessary mechanism for assuring the reciprocity of obligations. In this sense, the state exists to realize the innate right of interacting persons across the full range of their legal interactions. This right is innate in that it accrues to a person not through an act of acquisition, but as the immediate juridical concomitant of a person’s being. The innate 65 Raz (n 61) 211. 66 J Ebbinghaus, ‘The Law of Humanity and the Limits of State Power’ (1953) 10 Phil Q 14; M Wolff, ‘Julius Ebbinghaus, die Rechtliche Grenzen der Staatsgewalt und die Interpretation der Rechtlehre Kants’ in M Baum (ed), Staat der Freiheit (Steiner Verlag 2020) 1. 67 J Ebbinghaus, ‘Kants Rechtslehre und die Rechtsphilosophie des Neukantianismus’ in J Ebbinghaus, Gesammelte Schriften, Band 2, Philosophie der Freiheit, Praktische Philosophie 1955-​1972 (Bouvier Verlag 1988) 231, 244. 68 Kant (n 22) 34 [6:237]. 69 ibid 32 [6:236] (rightful honour).

The Rule of Law  197 right requires that a person who has not acted wrongly be regarded as ‘beyond reproach’70 and thus liable to no adverse legal determinations. One’s existence—​the very living of one’s life—​is not a wrong to anyone, nor is it a wrong to be who one is or to act in a way that does not injure anyone else’s rights. Because it deals with the juridical status of the person even apart from what the person has acquired, innate right comprises what Kant calls ‘the right to humanity in one’s own person’.71 From this, Ebbinghaus postulates a law of humanity that, by making innate right inviolable in the state’s treatment of interacting persons, sets the limit of the state’s authority. The purpose for the state’s existence defines the scope of its activity. Ebbinghaus’ analysis distinguishes between violations of the law of humanity—​ let us call them ‘inhumane orders’—​and injustice. Injustice occurs when an individual’s external freedom is restricted in a manner ‘that cannot be justified by the need to secure everyone against the arbitrary power of his fellow-​citizens’.72 (Ebbinghaus’ whimsical examples are laws that command the saluting of the governor’s hat or that prohibit the sniffing of violets.) A state restriction on liberty, even one that infringes the rights of subjects, counts as a mere injustice and not as an inhumane order ‘so long as the restriction imposed possesses the character of possible universality’. A general regulation of liberty, that is, one that applies equally to everyone, ‘does not annul anyone’s capacity to limit other people’s freedom by his own’. Such a regulation may be arbitrary by being unjustifiable in terms of the freedom of all, but because it is capable of universality through application to everyone ‘it does not go beyond the bounds of a possible harmony of liberty through law’. The Nazi measures against the Jews went beyond the limits of mere injustice so understood: ‘when Hitler poisoned the Jews, was this illegal only in the sense that it is illegal to condemn an accused man without giving him a fair hearing?’ The Nazi acts of genocide can be properly described not as injustices ‘but only as violations of humanity itself in the persons of the victims’. On Ebbinghaus’ analysis, the injustice of a law—​no matter how serious—​does not exempt anyone from complying with it. The point of the civil condition is to secure the reciprocal freedom of all in accordance with their innate equality. The state is necessary because the reciprocity of freedom is inconsistent with the exercise of any individual’s unilateral will in setting or enforcing or interpreting one’s right against another. In the exercise of its authority, the state enacts laws that determine, from its public and omnilateral standpoint, what justice requires. If an unjust determination by the state were grounds for non-​compliance with the law, the unilateral will would assume precisely the role that the state was established to pre-​empt. ‘Whoever declares that he will obey only the just expresses in reality his reluctance to obey at all.’73 The readmission of the prerogative will unilaterally to

70

ibid 34 [6:238]. ibid 36 [6:240]. 72 Ebbinghaus (n 66) 20. All the quotations in this paragraph are drawn from this page. 73 Ebbinghaus (n 66) 19. 71

198  Ernest J. Weinrib decide on its own compliance with the law would amount to the re-​establishment of the state of nature. Inhumane orders stand on a different footing. Inhumane orders annul a person’s capacity to interact with others in terms of their innate equality. They are, therefore, inconsistent with the point of state authority. This is the case ‘no matter how much it may have the formal characteristics of a law in accordance with the constitution of the state’.74 Unlike unjust laws, inhumane orders are not misapplications of state authority but non-​applications of it, because the state does not have the competence to violate the law of humanity. Nor are inhumane orders merely extreme instances of unjust laws, with the difference between them being a matter of degree. Rather, injustice and inhumanity are situated in different moral registers. An unjust law is a law that could under other circumstances be just.75 In contrast, an inhumane order could under no circumstances be a just law, as justice presupposes the possible universality of norms. Violations of humanity deny to the victims the capacity to assert claims to justice that are applicable to all. Inhumane orders, therefore, could never count as something that could implement justice. To bring out the ‘profound difference’76 that Ebbinghaus posits between injustice and inhumanity, it may be helpful to say that injustice stands to inhumanity as falsehood stands to nonsense: a false communication is a kind of communication—​a defective communication, to be sure, but a communication nonetheless—​whereas speaking nonsense is incompatible with the very idea of communication.77 A falsehood is something that might have been true but is not; nonsense is something to which truth does not apply. Similarly, an unjust law is a defective kind of law; an inhumane order is incompatible with the very idea of law. An inhumane order is a denial of humanity parading as law; it is a wolf in sheep’s clothing rather than a defective version of a sheep. This is why, as Ebbinghaus says, inhumane orders are ‘juridically completely meaningless’;78 that is, they are a kind of nonsense. Or to formulate this distinction in the terms developed in this chapter, an unjust law is a failure to carry out law’s regulative idea, whereas an inhumane order suffers from what is juridically a constitutive inadequacy that results in no law being made—​ producing what Ebbinghaus called a ‘pseudo-​law’.79 Consequently, Ebbinghaus concluded that inhumane orders ‘pass far beyond the limits of mere injustice’ because they violate humanity itself in the person of 74 J Ebbinghaus, ‘Positivismus: Recht der Menschheit—​ Naturrecht—​ Staatsbürgerrecht’ in J Ebbinghaus, Gesammelte Schriften, Band 1, Sittlichkeit und Recht, Praktische Philosophie 1929-​1954 (Bouvier Verlag 1986) 363. 75 Ebbinghaus (n 66) 20: ‘Under special circumstances, the requirements of general security may even forbid us to sniff the perfume of violets.’ 76 ibid 21. 77 This analogy wades into philosophically deep water, into which I do not want to advance any further; see eg C Diamond, ‘What Nonsense Might Be’ (1981) 56 Philosophy 5. 78 Ebbinghaus (n 67). 79 ibid 245.

