Scholars traditionally conceptualize private law around a commitment to the values of formal freedom and equality. Critics of the traditional view (including lawyer-economists) dispute the significance of a distinction between public and private law, construing private law as merely one form of public regulation. Both positions are flawed. The traditional position is conceptually misguided and normatively disap­pointing; the critical position confuses a justified rejection of private law libertarianism with a wholesale dismissal of the idea of a private law, thus denying private law’s inherent value.

This Article seeks to break the impasse between these two positions by offering an innovative account of the values that should, and to some extent already do, underlie the law of interpersonal interactions among private individuals in a liberal state. Rather than succumbing to the unappealing adherence to formal freedom and equality, private law should openly embrace the liberal commitment to self-determination and substantive equality. A liberal private law establishes frameworks of respectful interaction conducive to self-determining individuals. These frameworks are indispensable for a society in which individuals recognize each other as genuinely free and equal agents.


    1. The Traditional Conception: Private Law as a Locus of Formal Freedom and Equality
    2. The Critical Account: Private Law as the Continuation of Public Law by Other Means
    1. The Relational Form of Private Law
    2. The Normative Contents of Private Law
      1. On Interdependence and Personal Difference
      2. The Conception of the Person of Private Law
      3. Toward a Novel Approach to the Problem of Poverty in Private Law Theory
      4. The Role of Law and the Limits of Interpersonal Accommodation
    3. The Complexities of the Public–Private Distinction
      1. On Formal Equality and Independence
      2. Internal Contextual Factors
      3. External Commitments
    1. Accidental Harm to Life and Limb
    2. Residential Dwellings and the Workplace
      1. Residential Dwellings
      2. Workplace Accommodation
    3. Joint Projects and Other Collaborative Endeavors
    4. Affirmative Interpersonal Duties
      1. Mistaken Payment
      2. Private Necessity
      3. Responsibility of Property-Right Holders




Private law—the law of our horizontal interactions—offers many in­strumental benefits to society: Property and contract law help us assign and reassign entitlements, while tort law helps allocate responsibilities regarding those entitlements. Nothing that follows is intended to imply that these instrumental functions are insignificant. Private ordering may well provide a useful means to achieving important social ends. Yet private law is valuable beyond its contingent, external benefits: It is intrinsically valuable. The intrinsic value of private law lies in its con­struction of frameworks of respectful interaction—of just relationships—among genuinely free and equal individuals.

Private law is not reducible to common law. Its promulgation and ad­judication is not confined to the chambers of judges. 1 Some see the distinction between public and private law as turning on the ques­tion of whether the legal doctrine at hand is legislative or judge made. A representative observation appears in Martha Chamallas & Sandra F. Sperino, Torts and Civil Rights Law: Migration and Conflict: Symposium Introduction, 75 Ohio St. L.J. 1021, 1021, 1024 (2014) (observing, in comparing tort law and antidiscrimination law, that “[t]he distance [be­tween the Civil Rights Act and tort law] likely reflects their placement on opposite sides of the public–private divide, with Title VII . . . forming part of public law, while torts is a clas­sic, private law subject”). We disagree with respect to both the general claim that the pub­lic–private distinction is reducible to the statutory–judge-made distinction, and the particular claim that the wrong of discrimination is not part of the core of tort law. While the choice of institutional design is relevant to the pub­lic–private distinction, it seems indisputable that legislators are authorized to promulgate private law doctrines and, moreover, that they can do so just as well as their peers from the court. That Title VII is a piece of legislation does not in itself make it public law; that the common law is often associated with judge-made law does not make it any less coercive or statist than a statute. Rather, private law—whether judge made or statutory—is the law that governs our inter­personal relationships as free and equal persons; it stands in contrast to public law, which governs our interactions as patients of the welfare state or as citizens of a democracy. 2 Put another way, private law establishes the rights and duties individuals have against one another; public law, in contrast, pertains to individuals’ rights and duties as citizens or vis-à-vis the collective state at large. This seemingly banal definition highlights both private law’s relational form and its normative value. In this Article we develop a distinctly liberal conception of private law, founded on the commitments to individual self-determination and substantive equality. We demonstrate that this conception indeed accounts for much of pri­vate law as we know it. To be sure, private law does not always uphold these values in full. In certain spheres of human interaction, it undersup­plies such frameworks; in others, it fails to meet the demands of just rela­tionships. But these flaws, rather than undermining our theory of private law, highlight its significance as a source of internal critique that can push private law to live up to its implicit normative promise of securing just relationships: relationships of reciprocal respect to people’s self-deter­mination and substantive equality.

Our conception of private law stands in contrast to two broad, competing conceptions. The traditional view—shared by libertarians, modern Kantians and Hegelians, and liberal egalitarians—understands private law as a realm of prepolitical or apolitical interactions. 3 See infra section I.A (presenting the traditional view of the public–private distinction). There are, of course, important divergences among these groups, but for our purposes only their common denominator matters. These traditionalists conceptualize private law as that part of our law that is re­sistant to demanding interpersonal claims. To be sure, the traditionalist schools diverge on the question of whether public law should be guided by a commitment to people’s self-determination and their substantive equality (or not, as libertarians advocate). They all agree, however, that such commitments should not affect private law. Instead, they argue that private law should be guided by individual independence rather than by individual self-determination and that it should adhere to formal rather than substantive equality. 4 Put another way, the traditionalists argue that private law should focus on a com­mitment to negative, rather than positive, liberty. The reasons given for detaching these values from private law vary. Some argue that these values have no place in private law because private law precedes our social contract. Others invoke the classic division-of-labor argument that the state and its citizens bear different kinds of legal or moral responsibili­ties. See infra section I.A.

The traditional view has been subject to harsh and relentless criti­cism from many quarters—from Marxists and legal realists to critical legal scholars, feminist legal scholars, and lawyer-economists. 5 See infra section I.B (discussing the critique of the public–private distinction). These critics differ in many of their convictions, but they raise a shared opposition to recognizing the very existence of a distinction between private law and public law. They point out that private law has certain distributive effects and inevitably relies on public value choices. More­over, they argue that adhering to the public–private distinction tends to obscure certain regres­sive or otherwise oppressive features of private law and thereby shields it from scrutiny. Indeed, many critics conclude that private law should be treated as just one form of regulation, indistin­guishable from other regulatory regimes in both ends and means. Any conceivable distinction between private law and public law, they argue, is entirely con­tingent, deriving at best from their distinctive instru­mental characteristics.

Our approach, like the critique of the traditional view, resists at­tempts to naturalize and depoliticize private law. Private law does rely on public choices that run counter to the traditionalists’ liberal division-of-labor arguments. Yet we reject the critics’ reductionist account of private law, which ignores or marginalizes its noncollectivistic intrinsic value—the value of just relationships. Since private law is the law of our horizon­tal interactions, its roles cannot be properly performed by any other legal field. Only private law can forge and sustain the variety of frameworks for interdependent interpersonal relationships that allow us to form and lead the conception of our lives. Only private law can cast these frame­works of relationships as interactions between free and equal individuals who respect each other for the persons they actually are and thereby vindicate our claims to relational justice from one another. In sum, while the traditional account of private law as the law of independence and formal equality certainly warrants descriptive, conceptual, and normative criticism, the critics’ obliteration of its unique horizontal nature is no less unsupportable.

To anticipate one takeaway of our approach, consider the issue of housing discrimination. 6 See infra section III.B.1 (discussing the implications of a just-relationships conception of private law in the context of housing antidiscrimination law). Securing equal opportunity in the housing mar­ket is indisputably a desirable legal end for society. But do landowners bear any freestanding responsibility on this front? Both traditionalists and critics imply that the answer is no: For both groups, the current prohibition on excluding buyers or renters on the basis of race (or some other immutable characteristic) is contingent in that it depends upon the availability of other state-driven means for securing fair equality of opportunity for discriminated groups and ensuring that residential areas are sufficiently integrative. Our approach demonstrates that this view must be wrong.

The purpose of this Article is to break the impasse between the tradi­tionalists and their critics with a novel conception of private law, one that offers a charitable understanding of its distinctiveness. Part I fleshes out the competing understandings of private law that dominate the current discourse. Against the theoretical deadlock they have generated, we clar­ify the decidedly nonlibertarian values ingrained in private law properly conceived. The crux of our approach, developed in Part II, is a recons­truction of the public–private distinction that recognizes the significance of our interpersonal relationships and thus captures the irre­ducible core of both the form and content of private law. Our account does not eliminate all public concerns from private law; rather, it refines the interpersonal concerns standing at the moral center of private law. We do not claim that there is an intrinsic value in the separation of pri­vate law from public law; rather, we claim that there is an intrinsic value in private law itself, one worth retaining. By the same token, our theory does not ignore the traditionalists’ emphasis on independence and formal equal­ity; rather, we properly construe those values as subordinate to self-determination and substantive equality, the normative commitments that animate private law. Finally, our approach disconnects the misleading association between private law and adjudication by distinguishing be­tween private law’s core feature—structuring just interpersonal interac­tions—and the contingent institutional means of its vindication. This conclusion is particularly important for contemporary private law given the institutional limitations of the judiciary in our increasingly complex, interconnected environment.

This nuanced approach makes the task of translating our normative theory into legal doctrine far from straightforward. 7 One complication not fully addressed in this Article is the contexts in which corporate or governmental bodies are involved in horizontal dealings. Our account cap­tures corporations inasmuch as they are duty-bound toward natural persons (e.g., in hiring decisions). Cases involving corporations on both sides of the interaction as well as cases involving duties on the part of natural persons toward corporations stand, for the time being, outside the scope of our account. Any ultimate conclusion in these matters must presuppose a theory of the corporation; this Article does not develop one. But none of these complexities undermine the theory’s significance. To the contrary, a co­herent theory of just relationships, premised on individual self-determination and substantive equality, is more loyal to private law as we know it than any other competing account is. 8 See infra Part III (discussing just relationships as applied to negligence law, hous­ing and workplace discrimination, joint projects, and affirmative interpersonal duties). Indeed, traditionalists and critics (including lawyer-economists) alike fail to explain certain fundamental features of private law, the law of just relationships.

Other lessons of our theory of just relationships go beyond the explanatory plane. We challenge traditionalists who seek refuge in (their constructed version of) the common law from some of the real chal­lenges facing the law governing our interpersonal interactions in crucial contexts like housing and the workplace. We also upset the troubling scholarly schism regarding regulatory schemes that govern interpersonal relationships, such as the law of workers’ compensation. The traditional­ists oust these schemes from their purview entirely, while the critics are not careful to ensure that they are relationally just. Finally, focusing on private law’s commitment to the ideal of just relationships is important not only when it elucidates and reaffirms existing doctrine but also when it demonstrates its failings. Thus, for example, we show that the prevalent doctrine that renders racially restrictive covenants unenforceable fails (notwithstanding its many virtues) to appreciate their most troubling fea­ture, which should on its own render them invalid. 9 See infra section III.B.1 (explaining why racially restrictive covenants are intrinsi­cally invalid under a just-relationships conception of private law). We also allude to the overly hesitant approach of contemporary private law toward affirmative interpersonal duties and thus anticipate some reform once released from its traditional libertarian conception.

I. A Misleading Dichotomy

The traditional conception of private law and the prevailing criti­cism of that conception implicitly share common ground. Both express dissatisfaction with the straightforward, seemingly banal understanding of the public–private distinction—namely, that public law “is the law that pertains to government . . . or to the vertical relation between the govern­ment and individuals,” while private law regulates horizontal dealings among the private parties subject to that political authority. 10 Michel Rosenfeld, Rethinking the Boundaries Between Public Law and Private Law for the Twenty First Century: An Introduction, 11 Int’l J. Const. L. 125, 125–26 (2013). The tradi­tional approach finds this characterization of private law morally vacu­ous; after all, it implies that “even the Soviet Union had a private law.” 11 Alan Brudner, The Unity of the Common Law 54 (2d ed. 2013). Accordingly, traditionalists instill into private law values that dissociate it entirely from politics (broadly defined)—namely, from any “common ends” or “member obligations.” 12 Id. at 353. By contrast, the critics warn of the in­sidious risks of thus entrenching a libertarian private law. 13 See infra section I.B (discussing the critique of the traditional view). Moreover, because they agree that a law of interpersonal interactions cannot stand for any particular moral value, critics tend to renounce the public–private distinction altogether and instead conceptualize private law as a set of regulatory strategies with no (even potentially) unique moral significance.

A. The Traditional Conception: Private Law as a Locus of Formal Freedom and Equality

Within the broader traditionalist camp, libertarians construe private law as a prepolitical order. On this conception, private law is typified as a regime of strong property rights that both sets the boundaries of pro­tected domains and establishes strict rules for identifying valid transfer of entitlements. 14 See, e.g., Robert Nozick, Anarchy, State, and Utopia 160 (1974) (describing the robustness of owners’ rights); Randy E. Barnett, A Consent Theory of Contract, 86 Colum. L. Rev. 269, 270, 302–03 (1986) (describing the importance of strict rules in entitlement theory). This is unsurprising; such an understanding of private law—as governed by the ideal of people relating as formally free and equal persons—is foundational to the libertarian project. Libertarians typically conceptualize private law entitlements as the prepolitical base­lines for our social contract and, therefore, the bounds of its legitimate demands. 15 See Nozick, supra note 14, at 149. Hence, the three principles, which Robert Nozick famously asserted as the only principles the legal regime in its entirety should up­hold, correspond roughly to the three main branches of private law: the principle of acquisition of holdings (property), the principle of transfer of holdings (contract), and the principle of rectification of violations of the first two principles (tort). 16 Id. at 150–53; see also Alon Harel, Public and Private Law, in The Oxford Handbook of Criminal Law 1040, 1045 (Markus Dubber & Tatjana Hörnle eds., 2014).

More interesting, however, is the liberal-egalitarian canonical posi­tion. 17 See infra notes 18–22 and accompanying text. Liberals denounce the libertarian minimal state while endorsing—based on the traditional public–private distinction—a libertarian concep­tion of private law. Liberals insist that justice requires that the state go beyond the libertarian normative commitments to independence (i.e., negative liberty) and formal equality. But they assign to the state the sole responsibility for any additional positive obligations to facilitate individ­ual self-determination and ensure substantive equality. As such, liberals impose limited (if any) responsibility on individuals to engage with other individuals on terms that exceed formal equality and freedom. This idea of an institutional division of labor is the conventional foundation of the public–private distinction. 18 Lawyer-economists typically share this position for very different reasons. See, e.g., Louis Kaplow & Steven Shavell, Why the Legal System Is Less Efficient than the Income Tax in Redistributing Income, 23 J. Legal Stud. 667, 667–68 (1994) (arguing that institutional division is necessary to promote efficiency). But see Daphna Lewinsohn-Zamir, In Defense of Redistribution Through Private Law, 91 Minn. L. Rev. 326, 329–32 (2006) (arguing that efficiency is not the sole method of evaluating private law); Tsilly Dagan, Pay as You Wish: Globalization, Forum Shopping, and Distributive Justice 4 (June 10, 2014) (unpublished manuscript), http://ssrn.com/abstract=2457212 (on file with the Columbia Law Review) (“[T]he superiority of tax rules as a mechanism for redistributing income, which is convincing, perhaps indisputable, in a closed economy, is not self-evident in the conditions of today’s global economy.”).

A well-ingrained notion in liberal-egalitarian thought is that the state’s responsibility to ensure fair equality of opportunity is sufficient for realizing substantive equality and freedom. 19 Some liberal philosophers, however, seem to dissent from this conventional wis­dom. See, e.g., Samuel Scheffler, Equality and Tradition: Questions of Value in Moral and Political Theory 107–28 (2010) [hereinafter Scheffler, Equality and Tradition]. For a discussion of the connection between these views and our theory, see generally Hanoch Dagan & Avihay Dorfman, Justice in Private 6–9 (Nov. 25, 2015) (unpublished manu­script), http://ssrn.com/abstract=2463537 (on file with the Columbia Law Review). John Rawls, for example, argues that whereas state institutions, such as the tax system, enforce rules of distribution, private law institutions are supposed “to leave individuals and associations free to act effectively in pursuit of their ends and without excessive constraints . . . secure in the knowledge that else­where in the social system the necessary corrections to preserve back­ground justice are being made.” 20 John Rawls, Political Liberalism 268–69 (1993) [hereinafter Rawls, Political Liberalism]; see also John Rawls, Social Unity and Primary Goods, in John Rawls: Collected Papers 359, 371 (Samuel Freeman ed., 1999) (describing the “social division of responsibility” between citizens and the state). Ronald Dworkin similarly observes that equality is the sovereign’s virtue and that individuals do not have a “gen­eral duty to treat all other members of [their] community with equal care and concern.” 21 Ronald Dworkin, Law’s Empire 296 (1986) [hereinafter Dworkin, Law’s Empire]. He then compares the liberal-egalitarian and libertarian conceptions of equality, concluding that “[t]hough these two theories are very different from each other,” they are similar insofar as they do not apply the basic ideal of equality—that is, substantive equality—to the conduct of private individuals. 22 Id. at 299.

Thus, orthodox contemporary liberal political philosophy sees a convergence between the institutional division of labor and the moral division of labor. 23 Whether Rawls fully absolves individuals from responsibility to realize substantive equality and freedom has been contested. See Samuel Scheffler, Distributive Justice, the Basic Structure and the Place of Private Law, 35 Oxford J. Legal Stud. 213, 227–29 (2015). On one interpretation, this division of labor is principled—by securing background jus­tice, the state allows private persons to exercise their power to set and pursue their own conceptions of the good. See, e.g., Arthur Ripstein, The Division of Responsibility and the Law of Tort, 72 Fordham L. Rev. 1811, 1813 (2004) [hereinafter Ripstein, Division of Responsibility].
On another reading, the institutional division of labor derives from pragmatic concerns—the difficulty of evaluating the aggregate distributive effects of many types of interpersonal interactions—and is therefore contingent. See, e.g., Ronald Dworkin, Sovereign Virtue 117 (2000) [hereinafter Dworkin, Sovereign Virtue]; Samuel Freeman, Private Law and Rawls’s Principles of Justice 27 (2014) (unpublished manuscript) (on file with the Columbia Law Review). But even under this latter interpretation, the responsibilities that the demands of distributive justice may impose on individuals are not interpersonal. Rather, these responsibilities are a means for complying with the demands of belonging to a collectivity. They are thus contingent on a comparative assessment of the private and public supplies of collective responsibilities.
The most sophisticated and extensive articulation of this idea is in recent elaborations of Immanuel Kant’s conception of pri­vate law. 24 See, e.g., Arthur Ripstein, Private Wrongs 37 (2016) [hereinafter Ripstein, Private Wrongs]; Ernest J. Weinrib, Corrective Justice 263–96 (2012) [hereinafter Weinrib, Corrective Justice]. Although the above accounts diverge on many counts, the differences, as well as certain controversies as to these interpretations of Kant, are immaterial for pre­sent purposes. Although many details of the Kantian account are controver­sial, its core understanding of private law as a locus of personal independence and formal equality nicely echoes the mainstream liberal position—including that taken by liberal private law theorists 25 For the views of tort theorists, see, e.g., Ernest J. Weinrib, Correlativity, Personality, and the Emerging Consensus on Corrective Justice, 2 Theoretical Inquiries L. 107, 126–48 (2001) (discussing the positions held by Jules Coleman, Stephen Perry, and Arthur Ripstein).
For the views of property theorists, see, e.g., Thomas W. Merrill & Henry E. Smith, The Morality of Property, 48 Wm. & Mary L. Rev. 1849, 1882–83, 1895 (2007); Thomas W. Merrill, Property as Modularity, 125 Harv. L. Rev. Forum 151, 157–58 (2012) [hereinafter Merrill, Property as Modularity].
For the views of contract theorists, see, e.g., Daniel Markovits, Promise as an Arm’s Length Relation, in Promises and Agreements: Philosophical Essays 295, 307–12 (Hanoch Sheinman ed., 2011) [hereinafter Markovits, Arm’s Length Relation]; Seana Valentine Shiffrin, Paternalism, Unconscionability Doctrine, and Accommodation, 29 Phil. & Pub. Aff. 205, 227–30 (2000) [hereinafter Shiffrin, Paternalism].
—and thus illustrates well its implications.