The Rule of Law  199 the victims.80 This means that the victims have been denied the status to interact with others in reciprocal freedom. Even aside from mass murder, the restriction of property and contract rights on the basis of religion or racial origin deprives the victims of the possibility of acting legally in defence of their rights. ‘They are deprived, in principle, of any power at all to restrict the liberty of others in such a way that they could live among them.’81 Such a supposed law divests them of their condition as human beings and reduces them to the status of objects in their relations with others. ‘The law informs them, so to speak, that having no rights they cannot even be treated unjustly; and this precisely is what state power can never say to any of its subjects without exceeding the competence allowed it by the law of humanity.’82 The state cannot rightfully demand compliance with an inhumane order. Because that order does not conform to the minimal juridical condition for legality, it cannot emanate from the state’s legal authority. The person who issues it ‘must always be acting in his private capacity, even though he uses the irresistible power he wields to enforce his will upon his subordinate’.83 Within this approach, racist legislation, such as the legislation establishing apartheid, is paradigmatic of inhumanity. Of course, such legislation does not recognize the equal worth of all persons as bearers of an innate right. Instead, it assigns to a group a status that subordinates its members, affecting them not incidentally or sporadically or episodically, but in a way that comprehensively applies (and is calculated to apply) to all of them for the entirety of their lives. The comprehensiveness of their subordination signals the effort of state action to annul the significance of their innate right as persons. More specifically, racist legislation is inconsistent with the two aspects of innate right mentioned earlier in this section. First, it fails to treat persons who have not acted wrongly as being beyond reproach. By imposing adverse legal consequences on a person’s having a characteristic that no-​one can choose to have or not to have, it deliberately and systematically marks out the members of a particular group for who they are rather than for what they have done. In effect, it punishes a person for existing rather than for having committed a wrongful act. Secondly, it violates the innate equality of all persons by imposing disabilities that make this adventitious characteristic the basis of the rights, duties, and immunities that govern legal interactions. It may, for instance (as it did in Nazi Germany and as Ebbinghaus emphasized), impair their property or contract rights or disadvantage them in family disputes. This is inconsistent with the reciprocity of legal relations, because it deprives members of the victimized group of the possibility of obligating others in the same way that they are obligated.84 80 Ebbinghaus (n 66) 20. 81 ibid. 82 ibid. 83 ibid 22. 84 J Ebbinghaus, ‘Der Kantische System der Rechte des Menschen und Bürgers in seine Geschichtlichen und Aktuellen Bedeutung’ in J Ebbinghaus, Gesammelte Schriften, Band 2, Philosophie der Freiheit, Praktische Philosophie 1955-​1972 (Bouvier Verlag 1988) 249, 270.

200  Ernest J. Weinrib From Ebbinghaus’ account one can conclude that racist laws such as those that established apartheid in South Africa are not in conformity with the rule of law. This is the case, as Chaskalson suggested, even if Fuller’s desiderata of clarity, publicity, stability, and so on have been scrupulously satisfied. Those desiderata require that compliance with the obligations imposed by law be, as a practical matter, neither impossible nor futile. They take as their subject matter a constitutive feature of law, that law can create obligations that bind those who are subject to it to perform the actions that it mandates. However, the law’s power to create obligations is dependent upon an equally constitutive feature of law (and one that is indeed correlative to that power), namely that legal obligations result from the exercise of valid authority. From the juridical standpoint, this authority is what is missing in the enactment of racist laws, because from that standpoint no state has the authority to order a violation of anyone’s humanity. The Fullerian desiderata that Chaskalson mentioned bite on legal obligation, but legislation that is juridically devoid of authority generates no such obligation. On this analysis, apartheid was not an injustice; it was worse, a system of oppression that did not rise even to the level of an injustice. Being what Ebbinghaus termed a violation of humanity itself in the persons of the victims, it did not cross the threshold of being an emanation—​even an unjust emanation—​of legality at all.85

5.  Private Law Adjudication From these general reflections about the rule of law and the horrors of Nazism and apartheid, I return now to the more quotidian circumstances of private law. What is the significance of the rule of law as it pertains to private law? In the absence of codification or statutory interventions, the common law governs the legal relations between persons through the adjudicative process. This process, like all exercises of official power, is subject to the rule of law. Being concerned with constitutive aspects only, the rule of law focuses on adjudication as an open and public institutional proceeding that is impartially and disinterestedly responsive to the parties’ proofs and arguments, that determines liability in accordance with the parties’ rights and liberties, and that engages in the reasoned and systematic elaboration of a jurisprudence that applies generally and equally to all similarly situated parties. As was suggested in section 2 of this chapter, the rule of law comprehends both Fuller’s desiderata and the institutional separation of the legislative, executive, and adjudicative tasks. This means that, in the context of private law, Fuller’s desiderata

85 The same reasoning supports the conclusion that prohibitions of slavery, genocide, and apartheid are rules having the character of jus cogens under international law. See JA Frowein, ‘Ius Cogens’ in R Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2013) (http://​opil.oup​law. com); M Ragazzi, The Concept of International Obligations Erga Omnes (OUP 1997) 118.