Kant’s theory of private law builds exclusively on one underlying ideal: freedom cast negatively in terms of “independence from being constrained by another’s choice.” 26 Immanuel Kant, The Metaphysics of Morals 63 (Mary Gregor trans., 1991); see also Arthur Ripstein, Force and Freedom: Kant’s Legal and Political Theory 35 (2009) [hereinafter Ripstein, Force and Freedom]; Ernest J. Weinrib, The Jurisprudence of Corrective Justice 6 (2015) (unpublished manuscript) (on file with the Columbia Law Review). Independence implies that “each person is enti­tled to be his or her own master . . . in the contrastive sense of not being subordinated to the choice of any other particular person.” 27 Ripstein, Force and Freedom, supra note 26, at 4. Accord­ingly, independence requires that no one gets to tell you what purposes to pursue and is therefore “not compromised if others decline to accom­modate you.” 28 Id. at 14–45. Quite the contrary: “Because the fair terms of a bilateral interaction cannot be set on a unilateral basis, considerations whose justificatory force extends only to one party are inadmissible.” 29 Weinrib, Corrective Justice, supra note 24, at 36. For variations on this theme, see Jules Coleman & Arthur Ripstein, Mischief and Misfortune, 41 McGill L.J. 91, 109, 112 (1995) [hereinafter Coleman & Ripstein, Mischief and Misfortune]; Arthur Ripstein, Civil Recourse and Separation of Wrongs and Remedies, 39 Fla. St. U. L. Rev. 163, 181 (2011) [hereinafter Ripstein, Civil Recourse and Separation]. The principle of independence and accordingly, the requirement that the terms of people’s interactions manifest the formal equal independence of each interacting party 30 Kant, supra note 26, at 63. underlie modern Kantian accounts of the three building blocks of private law: property, contracts, and torts. 31 See infra notes 32–38 and accompanying text (summarizing the traditionalists’ conceptions of property, contracts, and torts).

Property. The Kantian principle of independence seeks to explain why property rights ought to be protected from external interference in the same manner as life and limb. The starting point of the explanation is the contention that there is no freedom-based justification for denying independent persons the possibility of exploiting external objects that are not already being effectively used (or controlled) by another. If peo­ple are to be allowed “to exercise their freedom by controlling external objects of choice,” these objects must be subject to the sole discretion of the choosing party, so that all others are bound by the proprietor’s unilateral will. 32 Weinrib, Corrective Justice, supra note 24, at 275; see also Ripstein, Force and Freedom, supra note 26, at 91 (“Your property constrains others because it comprises the external means that you use in setting and pursuing purposes; if someone interferes with your property, he thereby interferes with your purposiveness.”); Arthur Ripstein, Authority and Coercion, 32 Phil. & Pub. Aff. 2, 19 (2004) (noting that property requires the use and possession of an object).

Contract. Contractual consent also “gets its significance against the background of the basic [Kantian] right to independence that private persons have against each other.” 33 Ripstein, Force and Freedom, supra note 26, at 109. Contracts “enable free persons to exercise self-mastery together” 34 Id. at 108. and to “set and pursue their own pur­poses interdependently.” 35 Id. at 107. For A to gain access to B’s property, have an entitlement to B’s services, or enter into a joint venture with B, both A and B must use their “respective moral powers”; anything short of such a “united will” amounts to an attempt by A to convert B’s person or prop­erty into merely A’s own means. 36 See id. at 109, 114–15, 122–23, 127; Weinrib, Corrective Justice, supra note 24, at 153–54 (discussing “Kantian account of contractual entitlement”). Only by contract can B grant A “powers over [her] person and property in a way that is consistent with [her] exclusive power to determine how they will be used.” 37 Ripstein, Force and Freedom, supra note 26, at 127.

Tort. The same thin and formal conception of the person as a free and equal agent guides modern Kantians’ accounts of torts. Given the formally equal importance of each party’s independence, the terms of such interactions must be objectively set so as to preclude taking into ac­count the idiosyncrasies of the person whose conduct is being assessed. Incorporating such subjective considerations into the terms of an involuntary interaction would give one party to the interaction the stand­ing to determine these terms unilaterally, which would be in violation of formal equal freedom. 38 See id. at 171; Ernest J. Weinrib, The Idea of Private Law 147–52 (2012) [hereinafter Weinrib, Idea of Private Law] (highlighting the importance of objective tort law standards and, in particular, the standard of reasonable care); Ripstein, Civil Recourse and Separation, supra note 29, at 181 (arguing that private rights require objective standards).

For Kantians, the interpersonal respect we owe one another as free and equal persons means respecting each other’s abstract personalities. The “particular features—desires, endowments, circumstances, and so on—that might distinguish one agent from another” are “irrelevan[t].” 39 Weinrib, Idea of Private Law, supra note 38, at 82. The private individual is free by virtue of her capacity to set and pursue ends by deploying her person and property without being subordinated to others’ choices. Private individuals, moreover, are equal by virtue of having this capacity. They are thus “purposive beings who are not under duties to act for any purposes in particular, no matter how meritorious”; as such, they are subject to “a system of negative duties of non-interfer­ence with the rights of others”—namely, private law. 40 Weinrib, Corrective Justice, supra note 24, at 11.

Some modern Kantians argue that this understanding of private law is “juridical,” in that it “concerns itself only with values that reflect the distinctive nature of justification of private law.” 41 Id. at 28. But as Alan Brudner, a modern Hegelian, claims, the presentation of this view of private law as a logical necessity—a “mode of ordering ‘implicit’ in transactions”—fails because the law is not in fact “analytically determined” and the resort to the traditional understanding of private law is “morally contestable.” 42 Brudner, supra note 11, at 19, 21–22, 360. Furthermore, an argument from logical necessity sets an extremely high bar: There must be no possibility of any other coherent understanding of private law than as the law of interpersonal interactions among for­mally free and equal persons. Perhaps this requirement can be met in theory, but modern Kantians have yet to produce the required argu­ment. 43 See Hanoch Dagan & Avihay Dorfman, Against Private Law Escapism: Comment on Arthur Ripstein, Private Wrongs 10–15 (July 1, 2016) (unpublished manuscript) http://papers.ssrn.com/abstract=2769790 (on file with the Columbia Law Review) (arguing Ripstein’s theory fails to “withstand the test of moral intuitions”). Consequently, in order to justify such a libertarian private law, which presupposes “dissociated persons,” Brudner, like other modern liberals, returns to the traditional moral division of labor. Under this idea, the law governing our interpersonal relationships can, and thus should, uphold our independence by prescribing only “duties not to trans­gress personal boundaries” and relying on people’s public law rights—to which “the commonality” is accountable—to secure our “posi­tive right to the conditions of self-determination.” 44 Brudner, supra note 11, at 148, 352, 355.

B. The Critical Account: Private Law as the Continuation of Public Law by Other Means 45 To play on Carl von Clausewitz’s celebrated line. See Carl von Clausewitz, On War 119 (Anatol Rapoport ed., J.J. Graham trans., 1968).

Over the past century, legal realists, critical legal scholars, feminists, and lawyer-economists have attacked the traditional account of private law and the corresponding public–private distinction. 46 See, e.g., Jody Freeman, The Private Role in Public Governance, 75 N.Y.U. L. Rev. 543, 564–65 (2000) (summarizing critical legal studies scholars’ contention that the pub­lic–private divide is “incoheren[t]”). Many of these authors have been influenced by the earlier Marxist critique. See generally Gerald Turkel, The Public/Private Distinction: Approaches to the Critique of Legal Ideology, 22 Law & Soc’y Rev. 801, 805–09 (1988) (articulating Karl Marx’s position on the public–private divide). Read in their best light, these attacks correctly put forth two propositions: that this depic­tion of private law is neither inevitable nor apolitical and that at least some of its implications are normatively indefensible. 47 The first proposition was forcefully advanced in Hans Kelsen, Pure Theory of Law 280–83 (Max Knight trans., 2d rev. ed. 1967). We agree with these propositions. Some critics, however, go too far, adding to these claims more speculative contentions that amount to a dismissal of any possible distinction between private law and public law. 48 See infra notes 49–67 and accompanying text. This has the effect of abrogating any possible unique normative significance to private law as the law of interpersonal interactions. In so doing, these critics im­properly reduce private law to simply another form of allocation and regulation, indistinguishable from other regulatory regimes.

Legal realists and critical legal scholars direct much of their criticism at the traditionalists’ legal conceptions of property and contract. Because private law structures our daily interactions and thus tends to blend into our natural environment, the traditional discourse tends to “thingify” (or reify) its own contingent choices. This, in turn, causes people to perceive these choices as necessary (or at least, neutral and acceptable). 49 See, e.g., Felix S. Cohen, Transcendental Nonsense and the Functional Approach, 35 Colum. L. Rev. 809, 811–12 (1935) [hereinafter Cohen, Transcendental Nonsense] (criticizing the “vivid fictions and metaphors of traditional jurisprudence” for producing judicial decisions that “forget the social forces which mold the law”); Robert W. Gordon, New Developments in Legal Theory, in The Politics of Law: A Progressive Critique 413, 418–21 (David Kairys ed., 2d ed. 1990) (“[Legal] system building has the effect of making the social world as it is come to seem natural and inevitable.”); Robert W. Gordon, Unfreezing Legal Reality: Critical Approaches to Law, 15 Fla. St. U. L. Rev. 195, 212–14 (1987) [hereinafter Gordon, Unfreezing Legal Reality] (critiquing classical liberal attempts to find “innocuous or neutral rules” in private law). These critics show, further, that the traditional conception of private law as the realm of independence and formal equality is neither obvious nor incon­trovertible. 50 See, e.g., Libby Adler, The Gay Agenda, 16 Mich. J. Gender & L. 147, 192–93 (2009) (highlighting the shortfalls of formal equality in the context of same-sex marriage). Moreover, because the traditional understanding often turns out to serve “entrenched interests,” 51 Cf. John Dewey, Logical Method and Law, in American Legal Realism 185, 193 (William W. Fisher III et al. eds., 1993) (arguing that liberal legal systems, like all others, contain natural biases toward particular groups). these critics insist that, like public law, private law should be subject to a distributive analysis. 52 See, e.g., Aditi Bagchi, Distributive Injustice and Private Law, 60 Hastings L.J. 105, 106–07 (2008) (criticizing “affirmative arguments in favor of using [only] the state’s pow­ers of tax and transfer to effect redistribution”). Thus, since private property is not only “dominion over things” but “also impe­rium over our fellow human beings,” the law must address “the extent of the power over the life of others which the legal order confers on those called owners.” 53 See Morris R. Cohen, Property and Sovereignty, 13 Cornell L.Q. 8, 13 (1927). Traditional private law discourse impedes such an in­quiry by obscuring the distributive effects of law. 54 See generally Robert L. Hale, Coercion and Distribution in a Supposedly Non-Coercive State, 38 Pol. Sci. Q. 470 (1923) (arguing that law’s background rules are partly responsible for the inequalities in the distribution of income and power). It thereby also safe­guards the status quo from scrutiny and could even serve “to perpetuate class prejudices and uncritical assumptions which could not survive the sunlight of free ethical controversy.” 55 Cohen, Transcendental Nonsense, supra note 49, at 814–18, 840. For similar cri­tiques of other branches of private law, see Morris R. Cohen, The Basis of Contract, 46 Harv. L. Rev. 553 (1933); Louis L. Jaffe, Law Making by Private Groups, 51 Harv. L. Rev. 201 (1937); Karl E. Klare, The Public/Private Distinction in Labor Law, 130 U. Pa. L. Rev. 1358 (1982).

Feminist scholars similarly criticize the implications of the tradi­tional private law understanding of doctrines relating to the family. They underscore the contingency of the patriarchal family as well as the indis­pensable role of law in constructing this particular form of domestic rela­tions. Thus, they expose the flaws of traditional family law: In classifying the patriarchal family as “private” or “personal,” traditional family law adopts an extreme noninterference policy that obscures and perpetuates its injustice by shielding abuses—such as exploitation and battery of family members—from legal scrutiny. 56 See generally Ruth Gavison, Feminism and the Public/Private Distinction, 45 Stan. L. Rev. 1, 24 (1992) (noting that some “feminists challenge the [traditional family law] conclusion that the family should be free from interference by the state” because it “rests on false premises about the nature of family life” that have been used to “mask the exploitation and battering of family members”). Finally, feminists insist that domestic arrangements need to be publicly reviewed and that they, like any other part of private law, are appropriate subjects for theories of political and social justice. 57 See, e.g., id. at 16, 20, 23–25, 27–28.

Critics from each of these groups often take their arguments one step further with stronger—indeed, excessive—claims disputing the potential value of any possible alternative understanding of private law. They assert that “the division of law into public and private realms” is arbitrary and that all categories of private law are “delegation[s] of public power that [can] be justified only by public purposes.” 58 Morton J. Horwitz, The History of the Public/Private Distinction, 130 U. Pa. L. Rev. 1423, 1426 (1982). Private law, on this view, is “public law in dis­guise.” 59 Leon Green, Tort Law Public Law in Disguise, 38 Tex. L. Rev. 1, 1–2 (1959) (argu­ing that public-policy considerations should factor into the determination of all cases). Furthermore, “the theoretical distinction between public and private” is considered a legitimating device that “gives credence to the assumption that private activity is in fact purely private, so that the exer­cise of private power does not appear to be publicly sanctioned oppression.” 60 Alan Freeman & Elizabeth Mensch, The Public-Private Distinction in American Law and Life, 36 Buff. L. Rev. 237, 246–47 (1987); see also Catharine A. MacKinnon, Feminism Unmodified: Discourses on Life and Law 102 (1987) (arguing that the public–private distinction “keeps the private beyond public redress and depoliticizes women’s subjection within it”). This is why these critics celebrate the “decline of the pub­lic/private distinction” 61 Duncan Kennedy, The Stages of the Decline of the Public/Private Distinction, 130 U. Pa. L. Rev. 1349 (1982). See generally id. passim. and see the “explo[sion] [of] the private” 62 See Catharine A. MacKinnon, Toward a Feminist Theory of the State 191 (1989). as the prerequisite for “new possibilities for human contact.” 63 Freeman & Mensch, supra note 60, at 238.

The call to discard the public–private distinction implies that the division of labor between private law and public law is purely conven­tional and that at best it is a matter of institutional design based solely on the comparative advantages of the relevant regulatory devices. This posi­tion is currently most closely associated with the economic analysis of law, 64 See, e.g., Guido Calabresi, A Broader View of the Cathedral: The Significance of the Liability Rule, Correcting a Misapprehension, 77 Law & Contemp. Probs., no. 2, 2014, at 1, 1–4 [hereinafter Calabresi, A Broader View]. which in many other contexts is usually viewed as the nemesis of critical theory. 65 This is, to be sure, not the only point of convergence between the law and economics movement and critical legal theory: Both schools also tend to dismiss law’s normativity. As one scholar notes, “law and economics theorists deny the significance of a principled distinction between public law and pri­vate law” and tend to be indifferent toward—and at times even impatient with—theoretical efforts to establish such a differentiation. 66 See Harel, supra note 16, at 1040, 1050–51. Their basic view is that “[t]here is work to be done and it ought to be done in the best possible way,” with the choice between private or public agents (or private or public law) a “pragmatic” one that “depends on a comparison between the expected efficacy” of these possible agents “in performing the job.” 67 Id. at 1051.

To be sure, some of the economic analyses can be read as grounded in a commitment to autonomy (as self-determination) 68 See Hanoch Dagan & Michael Heller, The Choice Theory of Contracts (forthcoming 2017) [hereinafter Dagan & Heller, Choice Theory] (manuscript at 68–77) (on file with the Columbia Law Review). and thus may fit one of the two pillars of our own conception of private law—namely, sub­stantive freedom (the other being substantive equality). But autonomy is at best implicit in these analyses, while the explicit commitment of eco­nomic analysis is to maximizing aggregate welfare. This commitment necessarily generates an extreme instrumentalist approach to private law, and thus it should not be surprising that lawyer-economists tend to dis­miss the public–private distinction.

Whereas lawyer-economists seem content with such an undifferenti­ated legal domain, more critically oriented scholars tend to recognize both the troubling effects of the possible effacement of the public–private distinction and the resilient persistence of private law as a distinc­tive legal category. 69 Cf. Kit Barker, Private Law: Key Encounters with Public Law, in Private Law: Key Encounters with Public Law 3, 37–39 (Kit Barker & Darryn Jensen eds., 2013) (“[D]espite rumours to the contrary . . . , the private/public ‘distinction’ is far from dead. . . . [D]espite the pressures from the ‘public’ . . . there remain key aspects of private law’s ‘privateness’ . . . that are persistently important and that explain its resistance to being swallowed up in all that is ‘public.’”). As Ruth Gavison asserts, the feminist ideal is most certainly “not a state of affairs in which nothing is private.” 70 Gavison, supra note 56, at 42. Rather, feminists sometimes advocate changes “in the public/private mix,” given “the belief that women deserve more of . . . [the values of the private] than they presently receive.” 71 Id. at 29, 43. Similarly, in one of the canonical articula­tions of the critical legal studies critique of the public–private distinction, Alan Freeman and Elizabeth Mensch maintain that “one cannot dispute, and one should not demean, the liberating force” of these private law values. 72 Freeman & Mensch, supra note 60, at 256. But because they conceptualize “the basic model” of private law as one of “the exclusion of others” and the “affirmation of our alienated distance from one another,” they add that “[t]he dilemma is the extent to which what generates a moment of liberation soon serves to replicate, by use of the very same arguments, the world we are trying to change.” 73 Id.

II. The Justice of Private Law

These last observations help explain the contemporary theoretical deadlock as well as recent calls for a fresh new start. 74 John Goldberg, for example, recently announced that there is a “new private law.” John C.P. Goldberg, Introduction: Pragmatism and Private Law, 125 Harv. L. Rev. 1640, 1651 (2012). Goldberg notes four tenets of the new private law’s methodological commitments: recognition that “law is distinct from politics and morality” but that it is “not disconnected from them”; commitment to “conceptual legal analysis”; commitment to take law “seriously,” including through interdisciplinary study; and recognition that legal concepts are often influenced by the contexts in which they operate and the persons in charge of their administration. Id. at 1663. We fail to see, however, what pre­cisely renders these methodological commitments novel or even different enough from conventional private law theory to warrant the caption “the new private law.” Both the traditional and critical approaches understand the value of private law in similar terms: namely, as the practical expression of formal freedom and equality. But as established in Part I, they have opposite responses to this value, with the traditionalists endorsing and the critics denouncing or ignoring.

This understanding of private law is neither self-evident nor inevita­ble. Rather than idling in debate over the virtues and vices of the traditionalist conception of private law, we reconceive private law in a way that both is truer to democratic society’s liberal normative commitments and as it turns out, better accounts for much of our existing law. Our ac­count illuminates the irreducible value of private law in both form and substance, elaborating on the powerful intuition that private law ad­dresses our interpersonal relationships as private individuals rather than as citizens of a democracy or patients of the welfare state’s regulatory scheme. 75 We do not claim that it is entirely possible to disentangle our identities as individuals from our identities as democratic citizens or welfare-state patients. Rather, our point is that we should be careful not to conflate the social with the statist. Our relation­ships as individuals need not depend on the state, although it may in many contexts. For further discussion, see infra section II.C.3. It also puts to rest the misguided identification of private law with adjudication. In this Part, we elaborate the justice of private law—its underlying ideal of just relationships—and explore some of the implications of the reconstructed public–private distinctions it implies.

A. The Relational Form of Private Law

Private law addresses our interpersonal interactions by marshaling rights and obligations that take a relational form. This is, of course, mysterious from the standpoint of lawyer-economists and other critics, who see private law as just another means to serve our public goals. Yet private law does not deal with the parties to an interaction, taken sever­ally, but rather with the terms of their engagement with each other. It addresses, in other words, the rights and duties they bear in relation to one another as well as the frameworks of interpersonal interaction they sustain. A right to property, for example, corresponds with a duty against committing trespass. 76 See Avihay Dorfman & Assaf Jacob, The Fault of Trespass, 65 U. Toronto L.J. 48, 61–64 (2015); Avihay Dorfman, The Normativity of the Private Ownership Form, 75 Mod. L. Rev. 981, 993–95 (2012). It is a duty owed to the right holder in particular rather than to the entire universe of property-right holders. 77 Cf. John G. Fleming, The Law of Torts 79 (8th ed. 1992) (noting that lack of con­sent is fundamental to an action in trespass); Walter Wheeler Cook, The Power of Courts of Equity, 15 Colum. L. Rev. 37, 53 (1915) (emphasizing that trespass is an action in personam). This duty is owned by the right holder in the sense that it is an upshot of her basic Hohfeldian power to decide, within the limits set by the law of property, whether or not to seek its realization against those who are deemed liable to such power. 78 See Wesley Newcomb Hohfeld, Some Fundamental Legal Conceptions as Applied in Judicial Reasoning, 23 Yale L.J. 16, 44–54 (1913) (arguing that legal power is the correlative of legal liability). Similarly, the contractual obligation to keep one’s prom­ise is owed directly to the promisee, who, in turn, exercises an important measure of control over its fulfillment. 79 See Restatement (Second) of Contracts ch. 12, topic 4, intro. note (Am. Law Inst. 1981) (discussing a promisee’s authority to discharge a promisor’s duty to meet the requirements of a contract). Tort law, too, applies a relational form of rights and duties. A duty of care, for example, is not owed to the world at large 80 See, e.g., Palsgraf v. Long Island R.R., 162 N.E. 99, 99 (N.Y. 1928) (limiting one’s duty of care to a particular zone of danger). but rather carves out a class of potential victims whose relationship with, and proximity to, an injurer justifies the imposition of a relational duty. 81 See, e.g., Marshall v. Burger King Corp., 856 N.E.2d 1048, 1057 (Ill. 2006) (“The touchstone of this court’s duty analysis is to ask whether a plaintiff and a defendant stood in such a relationship to one another that the law imposed upon the defendant an obliga­tion of reasonable conduct for the benefit of the plaintiff.”); Coates v. S. Md. Elec. Coop., 731 A.2d 931, 936 (Md. 1999) (“Inherent . . . in the concept of duty is the relation­ship of the parties . . . .”).