The Rule of Law  201 should be understood in a way that is consonant both with the nature of private law and with the specifically adjudicative process of elaborating the norms of private law. So far as the nature of private law is concerned, legal norms operate within a bipolar relationship that actualizes the reciprocal freedom of both parties. Because rights are the external manifestation of this freedom, injustice occurs when one party has, or has done, something that is inconsistent with a right of the other party. From the standpoint of corrective justice, the point of liability is to correct this injustice by requiring the defendant to restore what is rightfully the plaintiff ’s or its monetary equivalent. Because within this bipolar relationship the injustice either does or does not exist, the legal success of one of the parties entails the legal defeat of the other. Failure to find the defendant liable for what has been an injustice to the plaintiff allows the inconsistency with the plaintiff ’s right to persist; conversely, finding the defendant liable where there has been no injustice to the plaintiff creates an inconsistency with the defendant’s right. So far as the adjudicative process is concerned, adjudication and legislation differ in the mode of creation that brings the norms of each into existence. To put the point extravagantly, legislative norms and adjudicative norms have different cosmologies. In the case of legislation, unless the statute is codificatory, a statutory norm has no legal existence whatsoever until the legislature duly enacts it. A change in the tax rate, for instance, imposes an obligation to pay the requisite tax only when the legislation that authorizes it takes effect. Before that point, the obligation is non-​existent.86 Legislation is thus a kind of creation ex nihilo that, through the authority of the legislature, transforms a norm’s non-​existence into the specifically legal form of existence that consists in the norm’s validity. In contrast, in the case of private law adjudication, the norm emerges not ex nihilo, but out of legally pre-​existing materials. These include not only the precedents, principles, and standards out of which a court might shape its decision, but also, where close precedents are unavailable, analogies from related legal contexts, and normative abstractions of greater and lesser generality appropriate to the law’s existing procedures and organizing categories. ‘Newly declared legal rules are not made by judges out of whole cloth. They are drawn by search and by analogy from the body of already existing rules and the complex relations of mankind.’87 Moreover, the existent material of private law includes not only the rules that govern the interaction of parties with each other, but also the reasoning that elaborates the rules from these pre-​existing materials. And whereas the justifications that legislators offer for their legislation may in some circumstances be

86 BVerfGE 127, 1, C, II, b (Spekulationsfrist 2010): ‘Only with its proclamation . . . is a norm legally existent.’ 87 Z Chafee Jr, ‘Do Judges Make or Discover Law’ (1947) 91 Proc Amer Phil Soc 405, 420.

202  Ernest J. Weinrib relevant to the interpretation of legislative provisions,88 the reasoning that supports private law decisions is not merely an aid to the interpretation of specific holdings, but is legally existent material that itself both organizes the pre-​existing material for articulating the norm at hand and contributes to the elaboration of subsequent norms. One should notice that the contrast being suggested is not between the legislative creation of new law and the judicial discovery of what the law always was. That view of adjudication has rightly been derided as a fairy tale.89 Adjudication no less than legislation features the creation of norms. The contrast, rather, is between two conceptions of creation, one of which transforms something devoid of legal character into a legal norm through an act of the legislative will, the other of which reorders existing legal material into a new legal norm through a reasoned process of adjudication. Instead of differentiating the creation of law from its discovery, one might say that on the adjudicative side, ‘the judges make law out of what they discover’.90 These contrasts between legislation and private law adjudication make it implausible to suppose that Fuller’s desiderata operate the same way for both. The rule of law should be understood in a way that reflects the distinct kinds of law, whether legislatively or judicially created, to which it applies. From the standpoint of private law, the most problematic of Fuller’s desiderata is the requirement of prospectivity. Private law decisions routinely apply to antecedent conduct. Indeed, since the point of these decisions is to adjudicate disputes that have arisen between the litigating parties, what else could they apply to? Application to the case at hand occurs even when the court announces a new doctrine or overrules an old one. Innovative judicial decisions are thus routinely retroactive, in the sense that they apply a new norm to a past event as though it was the law at the time of that event.91 As Benjamin Zipursky has observed in connection with innovative tort doctrine: ‘Retroactivity of new duty recognition is . . . virtually built into the structure of appellate adjudication.’92 The consequence of the introduction of a new doctrine is that parties to litigation are held to a norm of which they had no specific notice when the impugned conduct occurred. Jeremy 88 See eg Pepper v Hart [1993] AC 593. For a survey of common law jurisdictions see S Beaulac, ‘Parliamentary Debates in Statutory Interpretation’ (1998) 43 McGill LJ 287. 89 Lord Reid, ‘The Judge as Lawmaker’ (1972) 12 J Soc Pub Teach L (NS) 22: ‘There was a time when it was thought almost indecent to suggest that judges make law—​they only declare it. Those with a taste for fairy tales seem to have thought that in some Aladdin’s Cave there is hidden the common law in all its splendour and that on a judge’s appointment there descends on him knowledge of the magic words Open Sesame. Bad decisions are given when the judge has muddled the password and the wrong door opens. But we do not believe in fairy tales anymore’. 90 Chafee (n 87). 91 On this conception of retroactivity see B Juratowitch, Retroactivity and the Common Law (Hart Publishing 2008) 26. 92 B Zipursky, ‘Torts and the Rule of Law’ in L Austin and D Klimchuk, Private Law and the Rule of Law (OUP 2014) 139, 146.

The Rule of Law  203 Bentham had harsh words for such judicial law-​making: ‘When your dog does anything you want to break him of, you wait till he does it, and then beat him for it. This is the way you make laws for your dog: and this is the way the judges make laws for you and me.’93 One much discussed suggestion for mitigating the retroactive effect of a newly announced adjudicated norm is for the courts to engage in what has been termed ‘prospective overruling’.94 Under prospective overruling, the new norm applies to future cases, but does not apply (or has only a limited application) to cases that already have or could have arisen. In the purest form of prospective overruling, the new norm does not apply to the very case in which it was announced, with the consequence that the plaintiff, whose initiative and expense led to the norm’s formulation, is unable to enjoy its benefits. Fuller himself noted the irony of this: the only reward received by the plaintiff who successfully litigated to correct the law is that ‘the now admittedly mistaken rule is applied against him’.95 Less stringent versions of prospective overruling avoid this oddity by making the new norm available to the plaintiff in the instant case but not to plaintiffs in other pending cases, or available to plaintiffs in pending cases but not to parties with the same cause of action who have not yet commenced proceedings. By allowing a court holding to apply to future cases while abridging its impact on present ones, prospective overruling attempts to make an adjudicated norm behave like a legislated one. One way of formulating the difficulty with this is that it ‘amount[s]‌to the judicial usurpation of the legislative function’, and that it thereby ‘robs a ruling of its essential authenticity as a judicial act’.96 Prospective overruling sacrifices justice as between the parties in current controversies for the sake of providing notice of the norm for controversies yet to come. Whereas traditionally the precedential effect of judgments extends the principle that determines the case at hand into the future, prospective overruling divorces the future effect of the new norm from its current application. In its purest form, it treats victorious plaintiffs as unwitting lobbyists for future change rather than as litigants asserting their own rights in the face of an injustice suffered at the defendant’s hand. Instead of correcting the injustice between the parties, it allows that injustice to persist in the present case, in order to give fair warning of its nature to those who might commit it in the future. In its less stringent versions prospective overruling allows the plaintiff in the instant case to recover under the newly announced norm, while ruling out recovery

93 J Bentham, ‘Truth versus Ashhurst’ in J Bowring (ed), The Works of Jeremy Bentham, vol 5 (William Tait 1843) 235. 94 Of the many discussions of this see especially ML Friedland, ‘Prospective and Retrospective Judicial Lawmaking’ (1974) 24 UTLJ 170; RJ Traynor, ‘Quo Vadis, Prospective Overruling: A Question of Judicial Responsibility’ (1977) 28 Hastings LJ 533); Juratowitch (n 91) 99–​218. 95 Fuller (n 14) 57. 96 Re Spectrum Plus [2005] AC 680, para 28 (HL).

204  Ernest J. Weinrib for others who currently are similarly situated. The less stringent versions, like the purest form, subordinate the correction of injustice between the parties to the blunting of the effect of newly announced norms. In the less stringent versions, present application is an adjudicative exception to the quasi-​legislative function of the new norm. The issue in the less stringent versions is the extent to which this exception is to be recognized. From the standpoint of justice between the parties, there is no principled answer to this question, because the cases are differentiated solely on the basis of a manifestly arbitrary consideration: the timing of the initiation or resolution of claims. The distinction between the plaintiff who has already secured judgment, the plaintiff whose case is pending, and the potential plaintiff who has a cause of action but has not yet commenced proceedings does not justify treating their otherwise identical cases as being governed by different norms. To the contrary, the employment of a less stringent version of prospective overruling in order to make judicial law-​making comply with the rule of law’s insistence on prospectivity would fall foul of the rule of law’s requirement of generality.97 Accordingly, because it fails to vindicate the rights of the victorious plaintiff and others who are similarly situated, prospective overruling is not a satisfactory response to the challenge of retroactivity in private law adjudication. Prospective overruling seems ill-​fitted to what adjudication is supposed to do. The aporias posed by prospective overruling, however, suggest that it is worth asking how serious the challenge of retroactivity is to begin with. For this question, the contrast between legislation and private law adjudication is significant. As noted earlier, legislation and adjudication differ in the way they create law, the former creating legal norms ex nihilo and the latter creating them out of pre-​ existing material. From this difference follows another difference with respect to the creation of a rule today that directs what should be done yesterday. When such law-​making occurs legislatively, there is no yesterday. Creation ex nihilo takes effect only from the time at which it occurs; the time antecedent to that point is simply the period during which the obligation was non-​existent. And since the obligation was non-​existent at the time, legislation can subsequently breathe no life into it for the period for which present action is too late. To adapt Fuller’s felicitous formulation to this context, legislation that retroactively imposes obligations ‘talks in blank prose’,98 because the past conduct that it requires can no longer be the object of action by the person it obligates. Given that the rule of law comprises the conditions under which the law can subject human conduct to the control of rules, legislation that retroactively imposes obligations is alien to it. The same does not apply to obligations that are created by adjudication. These obligations arise out of pre-​existing legal material, material that is applicable to transactions that occur in the period that precedes the judicial decision. During

97 98

Juratowitch (n 91) 212–​14, 225. Fuller (n 14) 53.