(Private law’s relational form raises the question of how to distin­guish it from criminal law. This concern would be most relevant to those who conceive it as purely a publicization of the private power to vindicate interpersonal rights. 82 Many scholars defend the opposite view, conceiving crimes as wrongdoings against the public as a whole. See, e.g., Emile Durkheim, The Division of Labor in Society 62–63 (W.D. Halls trans., 1984); Herbert Morris, Persons and Punishment, in On Guilt and Innocence 31, 33–34 (1976); Michael S. Moore, The Moral Worth of Retribution, in Responsibility, Character and the Emotions 179, 179–80 (Ferdinand Schoeman ed., 1987). This view, however, detaches criminal law’s vertical enforcement structure from an underlying horizontal substantive right and thereby implies that criminal law should indeed be understood to extend, and even bolster, the force of private law. 83 See, e.g., Randy E. Barnett, Restitution: A New Paradigm of Criminal Justice, 87 Ethics 279, 287–94 (1977). Under this concep­tion, criminal law justifiably bears a significant resemblance to certain modern regulatory schemes that have partially replaced traditional pri­vate law institutions of adjudication and enforcement. 84 See infra notes 172–177 and accompanying text (explaining how some modern regulatory schemes are, at least partially, functionally equivalent to their traditional com­mon law counterparts). )

Private law’s relational form of legal ordering can be used for any number of external good causes, such as increasing overall social welfare. Tort law’s duty of care, for instance, makes society safer, and contract law allows people to further their own personal ends efficiently. But it is our contention that the relational form that characterizes private law also has value in and of itself, quite apart from its contribution to the realization of external goals. This is because private law is premised on people’s en­gagement with one another to achieve the ends they each pursue. To this extent, private law’s rights, obligations, and frameworks structure the pursuit of ends in a relational way. To illustrate, while a contractual prom­ise may enable both promisee and promisor to realize their respective desirable goals, the very manner in which the contractual transaction achieves this is of value, too, for it requires those who utilize it to recog­nize each other as parties to a joint endeavor. 85 See Daniel Markovits, Contract and Collaboration, 113 Yale L.J. 1417, 1448–64 (2004) [hereinafter Markovits, Contract and Collaboration] (arguing that contractual promises, like personal promises, promote the morals of respect and collaborative commu­nity); Joseph Raz, Promises and Obligations, in Law, Morality, and Society: Essays in Honour of H.L.A. Hart 210, 227–28 (P.M.S. Hacker & Joseph Raz eds., 1977).

Indeed, private law’s relational form of rights and obligations gener­ally facilitates the realization of certain projects through interpersonal interactions. At times, this interpersonal dimension is precisely the goal of the interaction—joining forces is the crux of projects such as marriage, whose essence is an underlying social interaction. In other con­texts, when enlisting others makes projects more feasible or practical (such as a supply contract for goods), parties engage in transactions for more in­strumental reasons. Different theories offer divergent explanations for the value that (arguably) inheres in private law’s relational form. Some articulate a thin and rather generic account of respectful recognition and more generally, liberal solidarity in various areas of pri­vate law; 86 See, e.g., Brudner, supra note 11, at 132, 155–59 (articulating that view in the context of property law); Avihay Dorfman, The Society of Property, 62 U. Toronto L.J. 563, 590–96 (2012) (same); Markovits, Contract and Collaboration, supra note 85, at 1448–64 (articulating that view in the context of contract law). others emphasize thicker types of private law engagements in particular social contexts. 87 See, e.g., Carolyn J. Frantz & Hanoch Dagan, Properties of Marriage, 104 Colum. L. Rev. 75, 81–94 (2004); Ian R. Macneil, The Many Futures of Contracts, 47 S. Cal. L. Rev. 691, 731, 805–06 (1974). We need not delve into these accounts be­cause our primary concern here is the contents of private law’s relational frameworks. But it is worth mentioning that all these different accounts imply that the distinctive feature of private law is the ideal of interper­sonal relationships it sets up rather than the specific legal mechanisms for addressing deviations from this ideal, be they the familiar one-to-one litigation or otherwise. 88 We thus reject the claim that the distinctive feature of tort law (or even private law writ large) is a “core idea of redress,” John C.P. Goldberg, The Constitutional Status of Tort Law: Due Process and the Right to a Law for the Redress of Wrongs, 115 Yale L.J. 524, 601 (2005); see also Benjamin C. Zipursky, Civil Recourse, Not Corrective Justice, 91 Geo. L.J. 695, 733–40 (2003), under which “a person who is wronged, but deprived by law of the ability to respond directly, is entitled to an avenue of civil recourse against the wrongdoer.” John C.P. Goldberg & Benjamin C. Zipursky, Civil Recourse Defended: A Reply to Posner, Calabresi, Rustad, Chamallas, and Robinette, 88 Ind. L.J. 569, 573 (2013); see also Benjamin C. Zipursky, The Philosophy of Private Law, in The Oxford Handbook of Jurisprudence and Philosophy of Law 623 passim (Jules Coleman & Scott Shapiro eds., 2002).
For other critiques of civil recourse theory, see, e.g., Avihay Dorfman, Private Law Exceptionalism? Part I: A Basic Difficulty with the Arguments from Bipolarity and Civil Recourse, 35 Law & Phil. 165, 177–85 (2016) (arguing that civil recourse fails due to its overinclusiveness, since the entitlement to substantive standing, as well as the entitlement to seek civil recourse against the right violator, cannot distinguish tort law from many other areas of public law, notably constitutional rights law).
As usual, law’s core significance lies in its every­day success rather than in its pathological failures. 89 Cf. H.L.A. Hart, The Concept of Law 79–88 (1961) [hereinafter Hart, Concept of Law] (explaining that legal norms are taken not only as predictions of judicial action but also as standards and guides for conduct and judgment and as bases for claims, demands, admissions, criticism, and punishment).

B. The Normative Contents of Private Law

The traditional conception of private law (and of the public–private distinction) also emphasizes its relational form. But its content differs dramatically from what we propose. Traditionalists construct an ideal of just terms of interaction around a formal conception of the free and equal person. 90 This conception of the person as free and equal and the one we shall defend in its stead, see infra notes 95–103 and accompanying text, are normative, rather than ontological, constructs that help flesh out what features of the human condition should be relevant to the analysis of rights and duties grounded in freedom and equality. Under this conception, people are equal in their interpersonal relationships if none is superior or subordinate to another and every person is free as against all others—entitled to set and pursue their own conceptions of the good. 91 Ripstein, Private Wrongs, supra note 24, at 288–95 (describing formal equality as central to horizontal relationships). Die-hard libertarians subscribe to this position because, for them, independence and formal equality are the only legitimate commitments of law, tout court. Liberal egalitarians, by contrast, take individual self-determination and substantive equality more seriously. They too, however, nonetheless exclude these values (at least in principle) from private law, insisting that the polity’s responsibility to these particular values is purely vertical in direction. 92 See supra text accompanying notes 17–22. They argue that the commitment to individual self-determination and substantive equality does not, and should not, govern people’s horizontal relationships and that so long as people respect one another’s independence and formal equality, they bear no responsibility for one another’s autonomy and need not be concerned with claims to substantive equality. 93 Taken to its logical extreme, this division-of-labor argument suggests that as soon as the state complies with its vertical obligations, the conventional conceptions of property as an absolute right to a thing valid against the world and contract as a means for delineat­ing boundaries of protected domains are all that free individuals need in order to form, pursue, and realize their good lives—including their preferred interpersonal arrange­ments. Cf. Merrill, Property as Modularity, supra note 25, at 157–58 (discussing the modularity model of property and how the standardization of exclusion rules varies with the identity of the party interacting with the property).

The critical account of the public–private distinction discussed above 94 See supra section I.B. is driven by a profound dissatisfaction with the traditionalists’ ideal of just terms of interaction among private individuals. This dissatisfaction is justified: The traditionalist view of private law disturb­ingly takes off the table the liberal commitment to individual self-determination (and not merely independence) and to substantive equal­ity (and not merely formal equality). 95 See supra notes 46–57 and accompanying text (summarizing the critique of the traditional view). Setting aside these canonical liberal values insofar as they concern private law is troubling in light of two aspects of the human condition: our interdependence and our per­sonal differences. If we take the facts of interdependence and personal difference seriously—if we appreciate both the vulnerability and the valuable options to which these social conditions give rise—we must acknowledge that the liberal commitment to individual self-determina­tion and substantive equality cannot be excluded from the law governing horizontal relationships. Our approach, then, adopts liberalism’s most fundamental commitments and should be read as a friendly attempt at amending a contingent, albeit significant, feature of its dominant articulations. 96 Some liberals may be resistant to this, invoking liberalism’s commitment to the legitimating features of public lawmaking. But unless one espouses a robust libertarian position, there is no reason to suspect that a legal regime that upholds independence and formal equality is at all more legitimate (or less coercive) than one that vindicates self-determination and substantive equality. See generally Hanoch Dagan, Liberalism and the Private Law of Property, 1 Critical Analysis L. 268 (2014) (criticizing the Hegelian effort to establish the legitimacy of private law libertarianism on liberal grounds).

To be sure, the implications of the commitment to the core liberal values of self-determination and substantive equality differ between the private and the public sphere. The source of this difference lies in the varying capacities in which people operate in private law and public law; simply put, our interactions as private individuals are of a different na­ture than our interactions as citizens. 97 For more on the distinction between people acting in their capacity of private individuals and of citizens, see 1 Bruce Ackerman, We the People: Foundations 230–31, 250–51 (1991). Our private obligations are shaped by reference to the particular interpersonal practices involved; 98 See infra section II.C.2 (discussing the impact of the pertinent practice on the specifics of interpersonal obligations). they are unencumbered, at least in principle, by the (potentially more demanding) public obligations of cocitizenship. 99 To this extent, our approach is remarkably different from those who call for constitutionalizing some aspects of private law. See, e.g., Tarunabh Khaitan, A Theory of Discrimination Law 201 (2015) (“[T]he antidiscrimination duty . . . [should] only [be] im­posed on those persons who have a sufficiently public character.”); Johan van der Walt, The Horizontal Effect Revolution and the Question of Sovereignty 22 (2014) (arguing that constitutional norms should be applied only insofar as these interactions are “situated in the context of majority-minority relations” and when disputes assume a “broader political dimension”). This qualitative differ­ence is well reflected in the parochial scope of our lawmaking practices among members of a political community, on the one hand, and the po­tentially universal scope of our interpersonal practices among persons, on the other. 100 This is why our conception of private law can, and probably should, inform the substantive law governing interpersonal interactions across national borders. For a prelimi­nary exploration, see Hanoch Dagan & Avihay Dorfman, Interpersonal Human Rights and Transnational Private Law 4 (May 19, 2016) (unpublished manuscript) (on file with the Columbia Law Review) (“[A] core set of interpersonal duties . . . [is] not limited to interac­tions among compatriots, but rather pertain[s] to relationships between individual per­sons as such . . . . [T]hese obligations . . . embody our interpersonal human rights.”).

This section refines and defends the ideal of just relationships prem­ised on reciprocal respect to self-determination and substantive equality. It also explains both the indispensable role of law in instantiating this ideal and the limits thereof.

1. On Interdependence and Personal Difference. — Our practical affairs are deeply interdependent, replete with interactions with others that range from the trivial, such as purchasing a coffee at a café, to the most valuable and intimate, such as those connected to family, friends, and work. These interactions can take either voluntary or involuntary forms: We invite, or are invited by, others to engage in joint projects. Those same projects often render vulnerable, or otherwise interfere with, the legitimate interests of other people, especially those who are outside the privity of the joint enterprise. The ability to successfully lead one’s life—and to relate to others as equals—is influenced at almost every turn by both of these types of interaction.

This fact of interdependence does not and need not affect the way libertarians understand private law. If independence (negative liberty) exhausts the requirements of freedom, the fact of interdependence only makes more imperative the requirement that private law vindicate per­sonal independence. But liberal egalitarians contest this decidedly thin understanding of freedom. They insist that an individual person is free not merely in the formal sense of not being subordinated to the choices of another but also in the more robust sense of being able to make mean­ingful choices about the direction of her life. 101 See John Rawls, Justice as Fairness: A Restatement 19–21 (Erin Kelly ed., 2001) (arguing that individuals are free when they act on their capacity “to have, to revise, and rationally to pursue a conception of the good”). In the formal sense, a person can be “free” simply because no one else is in a position of domi­nance over her. But this conception neglects concerns for the effective realization of that person’s ability to form and pursue her own concep­tion of the good. Rather, self-determination is necessary for people to lead the fully human life to which they are entitled. While this requires a measure of independence, it “is not something automatically guaranteed by a structure of negative rights.” 102 H.L.A. Hart, Between Utility and Rights, 79 Colum. L. Rev. 828, 836 (1979) [hereinafter Hart, Between Utility and Rights]. Therefore, if a just relationship requires reciprocal respect of each party’s claim to self-determination, re­lational justice cannot be exhausted by the negative duty of non­interference; it may, at times, require some affirmative interpersonal accommodation that takes account of certain personal circumstances or choices.

The traditional public–private distinction fails to sufficiently account for the role private law plays in constituting, facilitating, and authorizing such interdependent interactions. (These roles are vividly demonstrated, for example, in the common law rules that help solve collective action problems or oblige recipients of mistaken payments to reverse mistakes for which they have no responsibility. 103 See infra Part III (discussing these and other examples demonstrating this point). ) This failure reflects—and in­deed perpetuates—an undervaluation of the significance of our interper­sonal relationships to our conceptions of the good life. These errors are troublesome even in societies with just public law arrangements—in which all citizens have adequate opportunities to realize their full free­dom in their private lives. Our interdependence implies that our horizontal interactions are too significant to our autonomy and social equality to be so easily supplanted by vertical arrangements, however just they may be.

Further, the significance of interpersonal and interdependent en­gagements also implies that the substantive terms of the interactions themselves should be evaluated as just or unjust. Here too the traditional conception disappoints. Notwithstanding the fact of personal differ­ence—we all constitute our own distinctive personhoods on the back­ground of our unique circumstances—the traditional conception replaces a concern for people as real individuals relating to one another as free and equal agents with a concern for people as abstract beings. By assigning sole responsibility to address our personal differences to public law, traditionalists implicitly dismiss any demands private individuals may make on one another as a matter of relational justice.

Such an underestimation of this horizontal dimension of justice is deeply problematic. For persons to relate to one another as equals, the terms of their interaction must not reflect significantly unequal power or advantage of one party over the other. To meet this demand, the terms of private law interactions must reference the participants’ relevant per­sonal qualities, including their distinctive characteristics and circum­stances. Consider, for example, the tort law requirement that potential injurers accommodate the relevant constitutive features of their victims 104 See infra section III.A (highlighting relational justice in the context of tort law). or the property and contract law rules regarding residential and work­place accommodation. 105 See infra section III.B (highlighting relational justice in the context of residential and workplace accommodation). Only if the structure of the parties’ terms of interaction is predicated on the conception of the person as a substan­tively, not merely formally, free and equal agent can it guarantee a more-or-less fair relational starting point from which both parties can realize their respective freedoms. Therefore, to count as just relationships, the terms of the interaction must be determined with regard to the parties’ choices and circumstances to the extent that those choices and circum­stances are crucial to the ability of the parties to relate as equal and self-determining individuals given the persons they actually are. This prescription implies precisely the kind of accommodative structure that the traditionalists’ commit­ment to formal freedom and equality precludes.

Indeed, the respect that interacting parties are required to accord to one another should relate to more than merely their generic human capacity for choice. To be sure, a world that renders any specific accommodation redundant—say, with a certain technological improve­ment or through a reshuffling of a given social practice without undermining the good underlying that practice—would be an improve­ment over a world that requires such an accommodation. Yet while inter­personal respect does not require overemphasizing personal differences, there is a qualitative distinction between entirely eliminating the impact of a certain personal characteristic on a person’s life and merely ameliorating its practical effects through government action.

2. The Conception of the Person of Private Law. — Our emerging no­tion of accommodative terms of interaction raises the question of what features of the human condition are included in a thicker conception of the person. So far we have described what this conception is not: It does not reduce the person to an abstract bearer of generic personality. But what does it consist of affirmatively? To answer this question, it is im­portant to distinguish between two facets in the explication of our account of substantive equality and freedom.

The first facet consists of the notion that equality in this context is a relational ideal: It focuses on the parties’ equal standing with respect to determining the terms of their interaction. Equality, here, is relational not merely in the form that it takes; it is not merely about duties that per­sons formally owe to certain others. Rather, the relational ideal at the core of private law stands for the normative commitment underlying these duties. In particular, it picks out a commitment to the value of be­ing with others in relations of respectful recognition. 106 Cf. Scheffler, Equality and Tradition, supra note 19, at 176 (emphasizing the rela­tional roots of the liberal commitment to egalitarianism); Elizabeth S. Anderson, What Is the Point of Equality?, 109 Ethics 287, 313 (1999) (“[E]galitarians seek a social order in which persons stand in relations of equality.”). This commitment places constraints on the class of conceptions of the good and other personal choices that may require interpersonal accommodation. Not all choices can be the object of interpersonal respect among free and equal persons: Some personal choices, policies, and conceptions of the good deny certain others the very standing to relate to the deniers as equals—the animating ambition of both the murderer and the racist, for example, is the repudiation of their victims’ equal standing. In short, choices that are inimical to the ideal of relational equality cannot lay a compelling claim as plausible candidates for interpersonal accommodation.

The second facet—which renders more determinate our conception of the person as substantively free and equal—takes up the question of what features of the person’s situation should count for the purpose of being respected by others as the person she really is. For the sake of exposition, consider the distinction between choice and circumstance. 107 This distinction is used for exposition purposes only. In our analysis, both immutable features and deeply constitutive, chosen features can receive similar normative and legal treatment. Circumstances are strictly construed as encompassing only the immutable features of a person’s situation, such as race, sex, and disabil­ity. Thus, to facilitate respect for a person as substantively free and equal, just terms of interaction cannot allow the full costs of possessing such a feature to be borne by its possessor. 108 The term “costs” (as in bearing some of the costs of a person’s choice) is used broadly to include monetary and nonmonetary burdens that arise from possessing (or accommodating) the relevant features. Typically, circumstances that gener­ate disrespect for a person’s equal standing are related to traits that have been publicly branded as inferior, something reflected by the suspect classes enumerated in antidiscrimination laws. In principle, however, the demands of relational justice do not depend on such a public perception of inferiority. 109 The qualified language of the text is due to our second-order considerations that could justify a sort of numerus-clausus limitation on the list of suspect classes. See infra note 125 and accompanying text.

Choices, by contrast, consist in a more complex category of personal features. Relevant choices—those that are not inimical to the ideal of relational equality—can be placed on a spectrum based on their relative contribution to self-determination. At one extreme are “ground pro­jects,” or the choices that reflect the commitments that make us who we are. 110 The concept of “ground projects” derives from Bernard Williams, who character­izes them as those “projects which are closely related to [an individual’s] existence and which to a significant degree give a meaning to his life.” Bernard Williams, Persons, Character and Morality, in Moral Luck 1, 12 (1981). To be sure, nothing in our argument turns on Williams’s development of the concept of a ground project, including his psychological argument that the demands of impartial morality exert unreasonable pres­sure on the personal integrity of those who pursue such projects. Ground projects—especially religious, ethical, professional, and familial commitments—are fundamental to the meaning of a life. 111 Id.; cf. James D. Nelson, The Freedom of Business Associations, 115 Colum. L. Rev. 461, 463, 493–96 (2015) (illustrating that one’s identity develops from both individ­ual personhood and interaction with various groups). In emphasizing the importance of ground projects we are not proposing that the individual person is literally fully deter­mined by them (or by the culture or community to which they belong). After all, the idea of self-determination—the essence of leading an autonomous life—implies that it should always be up to the individual to decide what ground project to pursue. Cf. Rawls, Political Liberalism, supra note 20, at 30–32 (insisting that identities, commitments, and attach­ments that “give shape to a person’s way of life” are at bottom revisable). At the other extreme are choices that reflect preferences as to the realiza­tion of superficial ends, whose frustration bears very little, if at all, on one’s conception of the self. In between these poles, there are choices that involve commitments that, although valuable, do not shape the over­arching meaning of one’s life and therefore might not contribute toward defining one’s identity.