The Rule of Law  205 that period, the pre-​existing legal material is also publicly available to transacting parties and to their legal advisers. Of course, the parties and their advisers may not know the innovative way in which a court, drawing on arguments made before it, might use this material in the case at hand to create a new norm, but this possibility is nonetheless accessible to their professional estimation. Crucial to this estimation is the legal adviser’s knowledge of the history of the law’s development and of the techniques of legal argument—​themselves publicly knowable components of the legal material that pre-​exists the court’s announcement of its decision and that provides examples of transformative judgments. Even the judicial announcement of a new rule or standard ‘may represent a reasoned conclusion from familiar premises’.99 All transactions of private law take place, and are known to take place, within the matrix of this pre-​existing material. Similarly, all transactors are aware that in any controversy a court may rule in favour of their opponents. And all legal advisers are cognizant of the possibility (and have a professional responsibility to anticipate and advise as reasonable) that courts may distil from the pre-​existing legal materials novel arguments and novel rulings. Thus, although the court may announce a new norm that retroactively imposes an obligation, that obligation arises not out of a legal vacuum, but out of the intimations of legal material antecedently available to and binding on the parties. Unlike legislation, the judicial decision, although operating retroactively on the dispute at hand, does not direct the affected parties to obey a law with which compliance was temporally impossible because at the point of conduct there was nothing to comply with. Rather, the decision makes definitive what ought to and could have been done earlier on the basis of the law as it existed earlier. In this way, the retroactivity of innovative judicial decisions in announcing an obligation nunc pro tunc is consistent with the rule of law. If this is so, prospective overruling is the solution to a non-​existent problem. Central to this understanding of the rule of law in the context of private law is the role of legal argument. Legal argument, itself a pre-​existing practice in the adjudicative process, is the medium through which pre-​existing legal material is gathered and deployed in the production of the judicial decision. That decision and the norm that emerges from it in a given case cannot be viewed in isolation from the legal reasoning that supports them. In the private law context, the rule of law is, among other things, the rule of legal reasoning, applicable to the character of a court’s deliberation no less than to its conclusion. Conversely, if one focuses on the norm alone detached from its supporting reasoning, one might be tempted to view the norm as a variety of legislation. And then in the case of innovative adjudication one is confronted afresh by the mystery of creation ex nihilo and the perplexity of the new norm’s retroactive application to the parties to the litigation.100 99 LL Fuller, ‘The Forms and Limits of Adjudication’ (1978) 92 Harv LR 353, 392. 100 Legal argument is similarly salient in the related issue, pertinent to the rule of law, of when an enactment is unconstitutionally vague. Drawing on the jurisprudence of the European Court of Human

206  Ernest J. Weinrib In this light, one can now very briefly revisit the legal realist assertion with which this chapter began, that tort law (and by implication, private law more generally) is public law in disguise. When represented as something else in disguise, private law is inconsistent with the comprehensibility and publicness that the rule of law mandates. In setting out its reasons for judgment, a court engages in a law-​ making activity under the authority of the state that, like all such activity, is subject to the requirements of the rule of law. The reasoning through which courts support and elaborate the norms of private law is integral to the comprehensibility of those norms; it indicates a norm’s meaning and limits by exhibiting how the court weaves the pre-​existing material of private law into the fabric of its ruling. Moreover, the open and public presentation of the court’s reasoning is a conspicuous example of public promulgation that makes the law accessible and publicly knowable. The claim of disguise, in contrast, asserts that the professed reasons for judgment are counterfeits of the true determinants of liability. In effect, the realist adopts as an interpretive principle the notion that private law pervasively fails to comply with the rule of law. Given its radical implications for the legal order, this is an interpretive principle that should be regarded with scepticism.

6.  Non-​instrumental Law Towards the end of his great essay on the rule of law, Michael Oakeshott draws attention to the non-​instrumental character of a state envisaged as an association ruled exclusively by law.101 He had earlier in the essay defined the rule of law as a ‘mode of moral association exclusively in terms of the recognition of the authority of known, non-​instrumental rules (that is, laws) which impose obligations to subscribe to adverbial conditions in the performance of the self-​chosen actions of all who fall within their jurisdiction’.102 By ‘adverbial’ conditions, Oakeshott meant that they ‘relate to the how, the when, the where and to the circumstances of conduct, not to its substantive purposes or its consequences’,103 thereby leaving a Rights, the Supreme Court of Canada in R v Nova Scotia Pharmaceutical Society [1992] 2 SCR 606, 639 stated that: ‘It cannot be argued that an enactment can and must provide enough guidance to predict the legal consequences of any given course of conduct in advance. All it can do is enunciate some boundaries, which create an area of risk. But it is inherent to our legal system that some conduct will fall along the boundaries of the area of risk; no definite prediction can then be made. Guidance, not direction, of conduct is a more realistic objective . . . A vague provision does not provide an adequate basis for legal debate, that is, for reaching a conclusion as to its meaning by reasoned analysis applying legal criteria. It does not sufficiently delineate any area of risk, and thus can provide neither fair notice to the citizen nor a limitation of enforcement discretion. Such a provision . . . fails to give sufficient indications that could fuel a legal debate. It offers no grasp to the judiciary.’ 101 Oakeshott (n 20) 161. 102 ibid 136. 103 M Oakeshott, ‘Law’ in M Oakeshott, What Is History? and Other Essays (Imprint Academic 2004) 423, 424.

The Rule of Law  207 person free to engage in self-​chosen conduct. He goes on to observe that some have sought the virtue of such a state in a consequence to be enjoyed as an outcome of it, and that some have recommended it as being ‘instrumental to the achievement of “prosperity” understood as the maximum continuous satisfaction of the wants of the associates’. Refuting these suggestions only parenthetically, he continues: But the more discerning apologists (recognizing the inconsistency of attributing the virtue of a non-​instrumental mode of association to its propensity to produce, promote or even encourage a substantive condition of things) have suggested that its virtue is to promote a certain kind of ‘freedom’. But this is misleading . . . [T]‌his ‘freedom’ does not follow as a consequence of this mode of association; it is inherent in its character.

I want to conclude by applying the same observation to what I have presented in this and previous chapters.104 At this point, four considerations can be summarily stated. First, the non-​instrumental character of law as it has been presented here derives from the non-​instrumental character of reciprocal freedom and from the role of the state in actualizing that freedom. Reciprocal freedom entails that in governing the interactions both among persons and between persons and the state, the law must be consistent with the standing of all persons as ends in themselves, not as means. This is the imperative concomitant to what Kant termed each person’s innate right, to the acquired rights of private law that flow from the innate right, and to human dignity, the contemporary constitutional analogue to the innate right. Reciprocal freedom thus signifies a conception of interaction among persons that categorically precludes regarding the legal system as a device for treating them as means. Secondly, reciprocal freedom is a relational phenomenon. Its province is neither the freedom of any individual considered in isolation nor the freedom of all regarded as an aggregate of individuals, but rather the freedom that each individual has through the freedom of everyone in relation to everyone else. It cannot, therefore, consist in the unilateral exercise of will by any person or even by all persons. That version of freedom would merely be the ‘wild, lawless freedom’105 of the state of nature. For everyone to enjoy the reciprocity of reciprocal freedom, it must be confirmed and secured through an omnilateral, public, and systematic legal ordering. 104 I do this without intending to imply that the views presented in this book would all have been consonant with Oakeshott’s. Although he was sympathetic to the conception of corrective justice that I offer (as he generously wrote to me in a note sent decades ago when I was beginning to think along these lines), he would have been less receptive to what I say about distributive justice, the role of rights, or the decisive importance of Kant. 105 Kant (n 22) 101 [6:316].

208  Ernest J. Weinrib Thirdly, the role of the state is to establish and maintain through its legal system the conditions for the realization of reciprocal freedom. Accordingly, the law sets out the terms on which each person can act for self-​chosen purposes in a way that is compatible with capacity of others to act in turn for their own self-​chosen purposes. In private law this takes the form of the application of legal categories and principles that elaborate the compossibility of the freedom of all in their immediate interactions with each other. In public law, it takes the form of institutions of distributive justice that preserve reciprocal freedom in the face of threats to it that arise from the systemic operation of private law itself. And in constitutional law, it takes the form of enshrining the rights and freedoms that are the most salient specifications of human dignity. Fourthly, understood juridically rather than simply as a manifestation of state power, the law does not stand to reciprocal freedom as a means to an end. Law is the site of reciprocal freedom, and reciprocal freedom is law’s animating impulse. Only through the medium of law’s institutions and doctrines can reciprocal freedom maintain itself as something actual; and only by actualizing reciprocal freedom does the law treat persons with the dignity to which their standing as independent moral beings juridically entitles them. Thus, reciprocal freedom is immanent within, and exists by virtue of, the legal system, rather than being merely its outcome or consequence. Law, then, is not an instrument for producing reciprocal freedom, any more than the body is an instrument for producing life. Law is, rather, that through which reciprocal freedom has its being. And just as we live our lives with different degrees of bodily health and vigour, so too particular legal systems may realize reciprocal freedom to different degrees of adequacy and robustness. Because of this variability, just as living beings strive to maintain or improve their health, so reciprocal freedom is the normative principle that law must endeavour optimally to fulfil. In Oakeshott’s formulation, reciprocal freedom ‘does not follow as a consequence’ of legal ordering but is ‘inherent in its character’. This book has traced the peregrinations of reciprocal freedom, surveying its significance for the structure of the private law relationship, for rights and legal institutions, for distributive justice, for the relationship between private law and constitutional rights, and for the rule of law. Throughout, the argument has drawn attention to the abstractions pertinent to legal ordering and to the determination of those abstractions in the handling of substantive legal issues. It has treated law ‘not as an instrument to an independent blueprint for social life, but as a determining content to a more abstract form’.106 Amid the pluralism and instrumentalism of contemporary legal scholarship (noted in the book’s opening pages), this recourse to abstractions that unify legal discourse on the basis of a non-​instrumental conception of reciprocal freedom



106

M Stone, ‘Legal Positivism as an Idea about Morality’ (2011) 61 UTLJ 313, 337.

The Rule of Law  209 may seem questionable and even quaint. If so, it is worth attending to an observation of Julius Ebbinghaus, whose views have already loomed large in section 4 of this chapter. After explicating the significance of the Kantian law of humanity for the assessment of the Nazi orders, Ebbinghaus turned to his readers with the words: I hope you will agree that the theoretical considerations I have laid before you are not without practical significance. I often say to my students, ‘If you want to come to grips with the real moral issues of human life, you must go as deeply into abstractions as you can’. It is my deep conviction that at the bottom of the moral confusion in which we find the contemporary world lies the lack of a sufficiently abstract theory.107

This book’s exploration of reciprocal freedom, private law, and public right has been carried out in the spirit of these remarks.



107

Ebbinghaus (n 66) 22.

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.  References to notes are indicated by the note number (prefaced by a lower-​case letter ‘n’) following the page number.  abstraction and determination  2–​21, 23, 34, 71, 74–​75, 100–​1, 140, 208–​9 acquisition  acquired rights  31–​32, 34, 68–​69, 101, 102, 104, 107–​9, 116, 129–​30, 207 Kant’s theory of  61–​64, 68–​69, 98 ownership, relation to  61–​64, 66, 68–​69, 72n.10, 76 adjudication  corrective justice, and  6, 11–​12, 71, 110, 152 rule of law, and  200–​6 Aristotle  vii, 94–​95, 96, 147 assignment of contract rights  see public right Barak, Aharon  128–​29, 141, 171–​72, 175n.72, 175n.73 Bentham, Jeremy  202–​3 Blinkfüer case  161–​62 Canada Trust v Ontario Human Rights Commission  134–​35 Chaskalson, Arthur  195–​96 coherence  corrective justice conception of  9–​10, 11–​13 structural coherence and substantive coherence  12–​13 structure, relation to  8–​12 Coleman, Jules  3, 4 constitutional rights  balancing of  173–​80 corrective justice and  150–​51, 152, 160 distributive justice and  150–​51, 152, 157, 158 horizontality and  160–​67 human dignity and  117, 118, 122, 123, 124–​ 26, 127–​30, 131, 132, 140–​41, 169–​73, 174–​75, 176, 179–​80, 191, 193–​94, 208 illusory exercise of  160–​65 innate right and  127, 128–​29 negation of  167–​73 ownership and  65–​66 practical concordance and  175–​78