Sketching the contours of each of these categories of choice requires an elaborate theory of autonomy and an account as to what choices make a person’s life go well. For our purposes, the conception of “person” must encompass the first-category choices—ground projects—that consti­tute an individual’s self, the person one actually is. Conversely, third-category choices—mere preferences—do not have a strong claim to accommodation, and thus their costs should be fully internalized by the person who has made the choice. One familiar example is that of a plain­tiff whose choice of activity exhibits risk-preferring attitudes. The tort doctrine of assumption of risk properly absolves defendants from the duty to accommodate risky choices made by risk-preferring plaintiffs. 112 See Avihay Dorfman, Assumption of Risk, After All, 15 Theoretical Inquiries L. 293, 318 (2014) [hereinafter Dorfman, Assumption of Risk] (distinguishing between put­ting oneself in danger for the sake of, rather than in spite of, a known risk).

It is less clear whether accommodation duties must also apply to choices of second-category (intermediate) choices. The requirement to respect others on their own terms justifies integrating such choices in the thicker conception of the person. Nonetheless, because these choices have a less profound impact on the chooser’s self-determination, private law can (and probably should) insist on the chooser’s responsibility to moderate her demand to have her second-category choice accommo­dated by those with whom she interacts. To illustrate, the just terms of interaction between an employer and employee could entail that the for­mer should reasonably accommodate the latter’s absence from work due to important familial or religious commitments. 113 See infra note 203 and accompanying text (justifying the strong sentiment against excluding potential employees from the labor market based on certain defining personal characteristics like disability or religious affiliation). Yet the same rationale does not apply for an employee who is, for example, unavailable for work on certain days because of her interest in watching migratory birds pass­ing through; these kinds of leisure activities are clearly not first-category, ground-projects choices, and as such the employer bears no responsibil­ity to accommodate.

3. Toward a Novel Approach to the Problem of Poverty in Private Law Theory. — By reclaiming a thicker conception of the person for private law, our account provides a novel approach to the problem of economic inequality in private law theory. Typically, private law theorists defend the legitimacy of private law in two contrasting ways. Some argue that private law can and should be arranged to promote distributive justice, in which case private law joins tax law’s effort to bring about justice in holdings. 114 See, e.g., Aditi Bagchi, Distributive Justice and Contract, in Philosophical Foundations of Contract Law 193, 193–95 (Gregory Klass et al. eds., 2014) (arguing that principles of distributive justice can and should impose numerous constraints on contract law); Anthony T. Kronman, Contract Law and Distributive Justice, 89 Yale L.J. 472, 499 (1980) (discussing taxation and contractual regulation as methods of redistribution); see also infra note 181 and accompanying text (discussing distributive justice accounts of tort law). Others—private law libertarians—argue that private law must express principled indifference for considerations of economic inequalities, in which case public law alone is expected to do all the heavy lifting. 115 See, e.g., Dworkin, Law’s Empire, supra note 21, at 296; Weinrib, Corrective Justice, supra note 24, at 308.

On our account, by contrast, the problem of poverty in and around private law is not merely one of distributive justice. It is, instead, a prob­lem of relational justice. The question is not, then, whether to enlist the machinery of private law to promote justice in holdings across society but whether some instantiations of economic disparities ought to be taken into account when fixing just terms of interactions between individual persons even when thus fixing cannot come close to a scheme of system­atic redistribution of resources.

So the question is whether one’s low economic status can count as one of the personal traits whose existence calls for some measure of ac­commodation by others. While we recognize the possible existence of countervailing considerations—ascertaining one’s economic status can often be either overly intrusive or prohibitively costly, or both—we think that in principle the answer is in the affirmative. Indeed, at times, poverty or an immensely inferior economic starting point may come close to an immutable feature of a person’s situation, as when it is a surface manifestation of some disability or the upshot of an unusually poor back­ground (with very little opportunity to escape such a predicament). At other times, economic hardship that could warrant accommodation may arise from a person’s choice to follow a Franciscan-like conception of the good. In these and similar cases, accommodating one’s poor economic situation is an expression of respect for a person on her own terms. 116 The private law doctrine of assumption of risk sheds light on the implications of relational justice to the problem of economic inequalities and in particular, on private law’s infamous historical indifference to economic disparities. See, e.g., Lamson v. Am. Ax & Tool Co., 58 N.E. 585, 585–86 (Mass. 1900) (declining to consider plaintiff’s poor eco­nomic circumstances while faulting him for continuing to work in spite of increased risk of injury); see also Dorfman, Assumption of Risk, supra note 112, at 308–13 (discussing the doctrine’s traditional indifference to economic disparities and its later repudiation).

4. The Role of Law and the Limits of Interpersonal Accommodation. — Even friendly readers who find our account of relational justice among private individuals attractive may raise significant concerns: Are legal du­ties properly suited for expressing the ideal of just relationships? And are not such duties too intrusive on the autonomy of other persons? We ad­dress these questions to clarify the scope and limits of the requirements of interpersonal accommodation.

Consider first how legal norms in a liberal society interact with hu­man agency. Legal duties of accommodation purport to provide duty holders with mandatory reasons for action—in other words, a justifica­tion for why it is necessary to act in a certain way. 117 See generally Hart, Concept of Law, supra note 89. This implies that a critical distance exists between the normative grounds of a given reason and the motivation for conforming to its demands. The former cannot actually produce the latter; it can only influence persons to acquire it. This gap between reason and motivation is particularly important within the domain of legality. Typically, the law only compels persons to act in conformity with a demand rather than because of a particular reason. 118 We say typically because the gap between giving reasons and motivating is not a conceptual truth about law. Rather, it is a normative requirement to which the law govern­ing liberal societies must adhere. This limitation on the enforcement of motives re­flects both substantive considerations—such as those pertaining to the distinction between political and personal morality of right and virtue—and instrumental ones—notably the unverifiability of persons’ internal mental states. Cf. Hales v. Petit (1562) 75 Eng. Rep. 387, 397; 1 Plow. 253, 253 (asserting that the mere “imagination of the mind” is not punishable). Yet the morality of laws adhering to the ideal of just relationships turns not on the actual motivations of duty holders but on their having reason to act in a way that is respectful of others.

These observations help explain why the responsibility for uphold­ing just horizontal relationships requires a legal apparatus and cannot be fully delegated to social norms. To be sure, social norms may suffice inso­far as they respond to the dictates of just relationships and are taken to have an obligatory nature so that they in fact govern people’s interper­sonal relationships. But this is only because they would then be law-like. If, however, this is not the case—which is likely given our contemporary social environment—relying on social norms amounts, at best, to an indi­rect and opaque endorsement of private law libertarianism. This is deeply problematic because it would threaten the liberal state’s commit­ment to individual self-determination and substantive equality. There is, therefore, a clear role for law in upholding and promoting just relationships.

We recognize that law’s prescriptive effects are not limitless. In cer­tain cases, for example, legal intervention might backfire by crowding out internal motivations. 119 See Yuval Feldman & Tom R. Tyler, Mandated Justice: The Potential Promise and Possible Pitfalls of Mandating Procedural Justice in the Workplace, 6 Reg. & Governance 46, 48 (2012) (discussing the potential negative impact of formal legal regulation on individuals’ motivations for following such regulations). But notwithstanding these kinds of excep­tions, 120 See infra note 257 and accompanying text (discussing specific concerns about crowding out in the context of affirmative duties). incorporating interpersonal obligations into the law does not necessarily undermine their moral value. By the same token, although interpersonal practices diverge—some arise independently of political authority, others are the unique creations of such authority, and still oth­ers occupy an intermediate category involving some degree of legal facilitation—private law can be deeply involved in setting out the terms of interaction among those engaging in the vast domain of interpersonal practice. 121 To be sure, some practices will be rightfully exempt from any legal treatment. This may be due either to the crowding out concerns just discussed or to the limited scope of the legal application of relational justice. See infra notes 124–125 and accompanying text. Therefore, the responsibility for upholding just horizontal relationships cannot be fully delegated to social norms. 122 Further, in some cases involving the Hohfeldian power to impose duties on nonconsenting individuals—as with owner’s power to change nonowner’s normative situa­tion with respect to a resource—such a delegation may even be impossible.

Given that law must play this irreducible role, we now turn to the limited scope of the legal application of relational justice. We identify three important limits here. First, some limits emerge from the nature of legal prescriptions. For example, activities that turn on authenticity and sincerity, such as romantic love or friendship, should lie beyond the reach of the law; treating them as mandatory reasons for action neces­sarily destroys their inherent value. 123 The tension between legal duties and reasons for action that turn on authenticity is nicely captured by the Kantian distinction between officia iuris and officia virtutis. See Kant, supra note 26, at 31. Second, other limits of the legal application of relational justice come from the rule-of-law maxim of providing effective guidance to law’s addressees and thus also constrain­ing officials’ ability to exercise power. 124 See generally Hanoch Dagan, Reconstructing American Legal Realism & Rethinking Private Law Theory 202–03, 212 (2013) (discussing the rule-of-law require­ments of guidance and constraint); Hanoch Dagan, Doctrinal Categories, Legal Realism, and the Rule of Law, 163 U. Pa. L. Rev. 1889, 1902–03 (2015) (arguing that “case-by-case adjudication inhibits law’s ability to provide effective guidance” and to properly constrain adjudicators’ power). This maxim helps defuse the po­tentially intrusive and demanding aspects of accommodation by setting out clear categories and doctrines with which individuals can adequately discharge their duties, on the one hand, while allowing them to exercise their rights of accommodation, on the other. 125 Thus, one reason antidiscrimination laws invoke the numerus clausus principle is that this method allows employers (and others) to ascertain and assess precisely what ac­commodation requires. Historically, these antidiscrimination laws may be described as the state commandeering private individuals into the service of correcting past societal fail­ures. However, the enumeration technique can also be understood as the legal order’s means to a more inclusive commitment to the demands of relational justice among individual persons. This latter interpretation suggests that our account is consistent with the spirit of contemporary antidiscrimination law. These rule-of-law tech­niques create an intersubjective frame of reasoning that is capable of guiding participants’ deliberation and behavior by minimizing resort to individualized knowledge and radically ad hoc judgments.

Third and finally, there are limitations on the scope of relational jus­tice that derive from within this ideal itself. The duty of accommodation is not an all-encompassing requirement to accommodate each and every person in each and every area of their practical affairs. Rather, the duty typically establishes fair terms of interaction in and around one sphere of action; it applies to a particular context or event and with respect to one person (or class of persons) at a time. In the contexts of negligence law and workplace accommodation law, for example, limits to accommoda­tion derive directly from the just-relationships ideal. 126 See infra sections III.A–.B (demonstrating the limits of accommodation). This is because a duty of accommodation grounded in relational justice is a range prop­erty 127 For an explanation of the notion of the range property, see generally John Rawls, A Theory of Justice 508 (1971) [hereinafter Rawls, Theory of Justice]. : People, as noted, cannot be legitimately required to accommo­date choices that repudiate the status of others as free and equal per­sons. 128 See supra section II.B.2 (discussing this exception). This seemingly minimal constraint also implies that the burden to perform an interpersonal duty cannot be excessive because it must neither undermine the autonomy of either party involved nor create interpersonal subordination between the parties. This requirement, which limits the extent of accommodative duties, does not guarantee the degree of independence that a private-law-libertarian regime would secure. But this is justified because ensuring the independence of one party in these cases implies that the other party to the interaction would be denied both the equal power to determine the terms of interactions and the substantive freedom to act as a self-determining agent.

C. The Complexities of the Public–Private Distinction

Private law does not, and should not, govern the entirety of our so­cial life, but it is hard to deny its prominence in that sphere. Private law plays an integral role in some of our most important social contexts, from family and community to work and commerce. Our account—that of relational justice—appreciates both the significance of horizontal interactions and the impact of private law in shaping those relations. By rejecting critics’ wholesale dismissal of the distinctiveness of private law as a deeply troubling collectivization of the social dimension of life, rela­tional justice captures the rich normative implications that lie beneath the straightforward understanding of private law as the law governing our interpersonal relationships. Our approach underscores the signif­icant role of private law in structuring relations between people as free and equal individuals who are expected to respect one another as the persons they actually are. It highlights, in other words, the intrinsic value of private law, which lies in its minimal requirements of just relationships. This prescription may seem straightforward, but it is hardly so; several factors complicate the translation of these principles into the nuts and bolts of legal doctrine.

Some of these complexities highlight the risks of unreflective re­nouncement of the traditionalist conception. They may also explain the resilience of the traditional public–private law distinction in liberal cir­cles. Other complicating factors have the reverse effect: They suggest both that respecting self-determination and substantive equality means that private law and public law cannot be mutually exclusive and that the scope of private law and the degree to which its institutions comply with these values are contingent. This could explain the persis­tent suspicion among critics of the public–private distinction toward its rigidification. 129 See, e.g., Kennedy, supra note 61, at 1356–57 (arguing that it is impossible “to take the public/private distinction seriously as a description”). Their concerns, as well as those of the traditionalists, are valid and important to some extent; however, the traditionalists’ worries do not justify strict adherence to the conventional public–private dis­tinction, while those of their critics cannot justify the dogmatic repu­diation of the distinction. Instead, the complications explored in this section imply and help explain what lawyers already know: Beyond the abstract articulation of the demands of justice, the legal architecture of private law is complex and its relationship with public law is quite intricate.

1. On Formal Equality and Independence. — One complication arises from the roles of formal equality and independence—the values on which the traditional conception of private law focuses. Consider formal equality first. There are contexts in which formal equality is the all-things-considered best proxy for a state of affairs because the participants in these contexts are typically already situated in a relationship of, more or less, substantive equality. This may explain why the legal treatment of commercial contracts by and large conforms to formal equality. Contract theorists use this conformity to support the claim that formal equality is truly the foundational ideal of contracts in this particular context or in general. 130 See, e.g., Peter Benson, Abstract Right and the Possibility of a Nondistributive Conception of Contract: Hegel and Contemporary Contract Theory, 10 Cardozo L. Rev. 1077, 1172 (1989) (arguing that contracts are predicated upon an idea of “abstract equal­ity” among possessors of a capacity for choice); Markovits, Arm’s Length Relation, supra note 25, at 316–17 (asserting that contracts remain formally reciprocal and egalitarian even when substantively one-sided). But this conclusion does not hold.

Indeed, contract law applies any number of doctrines the basic or­ganizing idea of which is to exclude people whose capacities for contract making and contract keeping fall below a certain threshold for par­ticipation. Some of these doctrines take a categorical form—for instance, minors do not possess the legal personality to make an enforceable pro­mise. 131 See Restatement (Second) of Contracts § 12 (Am. Law Inst. 1981). Other doctrines, such as duress and undue influence, are less rigid but nonetheless manifest hostility toward some transactions based on the concern that one of the parties is not sufficiently competent to make and accept contractual promises. 132 See id. §§ 174–177. The doctrine that exemplifies this most dramatically is unconscionability, 133 See id. § 208. under which contract law ought to protect the vulnerable party—often, the “poor” 134 See Shiffrin, Paternalism, supra note 25, at 206. or the “weak, the fool­ish, and the thoughtless” 135 See S.M. Waddams, Unconscionability in Contracts, 39 Mod. L. Rev. 369, 369 (1976). —if: (1) she could exercise only formal and not “meaningful” choice and (2) the terms of the contract unreasonably favor the other party. 136 See Williams v. Walker-Thomas Furniture Co., 350 F.2d 445, 449–50 (D.C. Cir. 1965). A related example involves the prevailing rule of ex post fairness review of liqui­dated damages. Interestingly, the critics’ position on this rule seems convincing only for contract types in which formal equality is a reliable proxy for substantive equality. See Dagan & Heller, Choice Theory, supra note 68 (manuscript at 128). All these doctrines—as well as other, more covert means that courts use 137 See, e.g., Gordon, Unfreezing Legal Reality, supra note 49, at 209–10 (providing examples). —aim to reduce the risk that the dis­parities between the parties will prevent the contractual engagement from being between genuinely equally situated agents. These doctrines, in other words, constrain the permitted gap between the commitment to substan­tive equality and the use of formal equality as an imperfect yet adequate proxy. 138 Cf. Jedediah Purdy, The Meaning of Property: Freedom, Community, and the Legal Imagination 88, 112 (2010) (arguing that a property system properly designed needs to ensure “rules of recruitment” that reconcile “reciprocity, responsibility, and self-realization”).

To be sure, some or all these doctrines may currently fail to suffi­ciently limit this gap, perhaps due to overrestrictive interpretation by some courts. Yet en if the doctrines are not wholly effective, the struc­tural point remains: Contract law declines to enforce agreements made under circumstances in which formal equality is obviously an inadequate guarantee of substantive fairness. That is, doctrines like incompetency and unconscionability limit the excesses of treating contracting parties as formally equal. 139 Note that this is a structural, rather than empirical, claim: The various doctrines in question may serve as buffers against the excesses of treating parties to a contract as formally equal. They create the doctrinal opening for courts to turn what would otherwise be a freestanding ideal of formal equality into one that is conditional on its (loose) compatibility with substantive equality. 140 See Avihay Dorfman, Private Law Exceptionalism? Part II: A Basic Difficulty with the Argument from Formal Equality 19–21 (2016) [hereinafter Dorfman, Private Law Exceptionalism Part II] (unpublished manuscript) (on file with the Columbia Law Review).

Second, the value of independence also plays an important role in private law. But independence—unlike formal equality—is not a proxy for the realization of some other value; it is a real, albeit not ultimate, value unto its own. Although a liberal system of private law is ultimately committed to self-determination and not independence, it does not, and should not, dismiss or underrate the value of independence. A responsi­ble liberal account of private law must take seriously Isaiah Berlin’s cau­tionary words against too easily overriding people’s independence “in the name, and on behalf, of their ‘real’ selves” and his accompanying prescription that “some portion of human existence must remain inde­pendent of the sphere of social control.” 141 Isaiah Berlin, Two Concepts of Liberty, in Liberty 166, 167, 179 (Henry Hardy ed., 2012). Indeed, independence must be valued by every decent liberal polity. 142 See, e.g., John Stuart Mill, On Liberty 107–44 (1863) (arguing that individualism is an essential component of liberal society). Yet properly safeguarding peo­ple’s independence while keeping in mind that it is self-determination that justifies (and requires) that independence is challenging. In shaping our private law (and especially in constructing our affirmative duties), we must undertake what H.L.A. Hart described as the “unexciting but indis­pensable chore” of distinguishing “between the gravity of the different restrictions on different specific liberties and their importance for the conduct of a meaningful life.” 143 Hart, Between Utility and Rights, supra note 102, at 34–35.

Thus, an autonomy-based private law system is not reluctant to re­strain the independence of some people when its significance to their self-determination is minimal and upholding that independence could jeopardize their (or others’) self-determination or undermine the substan­tive equality among persons. (A particularly vivid example for this comes in the context of private law rules regarding joint projects. 144 See infra section III.C (discussing private law’s support for joint projects). ) But a lib­eral private law would treat people’s independence with greater caution in the absence of strong opposing normative pressure—namely, when there is no threat to self-determination and formal equality roughly approximates substantive equality. Moreover, it would certainly uphold independence when this is crucial for ensuring self-determination.

2. Internal Contextual Factors. — Complicating factors in the transla­tion of the liberal commitment to self-determination and substantive equality into private law doctrine also emerge from contextual considera­tions, both internal and external to the particular social practice at hand. We begin with the internal considerations, which derive from the substantive good (or goods) that the social practice engaged in through the interpersonal interaction is understood to embody or constitute.

Because every practice is supposed to be rationally conducive to the pursuit of its underlying good(s), each such practice has its own internal logic that is typically informative regarding the specific contents of the relationally just terms of interaction in the particular context. In some cases, this logic could do the fine tuning necessary for turning the ab­stract injunction of just relationships into a workable set of rules. Here, contextual considerations will render intelligible our judgments concern­ing what it is for people to be in relationships of substantive freedom and equality by specifying, for example, the personal qualities that should be determinative in setting the terms of the particular category of interac­tion and how decisive they should be. (The law pertaining to the negli­gent infliction of physical harm will render vivid this point: The relevant qualities are the victim’s physical, mental, and cognitive disabilities, whereas other personal qualities do not warrant accommodation because they are typically irrelevant to our practice of transportation. 145 See infra section III.A (discussing these factors). )

In other categories of cases, context rules out the possibility of reconciling a particular practice with these liberal commitments, requir­ing that we consider discarding the practice or at least transforming it substantially. In some contexts, the reason will be the repressive nature of a practice: Slavery is an obvious example of a practice indisputably unde­serving of a charitable transformation. But in small-scale instances of flatly illiberal social practices, the option of transformation is often quite attractive. 146 Consider, for example, how feminist insights, which highlight violations of rela­tional justice, have helped reform many of our social practices. See, e.g., William N. Eskridge, Jr., Sexual and Gender Variation in American Public Law: From Malignant to Benign to Productive, 57 UCLA L. Rev. 1333, 1334 (2010) (discussing “the social and intel­lectual forces pressing Americans toward the notion . . . that most variations [in gender, sex, and sexuality] are tolerable, and . . . ought to be recognized as entirely benign”).