corrective justice  acquired rights, and  102 author’s exploration of  vii–​ix coherence, corrective justice conception of  9–​10, 11–​13 concept of  vii–​viii, x–​xi, 207n.104 constitutional rights, and  150–​51, 152, 160 correlative conception of  5, 7–​10, 26, 116, 118–​19 definition of  vii distributive justice, distinction from  vii, 94–​98, 99–​100, 110–​11, 116, 157–​58 entitlements and  34 fairness, and  8–​12 horizontality, and  118–​19, 140, 147, 150–​51, 152, 157–​58, 160, 174 innate right, and  97, 108–​9, 116, 157 innate rights, and  102 internal understanding of justificatory character of private law, as  14–​22 juridical aspect of private law, and  22–​25 legislation’s role in  110 liability, and  vii, 26, 118–​19, 174, 201 private law, and  vii–​viii, 94, 118–​19, 140, 152, 181–​82 public right, and  78–​79 reciprocal independence, and  100, 102, 104–​5, 106, 108–​9, 116 relational term, as  viii–​ix right to education, and  157–​58, 160 rights and  26–​29, 30, 31, 34, 41n.35, 46, 47, 112–​13, 201 rule of law, and  181–​82, 201, 207n.104 scope of  x–​xi social and economic rights, and  157–​58 source of internal understanding of justificatory character of private law, as  14–​18 structural conception of  1–​4, 5, 7–​10 tort law, and  115 wider legal order, in  94 will theory of rights, and  41–​42, 45–​46

212 Index Depue v Flatau  158–​59 dignity  see human dignity distributive justice  constitutional rights, and  150–​51, 152, 157, 158 corrective justice, distinction from  vii, 94–​98, 99–​100, 106, 108–​11, 115–​16, 157–​58 horizontality, and  179–​80 programmes of  107–​8, 127–​28 rule of law, and  181, 207n.104, 208 social and economic rights, and  159 wider legal order, in  94 Dworkin, Ronald  21–​22 Ebbighaus, Julius  24–​25n.37, 196–​200, 208–​9 fairness  corrective justice, and  8–​12 human dignity, and  108 structure, relation to  8–​12 Fuller, Lon  183–​85, 188, 189–​91, 193–​94, 195, 200–​1, 202, 203, 204 Gardner, John  3, 4 Governing Body of the Jumaa Musjid Primary School v Essay NO  153–​60 Grimm, Dieter  113–​14, 124 Hevra Kadisha of the Jerusalem Community v Kastenbaum  128–​29, 171–​73 horizontality  balancing of constitutional rights  173–​80 concept of  117–​23 constitutional rights, and  160–​67 corrective justice, and  118–​19, 140, 147, 150–​ 51, 152, 157–​58, 160, 174 determinacy function of  133–​35 development function of  135–​41 dignity function of  141–​48, 149, 153–​54 distinction between scope and operation  149–​51 distributive justice, and  179–​80 functions of  131–​32 human dignity, and  117, 118, 121–​22, 123, 124–​26, 127–​30, 131, 132, 139, 140–​48, 149, 151–​52, 153–​54, 155–​56, 169–​73, 174–​75, 176, 179–​80 innate right, and  121, 126–​31, 139, 142, 149, 151, 154–​55, 157, 172 negation of constitutional rights  167–​73 presuppositions of  123–​31 public right, and  155–​56 social and economic rights, and  151–​60 human dignity  acquired rights, and  207

balancing of rights, and  173, 174–​75, 176 constitutional rights, and  117, 118, 122, 123, 124–​26, 127–​30, 131, 132, 140–​41, 169–​73, 174–​75, 176, 179–​80, 191, 193–​ 94, 208 dignity function of horizontality  141–​48, 149, 153–​54 fairness, and  108 horizontality, and  117, 118, 121–​22, 123, 124–​26, 127–​30, 131, 132, 139, 140–​48, 149, 151–​52, 153–​54, 155–​56, 169–​73, 174–​75, 176, 179–​80 independence as  105, 113, 125 innate right, and  121, 127–​30, 207 liability, and  139 negation of rights, and  169–​73 private law, and  140–​41, 207 reciprocal freedom, and  208 reciprocal independence, and  151–​52 right to education, and  153–​54, 155–​56 rule of law, and  108, 191, 193–​94, 207, 208 social and economic rights, and  151–​52, 153–​ 54, 155–​56 social concept of  121–​22 independence  see reciprocal independence inducing breach of contract  80–​82 injustice and inhumanity, rule of law and  195–​200 innate right  acquired right, as  31–​32, 62, 126–​27 acquired rights, and  129–​30, 207 birthright, as  43, 126–​27, 154 civil condition, in  127–​28, 142 common law protections of  32 constitutional rights, and  127, 128–​29 corrective justice, and  97, 102, 108–​9, 116, 157 definition of  31, 55, 62, 68–​69, 121, 154, 196–​97 distributive justice, and  97, 108–​9, 116, 157 enforcement of  37–​38 freedom from constraint by another’s choice  31, 45n.50, 54, 56, 62, 68–​69, 100, 116, 121, 154 horizontality, and  121, 126–​31, 139, 142, 149, 151, 154–​55, 157, 172 human dignity, and  121, 126, 127–​30, 142, 149, 154, 172, 207 infringement of  54 injustice, and  196–​97, 199 innate equality, and  196 Kant’s legal theory, in  37–​38 see also Kant, Immanuel  legal relationships, and  54, 55 personhood, and  56

Index  213 private law rights, and  128–​29, 131, 157, 207 property right, and  54–​55 protections of, in common law  32 reciprocal freedom, and  37–​38 reciprocal independence, and  100–​3, 105–​6, 108–​9 relational term, as  100 right to education, and  154–​55, 157 right to use, and  54–​56 rule of law, and  196–​97, 199, 207 social and economic rights, and  151, 157 things do not have  56 usability of people and things, and  56–​61 interest theory of rights  see rights interpretation  18–​21 Jacob’s ladder  7 Kant, Immanuel  acquisition, on  61–​64, 68–​69, 98 breach of contract, on  81–​82 civil condition, on  68–​69, 70, 71, 92, 98, 141–​ 42, 155 entitlements, on  33–​41, 47, 68–​69, 100–​1, 127 freedom, on  141–​42, 186 government, on  186–​89, 192–​94 Grotius’s theory of property, and  53–​56, 65 Hohfeld’s theory of rights, and  ix, 35–​41, 47, 66, 94 human dignity, on  122, 125, 129–​30, 142, 149 humanity, on  44–​45 independence, on  100–​1, 105, 106, 112–​ 13, 125 innate equality, on  196 innate right, on  121, 126, 127–​30, 149, 197 justice, on  98, 99, 196–​97 market overt, on  76–​79 morality, on  92–​93 Nipperdey’s legal theory, and  125 ownership, on  57–​66, 68–​69, 98 positive law, on  151 private law, on  xi private right and public right, on  68–​69, 70, 72–​80, 81, 92–​93, 98, 128–​29, 142 reason, on  x–​xi relationship between use and exclusivity, on  48–​49, 53–​61, 68–​69 right to humanity, on  196–​97 rule of law, on  185 social and economic rights, on  151 state of nature, on  68–​70, 71, 98, 127–​28, 141–​42, 155 theory of rights  viii–​xi, 29–​33, 46, 66, 67, 94, 125 will theory of rights, and  41–​46

legality, positive and juridical aspects of private law  22–​25 Life Imprisonment case  184 Lüth case  117–​18, 122, 143–​44 Mephisto case  170–​71 Moses v MacFerlan  19–​20 mutual independence  see reciprocal independence non-​instrumental law, rule of law and  206–​9 nuisance, private  see public law Oakeshott, Michael  185, 206–​7, 208 ownership  acquisition, relation to  61–​64, 66, 68–​69, 72n.10, 76 Blackstone’s definition of right of property, and  60–​61, 64–​65 civil condition, and  66 constitutional rights, and  65–​66 exclusivity, relation to  47–​49, 52, 57–​58, 59, 60–​61, 64 Grotius’ theory of  49–​57, 65 independent right to use, existence of  54–​56 juridical relationship of persons through things 66 property law, in relation  65–​66 reciprocal freedom, and  65, 66 restrictions on  65–​66 rights, relation to  65–​66 state, relation to  65–​66 usability, relation to  48, 56–​61, 63–​64, 65, 68–​69, 87–​89, 90–​91, 92 use, relation to  47–​49, 60–​61, 64 Parabolantenne case  166–​67 positive law  Kant’s theory of  151 positive and juridical aspects of legality of private law  22–​25 Princess Soraya case  144–​48 privacy  137–​40, 142–​43, 144–​46, 174–​75, 176, 177–​78 private law  corrective justice, and  vii–​viii, 94, 118–​19, 140, 152, 181–​82 corrective justice as source of internal understanding of justificatory character of private law  14–​18 human dignity, and  140–​41, 207 interpretive and normative dimensions of internal understanding of justificatory character of private law  18–​22 Kant’s theory of  xi

214 Index private law (cont.) positive and juridical aspects of legality of private law  22–​25 public law in disguise, as  181–​82, 206 public right’s effect on  92–​93 rule of law, and  182, 200–​6 prospective overruling  203–​5 public right  assignment of contractual rights, and  82–​84 author’s exploration of  209 breach of contract, and  80–​82 concept of  ix determination of rights within civil condition, and  75–​79, 92 effect on private law  92–​93 horizontality, and  155–​56 omnilaterality of  70–​75, 185 ownership, and  ix private nuisance, and  89–​92 private right, transition from  67–​70 privilege to preserve property, and  84–​89 reciprocal freedom, and  ix reciprocal freedom as ‘final end of all’  186 rule of law, and  185, 186, 209 state as repository of  155–​56 systematicity of law, and  79–​92 Pufendorf, Samuel von  85–​89 Rechtsstaat  190–​95 republican form of government, rule of law and  185–​90 retroactivity of private law adjudication  202–​5 rights  acquired rights  31–​32, 34, 68–​69, 101, 102, 104, 107–​9, 116, 129–​30, 207 concept of  29–​33, 46 corrective justice, and  46 entitlements, and  30, 33–​41, 43, 46 interest theory of rights  41–​42, 46 progression from structure to rights  26–​29 subjective rights  30–​34, 37–​38, 41n.35, 46 will theory of rights  41–​46 see also constitutional rights; innate right; public right; social and economic rights 

rule of law  concept of  183 corrective justice, and  181–​82, 201, 207n.104 distributive justice, and  181, 207n.104, 208 functions of  183–​90 human dignity, and  108, 191, 193–​94, 207, 208 injustice and inhumanity, relation to  195–​200 innate right, and  196–​97, 199, 207 Kant’s theory of  185 non-​instrumental law, and  206–​9 private law, and  182, 200–​6 public right, and  185, 186, 209 Rechtsstaat, and  190–​95 republican form of government, and  185–​90 Simmonds, Nigel  32–​33 Snyder v Phelps  167–​73 Spryer of Zurich case  162–​67 Stone, Martin  15n.18, 24, 208 structure  abstraction, as  4–​8 coherence, relation to  8–​12 corrective justice as source of internal understanding of justificatory character of private law  14–​18 corrective justice as structural idea  1–​4 fairness, relation to  8–​12 interpretive and normative dimensions of internal understanding of justificatory character of private law  18–​22 positive and juridical aspects of legality of private law  22–​25 progression from structure to rights  26–​29 structural coherence and substantive coherence  12–​13 unity of a conceptual sequence  97–​100, 116 Vincent v Lake Erie  84–​85 will theory of rights  see rights Windscheid, Bernard  60–​61