Finally, in other cases, private law’s commitment to the ideal of just relationships will be inconsistent with the very point of the particular practice (which in itself is grounded on perfectly valid liberal found­ations). This may explain, and even justify, a robust practice of freedom of expression and the privileges it grants to participants to ridicule and even harm others. It may help explain private law’s tolerance of freedom of expression even when it manifests itself in complete disregard of the other’s personal qualities, which is to say their judgments, character traits, and personal circumstances (such as race). 147 See, for example, Snyder v. Phelps, 562 U.S. 443 (2011), in which the Supreme Court revoked damages for intentional infliction of emotional distress caused by undenia­bly harsh and harmful speech. To be sure, a robust practice of freedom of expression may be controversial (especially if the relevant torts, such as defamation, are best viewed as protecting human dignity, rather than merely reputation); however, the free speech illustration demonstrates that there can be liberal practices whose animating good brings pressure to bear against the normative commitments that generally inform relationally just terms of interaction. Arguably, a structurally similar observation can be made with respect to some of the economic harms generated by moderately regulated economic competition among market participants. 148 See Stephen R. Perry, Protected Interests and Undertakings in the Law of Negligence, 42 U. Toronto L.J. 247, 265 (1992) (“[I]n a market-based economy in a liberal society individual economic interests are inevitably subject to a broad range of interfer­ence . . . by other persons.”).

3. External Commitments. — Alongside considerations internal to the practice at hand, external commitments of the liberal state—both normative and pragmatic—may also place constraints on the conception of relational justice in private law. A liberal (as opposed to libertarian) state should be committed to the demands of both distributive justice (which focuses on justice in holdings) 149 See, e.g., Dworkin, Sovereign Virtue, supra note 23, at 12 (“Distributional equal­ity, as I describe it, is not concerned with . . . rights other than rights to some amount or share of resources.”); Rawls, Theory of Justice, supra note 127, at 61 (“The second princi­ple [of justice] applies . . . to the distribution of income and wealth.”). and democratic citizenship (which seeks to eradicate hierarchies in our relationships qua citizens). 150 See, e.g., Iris Marion Young, Justice and the Politics of Difference 16 (1990) (“The concepts of domination and oppression, rather than the concept of distribution, should be the starting point for a conception of social justice.”). An adequate conception of the public–private distinction must address private law doctrines that may undermine these commitments. 151 External commitments to distributive justice and democratic citizenship also help explain why public law cannot and should not be conceptualized solely around a notion of relational justice. To con­tend with such troublesome ramifications, such a conception could apply second-order considerations to adapt the doctrinal framework so that it responds to these concerns, while still meeting the demands of relational justice through private law. One way of achieving this is to restrict individual responsibility by shifting some of the burden onto public law, thereby preventing or limiting the conflict with distributive or democratic commitments. Similar intermediate solutions could be justified for pragmatic reasons, for example, when considerations of efficacy pull toward collectivizing the legal regulation of an essentially horizontal interaction. 152 Consider, for example, the law governing work-related accommodation. Here there is a significant relational duty. See, e.g., 42 U.S.C. § 2000e-2 (2012) (prohibiting employers from discriminating on the basis of enumerated characteristics of the employee).

Moreover, because there may be some overlap between the public responsibilities to ensure self-determination and substantive equality and the private obligations that our conception of relational justice entails, private law should be wary of diluting public responsibilities. Private law’s commitment to relational justice, in other words, should not be inter­preted as necessarily exhausting or supplanting these public responsibili­ties and the state obligations they entail. This is most acutely so in contexts in which satisfying relational justice in the legal implementation of an interpersonal practice can only be achieved through a private law doctrine constructed on top of a public law regulatory infrastructure (as in the law of consumer transactions). 153 See Joseph William Singer, No Freedom Without Regulation: The Hidden Lesson of the Subprime Crisis 58–95 (2015). It is also relevant in cases in which the primary responsibility should be private and relational—child sup­port, for example—but fulfilling the parallel public responsibility re­quires that the state assist in enforcing the relational responsibilities or even provide some insurance against noncompliance with them. 154 Cf. Anne L. Alstott, No Exit: What Parents Owe Their Children and What Society Owes Parents 49–72 (2004) (discussing the state’s role in incentivizing proper paternal care). See generally Jane Millar & Andrea Warman, Family Obligations in Europe 24 (1996) (outlining, by country, the various approaches to state intervention in the context of parental obligations toward dependent children); Robert I. Lerman & Elaine Sorensen, Child Support: Interactions Between Private and Public Transfers, in Means-Tested Transfer Programs in the United States 587 (Robert A. Moffitt ed., 2003) (discussing the state’s role in enforcing child support obligations).

Yet, as the child support example suggests, there are clear cases in which private law’s terms of interactions should not be enlisted to serve purely public ends. Consider the simple tort duty against committing as­sault and battery. Could it be morally possible to eliminate its relational dimension for the sake of advancing some collectivist end, however desirable it might be? Other features of private law are no less crucial for it to remain the practical domain of relational justice—for example, it is hard to imagine an acceptable legal regime that would eliminate all the normative powers allowing persons to create contractual obligations and control external objects as owners. 155 Cf. Avihay Dorfman, Private Ownership, 16 Legal Theory 1, 12–16 (2010) (discussing the Marxist conception of personal property as an illustration of a legal regime that eliminates such normative powers).

Lastly, we do not deny that in some contexts a private law framework can be legitimately enlisted to serve irreducibly public values, whereby the state commandeers the support of private individuals to enhance col­lective goals. The incentives set by copyright law and patent law, for example, can (arguably) be understood in terms of delegating society’s collective interest in fostering culture, research, and development to pri­vate individuals and firms. 156 See, e.g., U.S. Const. art. I, § 8 (“The Congress shall have Power . . . [t]o promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” (emphasis added)). When private bodies are thus publicly en­listed to serve a public role, privatization-or-collectivization debates can properly revolve around considerations of comparative institutional competence. Nonetheless, as long as these public values are promoted through private law doctrines, it is still meaningful, indeed important, to evaluate not merely their (external) regulatory performance but also their (intrinsic) performance: whether they establish, facilitate, and sus­tain the ideal of just relationships.

III. The Law of Just Relationships

We are now ready to move from legal theory to legal doctrine. By highlighting the significance of the ideal of just relationships to contemporary private law, we hope to demonstrate what gets lost if the public–private distinction is completely rubbed out or, alternatively, if we accept the traditional conceptualization of private law as a fortress of independence and formal equality. So we turn to four broad areas that exemplify the fact of interdependence and taken together, encompass significant portions of private law.

For each case study, private law casts (as it should) interpersonal interactions as frameworks of relationships between self-determining individuals who respect each other as the persons they actually are. These case studies show that a private law that adheres to the ideal of just relationships places demands on the conduct of private individuals in particular and that these demands are necessary for people to be in rela­tionships of genuine freedom and equality. Moreover, such demands are not, and certainly need not be, overburdening; it is possible to safeguard against excessive infringements of independence without subscribing to the libertarian freedom-as-independence school of thought.

Finally, the case studies provide opportunities to explore the complexities of the public–private distinction and illustrate how contex­tual considerations refine the incorporation of relational justice into the actual operation of private law. They demonstrate why neither of these complexities justifies discarding the intrinsic value of private law by conceptualizing it either as the “law for persons regarded as ends outside of human association—as morally self-sufficient atoms” 157 Brudner, supra note 11, at 353. (as the traditionalists posit) or as just another garden-variety mode of public regulation (as some critics assert). Although resolving the evaluative questions our account raises requires judgment and entails contextual considerations, this does not strip it of significance. This may alarm formalists, yet for us it is a rather banal truism that reflects the phenome­nology of arguing about law in a particular, detailed context.

A. Accidental Harm to Life and Limb

The fact of interdependence implies that the possibility of leading a good life requires a sustained effort by society to mitigate the negative side effects of people’s otherwise legitimate pursuit of ends. For instance, going to visit a friend may involve acts, such as driving or crossing the road, that expose oneself and others to accidental but substantial risk of harm—the most prominent type being physical harm (including death). It is not surprising that a—or the—paradigmatic tort in many developed countries since the days of the industrial and automobile revolutions has been the negligent infliction of loss to life and limb. 158 See, e.g., Arthur Ripstein, Equality, Responsibility, and the Law 48 (1999) [hereinafter Ripstein, Equality]; G. Edward White, The Unexpected Persistence of Negligence, 1980–2000, 54 Vand. L. Rev. 1337, 1346 (2001). The tort of negli­gence responds to the problem of accidental harm by establishing fair terms of interaction—standards of care—between individuals. These terms respond to several demands, including the preservation of the equal freedom of those involved and the generation of incentives to take cost-justified precautions. What makes this peculiar response particularly challenging is the fact of personal difference, since the actual compe­tency to constrain risky conduct may vary radically across individuals. 159 The locus classicus is Oliver Wendell Holmes, Jr., Lecture III, Torts: Trespass and Negligence, in The Common Law 71, 71–117 (Harvard Univ. Press 2009) (1881).

To understand what could count as relationally just terms of interac­tion in these contexts, consider the case of a person with diminished mental capacity who is hit by a car while crossing the street. The victim’s disability can affect the terms of the interaction and ultimately, the resolution of this case in two important ways. First, it can partially deter­mine whether the injurer’s conduct is negligent at all. Any nonarbitrary attempt, for example, at identifying the “reasonable” speed limit presup­poses a prior judgment of what counts as reasonable conduct on the part of a potential victim reacting to an approaching car. In other words, the method of assessing the responding victim’s conduct partially constitutes the contents of the duty of care owed by the potential injurer. Second, the victim’s disability may, under the doctrine of comparative negligence, determine the scope of the liability that can be imposed on a negligent injurer: Excluding the disability as a relevant consideration reduces the injurer’s scope of liability, and vice versa.

Establishing the terms of interaction between injurers and victims re­quires determining which qualities and circumstances an injurer should accommodate and which should be excluded. Relationally just terms of interaction require that, subject to the existing conventional tort-law thresholds, such as reasonable foreseeability, the duty of care owed by an injurer to a victim be partially set by the latter’s mental or physical capac­ity. 160 Under the requirement of reasonable foreseeability, the amount of care required accounts for the possible presence of disabled persons as a matter of statistical foreseeabil­ity (reflecting the frequency and distribution of a given vulnerability across society). Tortfeasors are not expected to know the exact numbers, but they are certainly expected to be aware of—and be open to recognizing—the very existence of vulnerable persons in their society and of the possibility that at least one of them might be within the zone of foreseeable danger relevant to the tortfeasors’ risky conduct. See, e.g., Haley v. London Elec. Bd. [1965] AC 778 (HL) 805 (appeal taken from Eng.) (“In view of the large num­ber of blind persons who . . . are users of the road it cannot be said that the risk of causing them injury is so small as to be minimal and therefore to be excluded from the realm of foreseeability.”). The injurer must be held responsible to take extra care—that is, incur additional costs—to protect the mentally disadvantaged person, rather than merely the nondisadvantaged person, from the injurer’s dan­gerous activity. This rule, which current law by and large applies, 161 See, e.g., Noel v. McCaig, 258 P.2d 234, 241 (Kan. 1953) (“Since knowledge and appreciation of the peril are essential elements of contributory negligence, it is obvious that any inquiry into the age, experience, and mental capacity of the plaintiff is material where contributory negligence is invoked as a defense.”); Johnson v. Primm, 396 P.2d 426, 430 (N.M. 1964) (“If plaintiff was not a rational being and such condition proximately resulted from defendant’s negligence, her conduct is not to be judged by the same stand­ards as would apply to an ordinary or average adult . . . .”); Campbell v. Cluster Hous. Dev. Fund Co., 668 N.Y.S.2d 634, 635 (App. Div. 1998) (“The degree of reasonable care is meas­ured by the plaintiff’s physical and mental infirmities, as known by the defendants . . . .”); Stacy v. Jedco Constr., Inc., 457 S.E.2d 875, 879 (N.C. Ct. App. 1995) (“[O]ne whose men­tal faculties are diminished, not amounting to total insanity . . . is not held to the objective reasonable person standard. Rather, such a person should be held only to . . . the standard of care of a person of like mental capacity under similar circumstances.”). re­flects the proper understanding of a just relationship characterized by reciprocal respect and equal self-determination.

This analysis of relational justice challenges the symmetrical treat­ment of the parties under the economic analysis of law, 162 See, e.g., William M. Landes & Richard A. Posner, The Economic Structure of Tort Law 124, 126 (1987). For more on the tension between negligence law’s asymmetric treatment of plaintiff and defendant fault, on the one hand, and the economic analysis of tort law, on the other, see Avihay Dorfman, Negligence and Accommodation passim (May 9, 2016) [hereinafter Dorfman, Negligence and Accommodation] (unpublished manu­script) (on file with the Columbia Law Review). given the qualitative difference between the respective vulnerabilities of victim and injurer and the corresponding significance this difference is accorded in tort law. 163 See supra text accompanying notes 110–112 (discussing these qualitative differ­ences); see also Gregory C. Keating, Is Cost-Benefit Analysis the Only Game in Town? 18 (June 11, 2016) (unpublished manuscript) (on file with the Columbia Law Review). Compare this to tort law’s symmetrical treatment of the parties in categories of cases that involve qualitatively similar vulnerabilities, as reflected in the doctrine of nuisance. Under this doctrine, the plaintiff’s idiosyncratic sensitivities are irrelevant to determining whether the defendant unreasonably interfered with the plaintiff’s use and enjoyment of her land. See, e.g., Langan v. Bellinger, 611 N.Y.S.2d 59, 59–60 (App. Div. 1994) (requiring “objec­tive evidence” of nuisance); W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 88, at 628 (5th ed. 1984) [hereinafter Prosser & Keeton]. It also stands in sharp contrast to the requirement in private law’s traditional conception of upholding the parties’ independence and formal equality. This requirement implies an objective standard of due care, one that disallows consideration of the victim’s idiosyncratic charac­teristics (including diminished mental capacity) to unilaterally figure in determining the terms of the parties’ interaction, thus exempting the injurer from attending to her victim’s special circumstances. 164 See Coleman & Ripstein, Mischief and Misfortune, supra note 29, at 112 (“Be­cause fault is supposed to measure the costs of activities fairly and across individuals, it cannot be understood subjectively in terms of good faith efforts at care.”); see also Weinrib, Idea of Private Law, supra note 38, at 169 n.53, 183 n.22 (“[T]he plaintiff cannot demand that the defendant should observe a greater care than the plaintiff with respect to the plaintiff’s safety.”); Ripstein, Civil Recourse and Separation, supra note 29, at 181 (“The unusually sensitive plaintiff gets no solace from the law.”).
In an attempt to reconcile the traditional view with a departure from the objectively fixed standard of due care, Ripstein argues that a plaintiff’s physical disability could be allowed to partially determine the amount of care owed by the defendant. See Ripstein, Equality, supra note 158, at 111–13. Comparing a physically disabled plaintiff to icy road conditions, Ripstein claims that allowing the sensibility of the disabled plaintiff to adjust the amount of care required of the defendant is as justifiable, in terms of formal equality, as adjusting the level of care to accommodate the changing road conditions. In fact, he argues, both a plaintiff’s physical disability and icy road conditions are compatible with an objectively fixed standard of due care. Id.
Ripstein’s analogy, however, does not hold up. The concept of formal equality cannot coherently be applied to inanimate road conditions; rather, it is intelligible only with re­spect to the sensitivities and conditions of a human agent. Concerns of inequality can only arise when the law allows for the idiosyncratic features of one person to determine the ex­tent of care owed by another. To be sure, we do not argue that a person with diminished mental or physical capacity herself sets the terms of the interaction by exercising some pseudo-authority to decide whether to act. See id. at 112 (discussing the case of voluntary intoxication). Rather, the point is that the law fixes terms of interaction that are incon­sistent with the formal equality of the parties when such terms demand that one party tolerate the other’s idiosyncratic features. Ripstein’s failed analogy takes the traditional approach back to its point of departure: It disallows plaintiffs’ idiosyncratic features to count in fixing the terms of their interactions with defendants. See Dorfman, Negligence and Accommodation, supra note 162 (manuscript at 2) (discussing the law’s asymmetric treatment of plaintiffs and defendants).
In so do­ing, the traditional view fails to respect the victim on her own terms—that is, her sensibilities.

Traditionalists who acknowledge the offensiveness of burdening the victim with the entire cost of her own particular circumstances will likely assert that it is entirely the state’s responsibility—through a public law solution such as a national insurance scheme—to rectify the excesses of their conception of private law. 165 Even libertarians seem to subscribe to this position. See, e.g., Nozick, supra note 14, at 78–79, 82–83, 87, 115. Interestingly, critics of the public–pri­vate distinction reach a similar conclusion: For these scholars (especially from the law-and-economics school), the identity of the agent responsi­ble for the necessary accommodation—whether the private injurer or the state—is a matter of institutional design. 166 See Guido Calabresi, Ideas, Beliefs, Attitudes, and the Law 39–40, 66–67 (1985) (critiquing the traditional common law approach to risk-bearers and proposing an alterna­tive method of accommodation); see also Gregory C. Keating, The Priority of Respect over Repair, 18 Legal Theory 293, 314 (2012) [hereinafter Keating, Priority of Respect] (comparing compensation methods under tort and administrative regimes).

Indeed, the key difference between the traditional and critical ap­proaches is that the former dismisses relational justice, whereas the latter renders it wholly contingent. Neither approach, however, takes relational justice seriously. The potential injurer and victim cannot be in a just rela­tionship without allowing the diminished capacity of the victim to have some measure of influence on motorists’ duty to moderate their risky activity when they approach the potential victim. 167 Indeed, relationally just terms of interaction require imposing some of the costs that emanate from one party’s circumstances onto the other party to the interaction. Despite its distributive implications, this cost-internalization requirement is anchored in a concern for the terms of the relationships between individuals, not in considerations of justice in the holdings of persons taken severally. This is why the requirement to accommo­date a disabled person’s vulnerability does not draw on the distinction between the per­son’s brute luck and option luck, nor even between her bad luck and personal choice. The distinction between brute luck and option luck is made famous in Dworkin, Sovereign Virtue, supra note 23, at 73. Overlooking the vic­tim’s special makeup and circumstances in determining interpersonal duties in such cases is incompatible with an ideal of relating as genuine equals.

Filling in the contents of the accommodative structure of negligence law raises, of course, concerns beyond the context of diminished mental capacity. There are myriad contexts in which questions arise regarding the appropriate scope and extent of accommodation. The characteristics of the relevant practice help answer some of these questions. Most im­portantly, the personal qualities that should be accommodated by a duty of reasonable care must be connected to the kind of interdependence that brings the injurer and victim together. In this context, the relevant qualities are those associated with both the ability to decide where and when to cross the street and the competency to respond to the surround­ing environment (notably, approaching cars). Physical, mental, and cog­nitive disabilities are the first to come to mind, along with other forms of insufficient ability to adapt oneself to the potential risks of the road. 168 For familiar examples, see, e.g., Vaughan v. Menlove (1837) 132 Eng. Rep. 490, 492; 3 Bing. (N.C.) 468, 472 (noting that prudence varies with the faculties of men); Holmes, supra note 159, at 99 (describing the person who was “born hasty and awkward”); Prosser & Keeton, supra note 163, § 32, at 176 (describing the person suffering from “weaknesses of old age”); Richard A. Posner, A Theory of Negligence, 1 J. Legal Stud. 29, 31 (1972) (describing the person who is “clumsier than average”). Other personal qualities, by contrast, may not warrant accommodation because they do not reasonably connect to the sort of interdependence relevant to the practice at hand. For instance, it makes no sense to take into account the victim’s political sensibilities or hobbies in setting the contents of the injurer’s duty of care.

Moreover, contextual considerations can constrain the extent of the required accommodation. Inherent in the idea of just relationships is a stopping point to the duty to accommodate the circumstances of the potential injurer. 169 See supra text accompanying notes 127–128 (discussing the accommodative duty’s inherent stopping point). It is self-defeating to convert the injurer into a mere instrument for respecting the victim as a free and equal person. The no­tion of accidental harm implies that the risks created are incidental to the injurer’s pursuits of an otherwise legitimate end and not for the illegitimate purpose of putting others at risk. Accordingly, a requirement to take extraordinary care toward victims like the mentally disadvantaged might adversely affect the injurer’s autonomy to pursue worthwhile ends. Therefore, although an accommodative duty of care should be costly for an injurer to discharge in light of—and in recognition of—the victim’s peculiar sensibilities, it must not be prohibitively so. 170 There will also be harder, though not intractable, cases. For instance, should the same analysis apply to the case of a disabled injurer? See Dorfman, Negligence and Accommodation, supra note 162 (manuscript at 32–37). Should the accommodative struc­ture of negligence law be sensitive to the choices of victims or to their conceptions of the good? See id. (manuscript at 44–46). Can considerations other than relational justice, such as those of distributive justice, override the demands of an accommodative duty of care? See Gregory C. Keating, Rawlsian Fairness and Regime Choice in the Law of Accidents, 72 Fordham L. Rev. 1857, 1870–86 (2004) (discussing accident law from a fair­ness perspective); Ariel Porat, Misalignments in Tort Law, 121 Yale L.J. 82, 97–107 (2011). And so on. It is beyond the scope of this Article to address these questions.




There are two points worth considering before proceeding to the next category of cases. First, our analysis, like the traditional approach, 171 See supra note 38 and accompanying text. presents negligence law as a straightforward expression of the commit­ment to relationally just terms of interaction in the context of accidental harm to life and limb. Contrary to the traditionalists, however, we do not argue that negligence law is essential for this task. Relational justice only requires that the injurer be subject to an obligatory reason to accommo­date, within limits, the person the victim actually is. It follows that compliance with a reason to accommodate can, if necessary, be secured without the support of the adjudicatory and remedial aspects of contemporary negligence law. This conclusion could be particularly significant if these aspects turn out to be flawed in terms of public values, such as distributive justice or social welfare, or even private law values insofar as the existing private law institutions of adjudication fail to re­spond effectively to the increasing demand for dispute resolution.

For instance, New Zealand famously repudiated most aspects of the traditional tort of negligent infliction of physical harm and created, in­stead, a public insurance scheme subsidized by the general tax coffers. 172 See, e.g., Peter Davis et al., Compensation for Medical Injury in New Zealand: Does “No-Fault” Increase the Level of Claims Making and Reduce Social and Clinical Selectivity?, 27 J. Health Pol. Pol’y & L. 833, 835–36, 851–52 (2002). The insurance system presumably promotes distributive justice and according to some studies, attains efficacy. 173 See, e.g., id. at 851–52; see also Craig Brown, Deterrence in Tort and No-Fault: The New Zealand Experience, 73 Calif. L. Rev. 976, 1002 (1985). Tort theorists often invoke the New Zealand case as exemplifying a radical transition from a legal order grounded on private law to one grounded exclusively on public law. 174 See, e.g., Brudner, supra note 11, at 270; Robert Stevens, Torts and Rights 324 (2007); Coleman & Ripstein, Mischief and Misfortune, supra note 29, at 128 n.56. But these scholars overstate the shift in question and, moreover, fail to acknowledge that negligence law’s traditional scheme of adjudica­tion is not a necessary condition for securing just relationships in the context of risks to life and limb. 175 Cf. Calabresi, A Broader View, supra note 64, at 2 (citing New Zealand as an example of how tort law has both a public and private function). This is because New Zealand has not abolished the legal doctrines—particularly injunctive relief and punitive damages—that ensure compliance with the reason for discharging the accommodative duty of care. Thus, a victim can seek punitive damages against an injurer who actively disregards the reason she must have (independent of tort law) for accommodating the victim by exercising appropriate care. Until recently, this same doctrine had been applied even in cases of negligence such as medical malpractice and not only in assault and battery circumstances. 176 See Susan Couch v. Attorney Gen. [2010] NZSC 27, [2010] 3 NZLR 149 at [2].

A somewhat similar analysis holds for the typical workers’ compensa­tion scheme, which generally prescribes that injured workers are not enti­tled to sue their employers or coworkers for work-related injuries. 177 See, e.g., Iowa Code § 85.20 (2016) (noting the “rights and remedies” arising under the State’s workers’ compensation act are “exclusive”). However, the nonapplication of tort law in this context has not elimi­nated the interpersonal tort duty of care employers owe to employees; this is crucial for the terms of the employer–employee interaction to count as relationally just. 178 But see Brudner, supra note 11, at 315 (viewing schemes of “social insurance” as tort law’s antithesis); John C.P. Goldberg & Benjamin C. Zipursky, Torts 394–95 (2010) (arguing that there is “something troubling” about workers’ compensation programs that “place their beneficiaries in the position of supplicants asking for benefits rather than litigants pursuing claims as a matter of right”); Gregory C. Keating, The Theory of Enterprise Liability and Common Law Strict Liability, 54 Vand. L. Rev. 1285, 1295–96, 1317 (2001) (“Employment related accident costs should not remain concentrated on employ­ers. They should be dispersed across all those . . . who benefit from the imposition of the enterprise’s characteristic risks.”). Workers’ compensation schemes do not strip employees of their tort law right to bodily safety, under which they can compel their employers to ensure a reasonably safe work environment. Indeed, the duty of reasonable care can serve as the basis for enjoining the employer from exposing employees to an unsafe workplace. 179 See, e.g., Smith v. W. Elec. Co., 643 S.W.2d 10, 12–13 (Mo. Ct. App. 1982) (hold­ing that the plaintiff had adequately pleaded that his employer breached the duty of reasonable care and that injunctive relief would be an appropriate remedy). Moreover, workers’ compensation schemes do not release employers from their tort liability for any injury caused by their nonaccidental (i.e., intentional or reckless) misconduct. 180 See Johns-Manville Prod. Corp. v. Contra Costa Superior Court, 612 P.2d 948, 956 (Cal. 1980); Reeves v. Structural Pres. Sys., 731 So.2d 208, 211 (La. 1999); Delgado v. Phelps Dodge Chino, Inc., 34 P.3d 1148, 1156 (N.M. 2001).

Finally, taking a relational justice perspective on negligent infliction of physical harm can change the terms of one of the most fundamental debates in private law theory. Most leading noneconomic accounts of tort law presume that tort law expresses a commitment to either corrective justice or distributive justice (or a mix of both). 181 See, e.g., Jules L. Coleman, Risks and Wrongs 350–54 (1992) (approaching the question of tort law’s justice as giving rise to a choice between two forms of justice: correc­tive and distributive justice); Weinrib, Idea of Private Law, supra note 38, at 70, 72–75 (same); Peter Benson, The Basis of Corrective Justice and Its Relation to Distributive Justice, 77 Iowa L. Rev. 515, 540 (1992) (same); Gregory C. Keating, Distributive and Corrective Justice in the Tort Law of Negligence, 74 S. Cal. L. Rev. 193, 197 (2001) (same); Stephen R. Perry, On the Relationship Between Corrective and Distributive Justice, in Oxford Essays in Jurisprudence 237 (Jeremy Horder ed., 4th series 2000) (same). Whereas corrective justice is founded on a noncomparative conception of equality among formally free persons, distributive justice in tort law concerns the fair allocation of the costs of accidents according to some measure of merit. 182 See, e.g., Guido Calabresi, Some Thoughts on Risk Distribution and the Law of Torts, 70 Yale L.J. 499, 502, 532–33 (1961); Gregory C. Keating, The Idea of Fairness in the Law of Enterprise Liability, 95 Mich. L. Rev. 1266, 1267 (1997). Some liberal egalitarians who find the implications of corrective justice for tort law normatively disappointing are drawn to its competitor, distributive justice. 183 See Keating, Priority of Respect, supra note 166, at 319 (noting that “the [basic tort law] question of what rights people have is a question of distributive justice.”). Other liberal egalitarians, less skeptical of correc­tive justice’s moral underpinnings, suggest that tort law’s commitment to equality cannot be evaluated apart from the distributive patterns to which it gives rise or otherwise sustains. 184 See Peter Cane, Distributive Justice and Tort Law, 2001 N.Z. L. Rev. 401, 412; John Gardner, What Is Tort Law For? Part II: The Place of Distributive Justice, in Philosophical Foundations of the Law of Torts 335, 337 (John Oberdiek ed., 2014); Hanoch Sheinman, Tort Law and Distributive Justice, in Philosophical Foundations of the Law of Torts, supra, at 354, 372.
Under this orientation, corrective justice is an especially interesting context in which the law attends to distributive justice to a limited extent. See Gardner, supra, at 344. In this way, corrective justice enjoys “explanatory priority” over distributive justice, with the pur­suit of the latter “incidental” to tort law’s task of correcting injustices. Id.; Sheinman, su­pra, at 379–80. It is an open question whether and how this trend is inconsistent with the theory of corrective justice. See Avihay Dorfman, Philosophical Foundations of the Law of Torts, Notre Dame Phil. Revs. (Jan. 21, 2015), http://ndpr.nd.edu/news/55240-philosophical-foundations-of-the-law-of-torts/ [http://perma.cc/G7P8-JFHP].

The dichotomization of corrective justice and distributive justice and, by extension, the debate over whose side tort law ought to take is misguided: Relational justice represents a nondistributive conception of substantive, rather than formal, justice. Relational justice can, therefore, render tort law’s aspiration to do justice between persons both intelligi­ble and normatively attractive—in a way that distributive justice’s collec­tivistic aspirations and commitment to formal freedom and equality (respectively) cannot. 185 For further development of this point, see Dorfman, Private Law Exceptionalism Part II, supra note 140 (manuscript at 15–19).

B. Residential Dwellings and the Workplace

The first category of cases showed how our account of private law can inform and support the accommodative structure of tort law. Moreo­ver, these cases illustrate both the significance of contextual considera­tions and the persistence of the ideal of just relationships even when, for distributive or pragmatic reasons, the legal regulation of the activity at hand is largely collectivized—namely, when a regulatory, public law doc­trine has taken the lead. We focus now on the manifestations of private law’s accommodative structure in two other areas, property and contract, in the respective contexts of residential dwellings and the workplace. The dual purpose is to highlight the implications of our theory in these key doctrines as well as discuss certain contextual considerations that compli­cate matters but do not jeopardize the intrinsic value of private law.

1. Residential Dwellings. — Residential dwellings are understood in contemporary society as a person’s paradigmatic safe haven, as a bastion of individual independence, as shielding us from the demands of others and from the power of the public authority, and as providing us with an almost sacrosanct private sphere that serves as a prerequisite to our per­sonal development and autonomy. 186 See, e.g., Margaret J. Radin, Property and Personhood, 34 Stan. L. Rev. 957, 991–92 (1982) (“The home is a moral nexus between liberty, privacy, and freedom of association.”). While it may be unethical for an owner to refuse to let another into her home on grounds of religious objection, society defends the owner’s right to do so and conceptualizes such nonaccommodative behavior as the inevitable result of our residen­tial practices. 187 See Hanoch Dagan & Avihay Dorfman, The Human Right to Private Property, 18 Theoretical Inquiries L. (forthcoming 2017) (manuscript at 8–9), http://ssrn.com/abstract=2624428 (on file with the Columbia Law Review) (“[A] private property owner enjoys some measure of liberty to posit her subjectivity—her intention, judgment, and, indeed, point of view—as a source of legal claims over anyone else.”). In this context, the very point of our residential practices implies ruling out accommodation.

But the ownership of a residential dwelling also includes other normative powers that do not exclude commitments to relational justice. Suppose that A is interested in selling her dwelling or leasing it out but refuses to accept B as her buyer (or lessee) only because of B’s religious persuasion. Or suppose that B wishes to purchase A’s unit in a common interest development, but the board withholds its consent to the sale on racial grounds. Because buying or leasing a dwelling implies the fact of our interdependence, they expose certain classes of people—recall the fact of personal difference—to discrimination by some homeowners and landlords. The law justifiably interferes by insisting that these interac­tions be consistent with the demands of relational justice. Deciding on one’s residence is often a major act of self-authorship that plays an im­portant role in people’s construction of their ground projects. 188 See supra note 110 and accompanying text (discussing the concept of ground projects). The requirement that the parties recognize each other as substantively free and equal persons does not undermine the point of these residential practices.

The objection to discriminatory practices in this context seems indis­putable. Yet our theory nonetheless sharpens its private law implications by focusing on whether the responsibility in question must (at least also) be borne by the private seller or lessor. The positions of both advocates and critics of the traditional conception of private law are surprisingly similar and unsurprisingly disappointing.

Critics of the public–private distinction and scholars indifferent to it (e.g., lawyer-economists) are bound to treat the identity of the agent re­sponsible for eliminating discrimination in selling or renting residential dwellings solely as a matter of institutional design. For them, what mat­ters is that, at the retail level, members of groups that are discriminated against must enjoy fair equality of opportunity in their efforts to buy or rent the dwellings they prefer and at the wholesale level, that residential dwellings be sufficiently integrative. 189 See generally Elizabeth Anderson, The Imperative of Integration chs. 6–7 (2010) (defending the democratic egalitarian case for racial integration in various social domains, including housing).

Traditionalists, in turn, may be able to show that even if private law is founded on the thin commitment to independence and formal equality, there may be circumstances that justify stripping owners of their entitle­ment to exclude potential buyers. This is the case when nonowners do not have sufficient housing opportunities available to them, so that allow­ing owners to make their selling or renting decisions based on discriminatory considerations would make nonowners “fully subject” to the choices of these owners. 190 See Ripstein, Force and Freedom, supra note 26, at 292 (“The state cannot make an arrangement for a person inconsistent with his or her rightful honor. Therefore the state cannot set up a system of property that would allow one person to become fully sub­ject to the choice of another.”). But since, in principle, private owners and landlords neither exhaust nor control the supply of residential dwellings, there is no relationship of entailment between discriminatory practices on the part of owners and landlords and a state of dependence on the part of nonowners. 191 Our argument is not that a state of dependence cannot ever arise out of private owners’ discriminatory attitudes but rather that it cannot arise in any systematic manner as long as the state, acting on its duty to support the poor, provides—directly or via incen­tives—housing alternatives that sustain equality of opportunity for all nonowners.

Thus, under both the traditionalist and critical accounts, the prohibition against obvious discrimination by private owners is neces­sarily contingent: It depends on the extent to which the state carries out its responsibility to eliminate racial injustice in the context of residential dwellings.

By contrast, our theory of just relationships lays down a firmer, principled ground for this prohibition: Refusing to consider a would-be buyer on racial (or other discriminatory) grounds fails to respect this person on her own terms and so does not relate to her as a free and equal individual.

Relationally just terms of interaction between persons engaging in the context of buying or renting residential dwellings mandate that own­ers and landlords set aside certain considerations, such as their racist preferences, when making selling or renting decisions. In order for the involved parties to relate as free and equal individuals, the would-be buyer should not bear the consequences of adverse assumptions that owners assign to her based on the personal qualities she actually pos­sesses (or is even perceived 192 Arguably, discrimination is often an artifact of perception and can be harmful to people who are not, in fact, members of the class of persons the discriminator has been targeting. as possessing). 193 Some of the adverse consequences (or costs) mentioned are the upshot of the owner’s sheer biases, but they may also be associated with the economic value of the prop­erty. In any event, the duty to accommodate includes the accommodation of both kinds of adverse consequences, for it does not turn on whether or not it is economically rational to discriminate. And regardless of whether the state supplies sufficient housing options while sustaining integrative residential communities, private law must not leave intact (and thereby authorize) social relationships that violate the equal standing and the autonomy of the person subjected to discrimination. There is no way around this imperative to establish relationally just terms of interaction among persons engaging in buying or renting residential dwellings. The various pieces of fair housing legislation at both the federal and state levels, which prohibit discrimination in the sale or rental of residential dwellings based on such considerations as race, gender, nationality, religion, disability, familial status, and sexual orientation, 194 See, e.g., Fair Housing Act, 42 U.S.C. §§ 3601–3619 (2012). properly implement this prescription.

Our account also shows that the Supreme Court decision in Shelley v. Kraemer 195 334 U.S. 1 (1948) (holding that although state enforcement of private agree­ments to exclude people from residential dwellings based on race violates the Fourteenth Amendment, the underlying discriminatory private agreements do not). failed to acknowledge the existence and importance of the private law dimension of substantive equality. The Court held that judi­cial enforcement of racially restrictive covenants amounts to a violation of the vertical dimension of substantive equality—the dimension that captures the relationship between the state (acting through the courts) and the persons excluded by such covenants. 196 See id. at 20 (“We hold that, in granting judicial enforcement of the restrictive agreements in these cases, the States have denied petitioners the equal protection of the laws, and that, therefore, the action of the state courts cannot stand.”). The Fourteenth Amendment’s Equal Protection Clause constitutes the doctrinal expres­sion of this proposition. 197 See U.S. Const. amend. XIV, § 1 (establishing that no state shall “deny to any person within its jurisdiction the equal protection of the laws”). However, resort to constitutional law alone misses the significance of relational justice that ought to govern the terms of the interaction between the individual persons concerned. One of the basic difficulties with the Shelley ruling underscores the importance of relational justice: Racially restrictive covenants are voidable if, and only if, their enforcement is pursued through the courts. 198 See Shelley, 334 U.S. at 13 (concluding that “restrictive agreements standing alone cannot be regarded as a violation of any rights guaranteed to petitioners by the Fourteenth Amendment”). By implica­tion, then, these covenants are not illegal per se, and the same holds with respect to their private enforcement. This flaw is the product of a failure to appreciate the freestanding dimension of relational justice, which is the normative core of private law.

The scope and contents of the accommodative structure of the power and duty an owner bears in connection with her residential dwell­ing are partially set by reference to contextual considerations. 199 For more on contextual considerations, see supra section II.C.2. To begin with, the special standing private ownership accords to homeowners to make claims that would be otherwise illegitimate suspends many require­ments of relational justice outside the realm of selling and renting. Moreover, insofar as selling and renting are concerned, contextual considerations can also make a difference, as in the case when the leas­ing at hand entails the cohousing of the landlord and tenant, so that the internal logic of the practice of residential dwelling exempts owners from an accommodative duty. 200 The Fair Housing Act’s exceptions for single families and small, owner-occupied, multiple-unit dwellings seem to rely on this rationale but, arguably, overextend it. 42 U.S.C. § 3603(b)(1)–(2) (2012). There may be good policy reasons for these exceptions, but such exceptions are inconsistent with the demands of relational justice. In addition, contextual concerns can also shape the contents of the accommodation required for the terms of the interaction between the relevant participants to count as relationally just. Thus, the duty to accommodate need not affect the right of landlords to determine tenants’ maintenance obligations or similar leasing terms.

Regardless of what additional contextual refinements may be neces­sary, 201 One difficult issue, not addressed here, relates to the scope of accommodation that should be made in residential contexts that not only support owners’ quality of life but also serve as infrastructure for sustaining owners’ meaningful (e.g., religious or cul­tural) communities. Communities of this type require some demarcation from broader society and thus may require a measure of practical and symbolic exclusionism that goes be­yond the boundaries of the home. The duty to accommodate fair housing laws’ protected classes of people whose personal qualities are the outcome of chance cannot be qualified by such contextual concerns, yet intricate questions may arise regarding qualities that are subject to personal choice, such as religion and familial status. the terms of the interaction between owners and nonowners are not merely instrumental to realizing the public demands of justice in the residential dwellings context (and they certainly cannot be reduced to considerations of aggregate welfare). Requiring a private owner to set aside certain considerations, such as racist preferences, need not derive from a demand to support the state in its effort to fulfill its duty toward would-be victims of discrimination because it is fully grounded in private law’s commitment to relational justice. This commitment neither fulfills nor supplants the state’s obligation to curb discrimination in the housing market (including through the enlistment of the support of private own­ers to that end). Rather, it stands on its own, distinctive ground.

2. Workplace Accommodation. — A similar analysis can be applied to the context of workplace accommodation. Work, at least since the de­cline of feudalism, figures prominently in the mature lives of free and equal persons, as it generates both instrumental and noninstrumental value for our ability to do good by doing well in that practice. For many, work is the quintessential ground project. 202 See supra note 110 and accompanying text (discussing the concept of ground projects). Here, too, there is a strong sentiment against excluding would-be employees from the labor market due to personal qualities such as certain forms of disability, familial sta­tus, and religious affiliation. 203 See Elizabeth Anderson, How Should Egalitarians Cope with Market Risks?, 9 Theoretical Inquiries L. 239, 255 (2008) [hereinafter Anderson, How Should Egalitarians Cope] (noting ways in which unaccommodating work arrangements are objectionable from an egalitarian perspective); Daniel Markovits, Luck Egalitarianism and Political Solidarity, 9 Theoretical Inquiries L. 277, 298–301 (2008) (same); Sophia Moreau, What Is Discrimination?, 38 Phil. & Pub. Aff. 143, 143–48 (2010) (noting the effects and purposes of antidiscrimination laws); Seana Valentine Shiffrin, Egalitarianism, Choice-Sensitivity, and Accommodation, in Reason and Value: Themes from the Moral Philosophy of Joseph Raz 270, 297–98, 302 (R. Jay Wallace et al. eds., 2004) [hereinafter Shiffrin, Egalitarianism] (discussing how work accommodations facilitate certain types of freedom). Liberal egalitarians agree that the costs associated with such human qualities must not be borne exclusively (or even at all) by the would-be employees. Once again, the crucial question is who bears the responsibility to make the practice of working consistent with this truism; the answer to this question depends on the justification of this commonplace sentiment.

Under our account of private law, for the terms of the interaction between an employer and a would-be employee to count as relationally just, the responsibility in question must be borne, at least in part, by the employer. 204 Cf. Moreau, supra note 203, at 145–46 (acknowledging that the employer’s failure to accommodate is a personal wrong “akin to a tort”). It is not sufficiently clear, however, whether Sophia Moreau’s argument establishes the requisite connection between her proposed grounds of accommodation—the employee’s deliberative autonomy—and the necessity (in terms of justice) of imposing (at least part of) the duty to accommodate on the employer rather than merely on the state. Moreover, this responsibility should ground a negligence duty to exercise reasonable care in making relevant employment decisions, rather than merely a duty to refrain from making intentionally discriminatory decisions. 205 But cf. Jalal v. Columbia Univ., 4 F. Supp. 2d 224, 241 (S.D.N.Y. 1998) (“Title VII, however, provides no remedy for negligent discrimination . . . .”). Consider an employer who turns down a job candidate because the latter requests days off in accordance with her reli­gious calendar. By disregarding the candidate’s choice of religious prac­tice, the employer fails to respect her on her own terms as a free and equal person. A state effort to substitute the employer’s responsibility with workplace accommodation (whether directly or through subsidies) cannot rectify the employer’s failure. To the extent that the employer does not bear at least some of the costs of accommodation, there is no intelligible way to regard the employer as engaging in relationally just relationships.

The case of work-related accommodation demonstrates the possible tension between the demands of relational justice and some core distributive and democratic commitments. This tension is the product of the objective costs that work-related accommodation often entails; these costs do not turn on intolerant preferences on the part of employers, employees, customers, or even society at large. Employing a member of another religious faith, for example, could place substantial constraints (related to dietary observances, holy days, dress codes, etc.) on the employer in efficiently operating her business; similarly, constructing an accessible workplace may cost more than its inaccessible counterpart. In these and numerous other contexts—such as the case of a disabled person who cannot compete on equal terms with other candidates for a particular job—a duty to accommodate can impose nontrivial costs, including costs that, from the perspective of distributive justice, are soci­ety’s to bear. 206 Both Seana Shiffrin and Daniel Markovits seem to raise the case as a friendly amendment to the luck-egalitarian theory of equality by criticizing a principle of strict choice sensitivity. Neither, however, makes the connection between such a critique and the ideal of relationally just terms of interaction. In fact, their respective critiques implicate the state, rather than the employer, as responsible for accommodation; they rest their arguments on the employee’s freedom rather than on what it means for the candidate and the employer to relate as free and equal individuals. See Shiffrin, Egalitarianism, supra note 203, at 302. Markovits might accept this view, since contractual considerations, on his ac­count of the division of institutional labor, are modeled on formal equality. Markovits, Arm’s Length Relation, supra note 25, at 312, 317–18; see also Anderson, How Should Egalitarians Cope, supra note 203, at 255 (discussing collective solutions for eliminating various inegalitarian relations).

The tension between private and public responsibilities of accommo­dation may be even more acute. Integration through work is conducive, even if perhaps not essential, to the prospering of democracy. Work-related accommodations are, in some cases, the catalyst for the social integration of the disadvantaged. This tension also plays a crucial role in the social integration of members of heterodox religions and of national minorities. Although successful integration into society through work does not entail political integration, the democratic ideal of equal citizenship is hardly sustainable in its absence. 207 See generally Judith N. Shklar, American Citizenship: The Quest for Inclusion 79–88 (1991).

This means that considerations of both fair distribution and equal citizenship are at odds with the demands of relational justice. Specifying the metric by which the burden should be redistributed across members of society is a complex task, beyond the purposes of this Article. What is important here is that although employers’ accommodation costs are likely to be passed on to some extent to customers and workers, there is no reason to believe that the emerging distribution will mirror the distributive consequences of government-funded accommodation. 208 The most immediate reason is that government-funded accommodation can distribute the burden of workplace accommodation far more broadly to include those who stand outside the privity of the relevant employment (or consumption) relation. Furthermore, a government program can create this distributive effect far more systemati­cally than a private law duty of workplace accommodation owed by an employer to her employees.

This Article cannot decisively settle this clash between relational justice and external distributional and equality concerns. Yet our concep­tion of private law—that of just relationships—helps address this ques­tion. Under our account, the identity of the agent (i.e., private employers or the state) who bears responsibility for accommodating employees is not merely a question of institutional design. A society that fully collecti­vizes the recognition of the particular traits that constitute the person that an employee actually is (such as religious affiliation, familial status, and disability) fails to uphold the demands of relational justice. Even the most distributively and democratically just schemes of workplace accom­modation (let alone the most efficient ones) do not satisfy the demands of relational justice if private employers are not obligated to assume re­sponsibility for ensuring this state of affairs. Such schemes leave employ­ers with a too-shallow conception of their relationships with employees, as if their employees were merely abstract beings rather than the fully realized persons they actually are. 209 It is worth noting that this problematic collectivization is starkly different from the nuanced workers’ compensation schemes discussed earlier, which alter the balance between public and private responsibility but do not efface the latter. See supra text accompanying notes 174–181.

It is important to note, however, that the conflict between relational justice and the external requirements of fair distribution and equal citizenship is not zero sum. A duty of accommodation grounded in rela­tional justice is a range property. 210 See supra text accompanying note 127 (explaining this feature). Employers have no affirmative duty to add employees. Only when they decide to hire do they encounter a duty not to apply criteria irrelevant to hiring decisions. This limited scope of employers’ accommodative duty is not coincidental: Employers are enti­tled to retain their autonomy; they should not be converted into mere in­struments for respecting potential employees’ right to self-determination.

Furthermore, our account of relational justice does not require the full internalization of the costs of accommodation by the hiring em­ployer. To the contrary, employers should not be overwhelmingly subordinated to would-be employees to the point of self-effacement. This understanding is reflected in the various legal doctrines exempting employers from making accommodation arrangements that impose un­due hardship or providing tax incentives or other publicly funded benefits to ameliorate such hardship. 211 See, e.g., Americans with Disabilities Act of 1990, 42 U.S.C. § 12112(b)(5)(A) (2012). Another, perhaps even better, solution (available in other jurisdictions) is to use direct subsidies. See Eyal Zamir & Barak Medina, Law, Economics, and Morality 255 (2010). Indeed, a legal regime that pro­vides substantial coverage of employers’ costs is not inconsistent with the prescriptions of just relationships, provided that it does not unjustifiably dilute the demands of recognition that underlie the relational duty of accommodation. In other words, for government-provided carrots not to eliminate the employer’s obligation to respect the person the employee actually is, the employer must bear some non-nominal burden (monetary or otherwise). 212 This is especially apt in the case of the maximally egalitarian employer whose reluctance to accommodate arises not from subjective or irrational considerations but from purely economic ones.

C. Joint Projects and Other Collaborative Endeavors

The first two categories of cases reviewed above support our just-relationships conception of private law by illuminating both the failure of the traditional account to consider substantive autonomy and equality and the false promise of the critical account to discard entirely the public–private distinction. In this category of cases, the traditional conception of private law is not objectionable from an equality stand­point; formal equality is here actually a reasonable—at times even the best—proxy for substantive equality. Even so, the traditionalist concep­tion still fails in these cases because it is not sufficiently responsive to the liberal commitment to individual self-determination. In contrast, private law by and large takes seriously its unique role in facilitating self-determination given the fact of human interdependence. Private law’s commitment to freedom thus exceeds that dictated by its traditional portrayal.

To illustrate, consider cases in which the interests of a group of peo­ple are interlocked, such as when they share an interest in the same piece of property or are all subject to a common liability. Suppose one group member incurs some expense in this joint project—she pays for the re­pair of a damaged roof of a shared house—and thereby benefits the other members since it is impossible or infeasible to exclude them from this collective good. In some instances, the beneficiaries might actively indicate an unwillingness to pay for the benefit; in many others, the ex­pense-incurring member and the beneficiaries might never have communicated about the expense. If private law were to discount peo­ple’s self-determination and focus solely on upholding their independ­ence, it would be difficult to justify a requirement for beneficiaries to make restitution; in the typical case, the claimant can show neither harm inflicted by the defendant nor the defendant’s consent to the exchange. 213 See Nozick, supra note 14, at 95 (arguing that one cannot demand payment for voluntarily conferred benefits absent prior consent); see also Coleman, supra note 181, at 166–69 (describing the difficulties of linking consent and hypothetical ex ante contracting).

Fortunately, private law does not take this approach. 214 See, e.g., Restatement (Third) of Restitution and Unjust Enrichment §§ 23, 26 (Am. Law Inst. 2011) (requiring restitutionary payment to one who confers an economic benefit on another in performing a joint obligation or in holding jointly held property). When the parties’ interests are sufficiently interlocked to prevent the claimant from reasonably pursuing her self-interest without benefiting others, the law of restitution typically facilitates collective action by forcing the beneficiar­ies to pay their proportionate share of the collective good. This neutral­izes the potential free-riding that could undermine the jointly beneficial collective action and the parties’ self-determination. 215 The following discussion of the law of restitution draws in part from Hanoch Dagan, The Law and Ethics of Restitution 123–63 (2004) [hereinafter Dagan, Law and Ethics] (arguing that “restitution for self-interested conferrals of unsolicited benefits can help to overcome free-riding problems that may hinder jointly beneficial actions”).

The typical features of collective action problems, 216 See generally Russell Hardin, Collective Action 9–10 (1982) (noting individual self-interest motivates collective action); Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups 2, 8, 10–11, 21, 51, 60–61 (2d ed. 1971) (arguing that individuals will not act collectively due to disparate personal incentives); Michael Taylor, The Possibility of Cooperation 3 (1987) (describing collective action problems as resulting from rational egoism). which exemplify the significant impact of human interdependence on self-determination, illuminate the gap between the commitments to independence and self-determination. Although promoting the parties’ self-interests when these problems arise requires cooperation, the absence of legal intervention might hinder jointly beneficial action because the individual interest of each party might override their interest in the collective good. 217 This will be the expected outcome if no single member of the group is likely to derive sufficient personal benefit from the collective good to justify paying the entire cost of supplying it alone and no coalition of members can feasibly divide the costs among those members. See Olson, supra note 216, at 41 (describing how a single holdout can derail collective action). Propo­nents of the regulatory conception of private law, concerned that free-riding means inefficient underproduction of collective goods, support the solution offered by the law of restitution. 218 See Ariel Porat, Private Production of Public Goods: Liability for Unrequested Benefits, 108 Mich. L. Rev. 189, 206 (2009) (describing how expanding restitution duties would help solve free-riding problems even absent state intervention). But it is important to recognize that the liberal commitment to self-determination also entails restitution, if (and only if) restitution is fine tuned: that in a significant subset of collective action problem cases, people’s independence must recede for the law to properly ensure self-determination.

Indeed, when law’s nonintervention is likely to frustrate goals that require collective action, the commitment to autonomy could justify overriding restitution defendants’ explicit disinterest in participating in collective action and paying their share. For this to hold, however, two conditions must be met. 219 See Dagan, Law and Ethics, supra note 215, at 135. First, it must be objectively evident that the defendant’s proportionate benefit exceeds her proportionate share of the cost of providing the benefit and that the law’s intervention is neces­sary to facilitate the jointly beneficial collective action. Second, a defend­ant must be unable to point to any (credible) nonstrategic motive for not contributing to the collective good.

Together, these conditions ensure that restitution defendants are better off receiving and paying for the collective benefits than doing without them and, therefore, have no legitimate objection to the restitu­tionary obligation. The first condition refines the circumstances in which law’s nonintervention is likely to hinder goals requiring collective ac­tion—that is, cases in which individuals may refuse to pay their fair share based solely on the expectation that the efforts of others will yield the same good free of charge to them (or more cheaply). The second condition ensures that the divergence between the defendant’s explicit preference (not to participate in the collective action) and her presumable self-interest (to participate) is due to the payoff structure to which she and the other potential participants are subject and does not reflect her genuine subjective preferences.

The second condition, which echoes the doctrine of subjective devaluation, clarifies when our conception of private law departs from that of both the traditionalists and critics. Thus, on the one hand, private law libertarians “cannot fill the gap [of the defendant’s consent or wrongdoing] by deeming a benefit incontrovertible, because this simply bypasses what needs to be established: the defendant’s active involvement in the transaction.” 220 Brudner, supra note 11, at 250. This, for an independence-driven private law re­gime, is a strict prerequisite for liability. 221 See id. On the other hand, a utility-enhancing perspective is much more responsive to restitution claimants than its autonomy-enhancing counterpart is; restitution is denied only if the utility loss to the defendant, if forced to pay, exceeds the gain to the plaintiff from the collective action that restitution could facilitate. In contrast, the demands of autonomy under the liberal commitment to individual self-determination are more stringent, precluding restitution in cases of potential subjective devaluation even when there is relative certainty that the action is jointly beneficial overall. This normative diver­gence generates a doctrinal one. 222 These two alternatives were offered respectively by the two great authorities on the English law of restitution. Compare Peter Birks, An Introduction to the Law of Restitution 121–24 (rev. ed. 1989) [hereinafter Birks, Restitution] (arguing for restitution under specific factual circumstances), with Lord Goff of Chieveley & Gareth Jones, The Law of Restitution 25 (Gareth Jones ed., 6th ed. 2002) (arguing restitution may be re­quired for conferral of a benefit that could be realized but need not be).

Utility yields a restrictive interpretation of the subjective devaluation defense, which potential realizability in money can overcome because even if the beneficiary does not appreciate the conferred benefit, the market’s appreciation will ensure that restitution does not generate a utility loss. By contrast, autonomy is more demanding. Insist­ing on people’s right to order their own priorities means that a benefit’s value is deemed incontrovertible only if it has been actually converted into money or “it is inevitable that the defendant will [in fact] realize the benefit.” 223 Graham Virgo, The Principles of the Law of Restitution 81 (2d ed. 2006). Contemporary law largely takes the latter approach, reflected in the orthodox position that denies restitutionary liability for improve­ment of a defendant’s existing interest, as opposed to its preservation, which does yield liability (at least in some cases). 224 See Hanoch Dagan & Michael A. Heller, The Liberal Commons, 110 Yale L.J. 549, 611–13 (2001) [hereinafter Dagan & Heller, Liberal Commons]. Although the line between improvement and preservation is often blurry, a defendant’s objection to invest­ing in an improvement is more likely to express her genuine valuation rather than be a strategic holdout.




These restitutionary rules are only the tip of the iceberg. There are numerous other private law doctrines that serve individual self-deter­mination while going well beyond the strict injunctions of independence and thus the justified scope of private law under the traditional (liber­tarian) conception. Insights of lawyer-economists and critical scholars can explain the breadth of this category of doctrines. 225 For more detailed analyses, on which the next two paragraphs draw, see generally Hanoch Dagan, Inside Property, 63 U. Toronto L.J. 1, 3–10 (2013) (highlighting the im­portance of rules addressing the internal life of property); Dagan & Heller, Choice Theory, supra note 68 (manuscript at 107–25) (demonstrating the significance of the proactive legal support of many types of contractual interactions).

The economic analysis of private law forcefully demonstrates how many of our existing practices rely on legal devices for overcoming vari­ous types of transaction costs 226 Alongside these transaction costs, there are certain features of cooperative endeavors—most notably, affirmative asset partitioning—that are (almost literally) impossible to achieve without legal intervention. See Henry Hansmann & Reinier Kraakman, The Essential Role of Organizational Law, 110 Yale L.J. 387, 406 (2000). (information costs, bilateral monopolies, cognitive biases, and heightened risks of opportunistic behavior) that generate the participants’ vulnerabilities in most collaborative interper­sonal interactions. 227 See, e.g., Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L.J. 87, 93–95 (1989) (discussing information costs and opportunistic behavior); Russell B. Korobkin & Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 Calif. L. Rev. 1051, 1077–80, 1137 (2000) (discussing cognitive biases and bilateral monopolies). Merely enforcing the parties’ expressed intentions would not be sufficient to neutralize the inherent risks of such endeav­ors. If many of these endeavors are to become or remain viable options, the law must provide assurances to generate the trust so crucial for suc­cess. Even when parties follow their own social norms in their interaction, these background legal guarantees serve as a sort of safety net in the event of future conflict and thereby foster trust in the routine interactions. 228 See Dagan & Heller, Liberal Commons, supra note 224, at 578 (arguing that “background rules . . . can . . . creat[e] a formal ‘safety net’ that enables commoners, without taking prohibitive individual risks, to gain the benefits that flow from trusting one another”).

The law’s effects are not only material but also constitutive. Because private law tends to blend naturally into the fabric of our society, its categories are crucial in structuring our daily interactions. 229 See, e.g., Gordon, Unfreezing Legal Reality, supra note 49, at 212–14. Thus, many of our conventions—including social practices we take for granted (think bailment, suretyship, and fiduciary)—are, especially in modern times, legally constructed. 230 See Ian Ayres, Menus Matter, 73 U. Chi. L. Rev. 3, 5–8 (2006) (describing varying legal proposals to address employment discrimination based on sexual orientation). Even putting aside the transaction costs entailed in constructing these arrangements from scratch, were these conventions not to be legally coined, people would face “obstacles of the imagina­tion” that could preclude these practices. 231 Hanoch Dagan, Defending Legal Realism: A Response to Four Critics, 1 Critical Analysis L. 254, 266 (2014). Indeed, private law institu­tions play an important cultural role. Like other social conventions, they both consolidate people’s expectations and participate in constructing core categories of interpersonal relationships around their underlying normative ideals. 232 See Hanoch Dagan, Property: Values and Institutions 3–35, 77–84 (2011) [hereinafter Dagan, Values and Institutions] (illustrating the link between the structure of property rights and categories of human interaction); Michael Klausner, Corporations, Corporate Law, and Networks of Contracts, 81 Va. L. Rev. 757, 786–89 (1995) (discussing the accumulated outcome of the social learning effect and the network externalities phenomenon).

The material and constitutive functions of private law imply that con­tractual freedom, albeit significant, cannot do all the work there is to be done, and hence, for many cooperative types of interpersonal relation­ships, some measure of active legal facilitation is both desirable and necessary. Lack of legal support may sometimes undermine—perhaps even obliterate—these types of interactions and, in turn, people’s equal ability to pursue their conceptions of the good.

The unbridgeable gap between strict adherence to formal freedom and private law’s commitment to autonomy is rooted in people’s fallibil­ity—most notably their bounded rationality, cognitive failures, and the fact that they tend to prefer their self-interests over the interests of others. A theoretical account of private law could start from an ideal world in which no such imperfections exist. But at some point, these imperfections would have to be addressed, and a shift from an ideal to a nonideal theory of private law would be inevitable. 233 Cf. Rawls, Theory of Justice, supra note 127, at 245–46, 351–52 (distinguishing between an ideal and nonideal theory of justice). Indeed, it is hard to imagine how a purely ideal theory of private law could have practical relevance for doctrinal areas (such as those just discussed) in which human imperfections are not merely of peripheral concern but a systematic difficulty. Ignoring this difficulty would be self-defeating if a theory of law aims to provide guidance for, or justification of, the actual legal doctrines that govern the terms of interaction among private individuals.

To be sure, die-hard libertarians need not be alarmed by these propositions. As we saw in the restitution example, they can—and, to be normatively consistent, should—insist that defendants’ liability be limited only to what can be reliably founded on their actual consent. 234 Cf. Peter Benson, Gaps and Implication in Contract Law and Theory: An Alternative to the Default Rule Paradigm 1–15 (2014) (unpublished manuscript) (on file with the Columbia Law Review) (“[C]ontractually enforceable terms of performance are not limited to or exhausted by what the parties have expressly provided.”). That is, libertarians could insist that there should be no discrepancy between the ideal and nonideal theories of private law since both must strictly adhere to formal freedom and equality, irrespective of human imperfections. But this response does not work for division-of-labor liberals, who take seriously substantive equality and self-determination. For them, the route is unavailable at least insofar as they can now appreciate the significant horizontal implications of these values and therefore acknowledge that state-supplied public law cannot viably substitute for the relational obliga­tion of substantive equality and self-determination. 235 In theory, these division-of-labor liberals could alternatively argue that the private law doctrines that address strategic behavior in the context of joint projects and common endeavors embody a norm against abuse of rights, particularly the right to independence. While no one has yet developed an argument of this type, it would inevitably be suspected of stretching the anti-abuse-of-rights norm beyond its appropriate scope as a limiting principle. There may be diverging views on the scope and the details of the private law mecha­nism necessary for properly tackling this problem. 236 On one position on this issue, the commitment to individual self-determination requires that the state enable individuals to pursue their own conceptions of the good by proactively providing each major category of human activity—including commerce, work, residence, and intimacy—with a sufficiently diverse repertoire of different property institu­tions and contract types, each governed by distinct values or different balances of values. See Hanoch Dagan, Pluralism and Perfectionism in Private Law, 112 Colum. L. Rev. 1409, 1423 (2012); Dagan & Heller, Choice Theory, supra note 68 (manuscript at 138–54). But liberals cannot ignore the impact of private law on substantive freedom and equality; if the state takes a hands-off attitude, it will authorize a social structure guided by formal freedom, which undermines the commitment to self-determination. 237 Cf. Joseph Raz, The Morality of Freedom 162, 265 (1986) (arguing that the liberal state must provide “a multiplicity of valuable options” or else it “would undermine the chances of survival of many cherished aspects of our culture”).

D. Affirmative Interpersonal Duties

The private law doctrines that facilitate joint projects and coopera­tive endeavors subordinate people’s independence to their self-determination. The law is relatively confident that claims to independence in these cases are invoked only for strategic reasons and that the liability it imposes is in fact conducive to people’s self-interests. We turn now to the most contentious category of cases: when private law imposes on people affirmative duties in the service of the self-determination of others. Undoubtedly, such duties are flatly inconsistent with the proposi­tion that either formal freedom or formal equality is (or both are) the basic underlying value(s) of private law. 238 See Richard A. Epstein, A Theory of Strict Liability, 2 J. Legal Stud. 151, 198–99 (1973) (arguing against affirmative duties on grounds of “individual liberty”).

Indeed, from the traditional perspective, any legal duty to aid a se­verely distressed stranger necessarily subordinates the duty bearer to the stranger’s vulnerability and thereby denies the duty bearer both her inde­pendence and her formal equal standing vis-à-vis that stranger. Unsurprisingly, traditionalists often speak of “the rule against tort liability for failing to rescue” 239 Brudner, supra note 11, at 278. and regard it to be “an organizing normative idea in private law.” 240 Peter Benson, Misfeasance as an Organizing Normative Idea in Private Law, 60 U. Toronto L.J. 731, 733, 788 (2010) [hereinafter Benson, Misfeasance]. They claim that the distinction between misfeasance and nonfeasance is a conceptual feature of private law and a stable point for legal analysis, 241 Weinrib, Idea of Private Law, supra note 38, at 10. which is normatively crucial for justifying the division of labor between private and public law in a liberal state. 242 See Ripstein, Private Wrongs, supra note 24, at 53–80, 288–95; see also Ripstein, Division of Responsibility, supra note 23, at 1823–25 (discussing the misfeasance–nonfea­sance distinction as relevant to “determining the structure of the obligations between pri­vate persons”); Arthur Ripstein, Three Duties to Rescue: Moral, Civil, Criminal, 19 Law & Phil. 751, 764–65, 767–68 (2000) (illustrating that the division between private law and public law is partially informed by the conceptual differences between misfeasance and nonfeasance).

We deny neither the existence of a misfeasance–nonfeasance distinc­tion as a matter of positive law 243 See, e.g., Riedel v. ICI Ams. Inc., 968 A.2d 17, 22–23 (Del. 2009) (discussing the “early common law distinction between action and inaction”). nor its significance in determining the contents of interpersonal duties. But we do reject the attempt to read a foundational commitment to this distinction. The most plausible justification for the misfeasance–nonfeasance distinction is consistent with our account of private law as the legal ordering of relational justice among substantively free and equal persons.

Two uncontroversial observations undermine any attempt to situate the misfeasance–nonfeasance distinction at private law’s doctrinal core. The first is that reluctance to impose affirmative duties to aid others is not a unique feature of private law; indeed, this concern is not foreign to criminal law or, even more significantly, to constitutional law. 244 For criminal law, see Wayne R. LaFave & Austin Scott, Jr., Criminal Law 193 (1986) (discussing the traditional reluctance to impose criminal liability for omissions). See generally Graham Hughes, Criminal Omissions, 67 Yale L.J. 590 (1958) (discussing the history, conceptual basis, and current case law relevant to omissions in the criminal law context). For constitutional law, see DeShaney v. Winnebago County Department of Social Services, 489 U.S. 189, 195–96 (1989) (noting that the Due Process Clause is a “limitation on the State’s power to act” and not the imposition of an “affirmative obligation on the State”). See generally Frank I. Michelman, The Protective Function of the State in the United States and Europe: The Constitutional Question, in European and US Constitutionalism 156, 156–60 (Georg Nolte ed., 2005) (discussing the constitutional ques­tion of the scope of a government’s “protective” function and the different approaches in the United States and in Europe). The gen­eral reluctance to impose affirmative duties in the criminal law context, which does not feature a bipolar structure of duties and litigation, reveals that it is not the distinctively bipolar structure of private law that under­lies the restrained approach to affirmative duties. Similarly, the constitu­tional law of some jurisdictions (most notably, the United States) renders vivid the thought that it is not the horizontality of private law that guides the law to treat affirmative duties differently than negative limitations or prohibitions. For instance, some areas of U.S. constitutional law, such as the Due Process Clauses of the Fifth and Fourteenth Amendments, dis­play substantial hostility to affirmative duties. 245 DeShaney, 489 U.S. at 195–96. And unlike private law, these areas of constitutional law quintessentially express public law’s con­cern for vertical interactions.

The second observation is that the misfeasance–nonfeasance distinc­tion is not universally applied in all private law systems, nor is it categori­cally applied within common-law private law. Both of these observations undermine the significance of the misfeasance–nonfeasance distinction as a fundamental characteristic of private law. But while the first observa­tion shows that the placement of this distinction as a fundamental char­acteristic of private law is underinclusive, the second observation—to which we now turn—illustrates the opposite concern of overinclusiveness.

The private law of some jurisdictions across Europe and Latin America does not strictly adhere to this distinction, 246 See Jeroen Kortmann, Altruism in Private Law: Liability for Nonfeasance and Negotiorum Gestio 29–50 (2005) (discussing French and German approaches to nonfea­sance liability); Damien Schiff, Samaritans: Good, Bad, and Ugly: A Comparative Law Analysis, 11 Roger Williams U. L. Rev. 77, 79 (2005) (“[A]lmost every civil law jurisdiction in Europe, as well as in Latin America, recognizes various types of duties to rescue and related tort actions.”).
Recently, Ripstein has argued that these counterexamples can be explained away. Ripstein suggests that the French civil law (perhaps the most prominent jurisdiction to enforce affirmative tort duties) takes instances of violating the criminal code as sufficient evidence for the existence of fault, so that “the presence of the duty to rescue in French civil law is an instance of a more general addition of private liabilities based on duties that are not private duties.” Ripstein, Private Wrongs, supra note 24, at 60 & n.15. However, the French form of incorporation need not be importantly different from tort liability for the violation of some statutory provisions that are considered relational (typically through the doctrinal lens of the negligence per se rule). Indeed, a statutory provision that is best con­strued as designed for the protection of other persons places a requirement on parties to act in a certain way, the breach of which is “more than some evidence of negligence. It is negligence in itself.” Martin v. Herzog, 126 N.E. 814, 815 (N.Y. 1920). One variation of this rule is that the statutory provision can serve not merely as sufficient evidence for proving faulty conduct (which is the point of the negligence per se rule) but also as the source of the relational tort duty (owed to the plaintiff class). See Restatement (Third) of Torts: Liab. for Physical & Emotional Harm §§ 14, 38 (Am. Law Inst. 2005). Thus, there is noth­ing in the form of incorporating some “external” duties to the legal practice of tort law that renders the French civil law foreign or less “private.” The only question, then, concerns the content of the particular statutory provision on the basis of which violators are held liable in torts—whether or not it imposes a duty to rescue.
and there is no rea­son to believe that the private law, say, of France, is so essentially distinct from the common law just because it imposes affirmative duties. 247 Weinrib concedes that non–common-law legal systems are far less hostile to imposing a duty on nonfeasant individuals, but he explains this divergence as a matter of quantitative, rather than qualitative, difference. See Weinrib, Idea of Private Law, supra note 38, at 154 n.17. However, this maneuver is inconsistent with his overall argument against liability for nonfeasance. For one can claim that, certain exceptions notwithstand­ing, liability for nonfeasance is either incompatible with the bipolar structure of doing and suffering or compatible with this idea of private law as long as the law acknowledges, in some measure, the difference between liability for misfeasance and nonfeasance. See id. at 153–54 & n.17. More­over, even common-law private law imposes affirmative duties to aid strangers in some nontrivial cases. 248 See, for example, the celebrated case of Tarasoff v. Regents of the University of California, 551 P.2d 334, 340 (Cal. 1976) (finding a duty on the part of a therapist to use reasonable care to protect the intended victim of the therapist’s patient).

These observations of under- and overinclusiveness demonstrate that the rule against liability for nonfeasance cannot be taken as a “stable point” of private law. 249 Some traditionalists attempt to rescue the notion of the privileged status of the misfeasance–nonfeasance distinction in private law by reinterpreting it to reflect the maxim of “damnum absque injuria,” under which the defendant is not liable to a plaintiff unless the latter holds a right against the former. See Benson, Misfeasance, supra note 240, at 747 n.39 (“Historically, interference with the person or possessory and property rights of another was a paradigm instance of misfeasance.”); see also Ripstein, Private Wrongs, supra note 24, at 55–59 & n.6 (“You are entitled to constrain the conduct of others with respect to something only if you are entitled to determine the purposes for which it is used.”); Weinrib, Idea of Private Law, supra note 38, at 153 (“[T]he common law recog­nizes that for the injured person to recover, the suffering must be the consequence of what the defendant has done.”). The motivation for making this argument is to show that it applies far beyond rescue cases, including unrelated doctrines. See Ripstein, Private Wrongs, supra note 24, at 63–64 (providing examples of the absence of a duty to rescue in tort law); Benson, Misfeasance, supra note 240, at 737–43 (detailing cases of pure eco­nomic loss).
This account, however, is conclusory, in that it merely restates the crucial questions: What rights do we have and, ultimately, why do we have them? The problem with the claim that the misfeasance–nonfeasance distinction merely reflects an antecedent system of rights is that it implies that the distinction between the two depends entirely upon the applicable system of rights. The conclusory character of this account is on vivid display in two familiar tort cases. In the classic nineteenth-century English case Winterbottom v. Wright, the court defended the absence of a duty of care owed by the manufacturer of a defective product to its end-user by concluding that
“[t]his is one of those unfortunate cases in which there certainly has been damnum, but it is damnum absque injuria; it is, no doubt, a hardship upon the plaintiff to be without a remedy [for a negligently inflicted bodily injury], but by that consideration we ought not to be influenced.”
(1842) 152 Eng. Rep. 402, 405–06; 10 M. & W. 109 (Ex.). Some seventy years later, in MacPherson v. Buick Motor Co., then-Judge Benjamin Cardozo embraced the opposite posi­tion, concluding that, in the context of manufacturer’s liability for defective products, if the manufacturer “is negligent, where danger is to be foreseen, a liability will follow.” 111 N.E. 1050, 1053 (N.Y. 1916).
They also suggest that the conservative approach to affirmative interpersonal duties has its basis in general moral princi­ples rather than in considerations unique to private law. One justification for this approach is a concern with excessive interference with auton­omy. 250 See, e.g., Robert L. Hale, Prima Facie Torts, Combination, and Non-Feasance, 46 Colum. L. Rev. 196, 214 (1946) (criticizing “judicial reluctance to recognize affirmative duties” for its basis upon the assumption that “when a government requires a person to act, it is necessarily interfering more seriously with his liberty than when it places limits on his freedom to act”). This concern tracks the distinction between creating, having a contributory role in creating, and having no role in creating the risk of harm to the life and limb of another. Its focus is on the relationship be­tween a person’s agency and another person’s risk of being exposed (most paradigmatically, risk to her person). The category of misfeasance picks out core instantiations of this relationship, whereas the category of nonfeasance features its absence. 251 However, on this characterization, cases of pure economic loss or certain types of nuisances (such as interfering with the free flow of light onto the plaintiff’s land) do not fall in the category of nonfeasance. It is of course a separate question whether or not a no-duty rule should apply in these cases.

Indeed, the imposition of a duty (say, of care) has different norma­tive implications for the duty holder’s autonomy depending on the cate­gory, mis- or nonfeasance, under which a given case falls. It is one thing to place limits through a negative duty on a person’s course of action; it is quite another to dictate through an affirmative duty what this course of action should be. 252 But see Liam Murphy, Beneficence, Law, and Liberty: The Case of Required Rescue, 89 Geo. L.J. 605, 649 (2001) (arguing that cases involving misfeasance and nonfeasance, particularly those involving rescue, do not have significantly different normative implications). Put differently, it is one thing to require people to moderate their pursuits of ends when their ends put others at risk; it is quite another to compel them to make the vulnerability of others their mandatory ends. This understanding of the misfeasance–nonfeasance distinction does not imply that people should not bear duties to aid oth­ers. It does imply, however, that all else being equal, considerations of autonomy can be weightier when determining what should be people’s moral and legal responsibility to aid others. This constraint illuminates that affirmative interpersonal duties must take into serious account the self-determination of both parties to the interaction. In particular, it sin­gles out cases of easy rescue in which the responsibility placed on the duty bearer certainly infringes on her formal freedom but does not seri­ously jeopardize her security or other autonomy-supporting interests.

In principle, therefore, a private law committed to relational justice and, moreover, attuned to the fact of interdependence must make the requisite normative room for more affirmative interpersonal duties. 253 See Ernest J. Weinrib, The Case for a Duty to Rescue, 90 Yale L.J. 247, 251, 262, 293 (1980) (arguing that there should be an affirmative duty to easy rescue and that it would fit into the common law’s current understandings of liberty). It should be clear at this point that Weinrib has since retreated from this argument. See supra note 247 (discussing Weinrib’s updated, traditionalist position). The doctrines we consider below can further clarify this commitment. They each demonstrate that substantive, rather than merely formal, free­dom underlies private law’s existing affirmative interpersonal duties. 254 One doctrine—private necessity—even gives further support to formal equality’s role as an imperfect proxy for its substantive counterpart. Moreover, these doctrines manifest a (perhaps overly) cautious approach to the legitimate imposition of affirmative interpersonal duties. These pockets of liability for nonfeasance do not fully satisfy the demands of relational justice or exhaust the manifestations of the duty to aid in our private law (as the Tarasoff decision suggests). 255 Tarasoff v. Regents of the Univ. of Cal., 551 P.2d 334, 340 (Cal. 1976) (acknowledging a duty on the part of a therapist to use reasonable care to protect the intended victim of the former’s patient). There may, of course, be considerations that weigh against enforcing an otherwise legitimate pri­vate law duty to aid others. Imposing an obligation to aid may dilute the ethical value of altruism 256 See Douglas J. Den Uyl, The Right to Welfare and the Virtue of Charity, in Altruism 192, 192–93, 197, 205, 222–23 (Ellen Frankel Paul et al. eds., 1993). and, pragmatically, may also make it difficult to draw lines between easy and hard cases. 257 See Saul Levmore, Waiting for Rescue: An Essay on the Evolution and Incentive Structure of the Law of Affirmative Obligations, 72 Va. L. Rev. 879 passim (1986). It is beyond our current ar­gument to assess whether these considerations justify the limited role of affirmative interpersonal duties in contemporary private law. But some alarming evidence that the common law’s traditional reluctance to im­pose affirmative duties of easy rescue may be groundless 258 In the familiar case of Handiboe v. McCarthy, 151 S.E.2d 905, 907 (Ga. Ct. App. 1966), for example, the court found that a property owner owes no affirmative duty of easy rescue to save a drowning four-year-old licensee. The court invoked the “general rule,” according to which “the fact that a person sees another who is injured does not, in itself, impose on him any legal obligation to afford relief or assistance . . . .” Id. However, apply­ing the rule to this case seems a mere rationalization. In particular, the court relied on a rigidly formalist analysis of the situation, asserting that “[t]he mere fact that such child is an infant of tender years and unable to appreciate the danger of a particular situation [e.g., a swimming pool with a ‘slippery and slimy’ bottom on the defendant’s yard] as readily as would an adult does not alter the relation of the parties.” Id. at 906. suggests that private law must develop such a requirement in a more systematic fashion.

1. Mistaken Payment. — The case of mistaken payment is often de­scribed as the law of restitution’s “core case.” 259 Peter Birks, Unjust Enrichment 3 (2d ed. 2005). The basic rule governing such cases prescribes that, in principle, a recipient of a mistaken payment “is liable in restitution.” 260 Restatement (Third) of Restitution & Unjust Enrichment § 5 (Am. Law Inst. 2011). Absent negating considerations, such as reli­ance on the part of the recipient, restitution seems appropriate given that “the plaintiff’s judgment was vitiated in the matter of the transfer of wealth to the defendant.” 261 Birks, Restitution, supra note 222, at 147. This form of restitutionary liability is broadly accepted. 262 This acceptance is true at least in the “private” contexts of focus here—that is, con­texts in which neither the transferor nor the transferee is an institution. On the institutional context, see Dagan, Law and Ethics, supra note 215, at 60–63, 67–80. But as Brudner convincingly argues, the traditionalist at­tempts to account for this doctrine necessarily fall short.

If private law is to address only our independence and formal equal­ity, it must, by definition, be indifferent to whether the transferor’s mistake “thwarts the attainment of [her] intended goal” as long as the mistake “was not forced or manipulated by fraud.” 263 Brudner, supra note 11, at 242–43. Moreover, imposing liability in such cases offends formal equality because it enlists the recipi­ent, who is “a purely passive beneficiary,” for the task of remedying “the [transferor’s] unfortunate mistake”—“the consequences of her own freely willed activity”—for which she bears no responsibility. 264 Id. at 247, 253. For similar critiques, see generally Dennis Klimchuk, Unjust Enrichment and Corrective Justice, in Understanding Unjust Enrichment 111 (Jason W. Neyers et al. eds., 2004); Stephen A. Smith, Justifying the Law of Unjust Enrichment, 79 Tex. L. Rev. 2177 (2001). Indeed, given that in these cases, “the [transferor’s] possessory title is good against the [recipient],” Brudner concludes, the transferor’s demand of restitution is tantamount to unilaterally “subordinating the [recipient] to [the transferor’s] ends.” 265 Brudner, supra note 11, at 252.

This conclusion deals a strong blow to the traditional conception of private law. But it need not be a verdict against the law of mistaken pay­ments in itself, which is quite consistent with the commitment to individ­ual self-determination (and at the very least, is not inconsistent with the demands of substantive equality). Once we reject the strict binarism of the traditional conception of private law and accept that, in shaping the law of interpersonal relationships, we must sometimes make the type of unexciting but indispensable judgments to which Hart alluded, 266 See supra text accompanying note 143 (discussing the types of judgments en­tailed by a commitment to respect people’s autonomy). it be­comes clear that mistaken-payments law’s duty to aid others is unobjectionable.

For a private law that concerns itself with self-determination, “to be free is to act from purposes that are self-authored and to be able to view one’s life as broadly expressive of one’s projects and goals”; therefore, such a private law—the currently prevailing private law—pays heed to “the misalignment between the plaintiff’s reason for acting and the outcome she produced.” 267 Brudner, supra note 11, at 253–55. Furthermore, restitutionary liability is also conducive to self-determination because it expands people’s freedom of action by reducing the freezing and chilling effects of making mistakes under a no-liability regime. 268 See Dagan, Law and Ethics, supra note 215, at 43–44. Finally, the affirmative obligation it im­poses on the recipient is a modest one—a trivial burden that neither jeopardizes her self-determination nor seriously undermines her inde­pendence. 269 Id. at 43. Reciprocal respect to self-determination fully justifies this duty because it implies that the recipient should not be oblivious to the mistaken party’s circumstances.

Justifying the law of mistaken payments on self-determination grounds not only accounts for the presumptive rule of restitution but also clarifies the other details of this doctrine. Mistaken payments are typically not immediately and costlessly discovered: Recipients sometimes fail to notice the mistake and dispose of their income in the belief that the conferred payment is rightfully theirs. In such cases, the recipient’s autonomy is also at stake because requiring recipients to always be pre­pared to return any benefits they receive would severely hamper the secu­rity and stability of their affairs. Therefore, an autonomy-based law of mistakes must assign entitlements and liabilities through careful reconciliation of our liberty with security and stabil­ity, as exemplified by the familiar change-of-position defense. 270 See id. at 38–39, 45–52.

2. Private Necessity. — Consider the common law doctrine of private necessity and in particular, the entitlement of an individual in severe dis­tress to use another’s property to save her person or property. 271 Private necessity applies not only to cases in which the person of the defendant is at risk but also when only her property is at risk. See Vincent v. Lake Erie Transp. Co., 124 N.W. 221 (Minn. 1910) (regarding defendant at risk); Ploof v. Putnam, 71 A. 188 (Vt. 1908) (regarding property at risk). In nor­mal circumstances, the status of the interacting parties as formally free and equal justifies the requirement to secure the ex ante consent of the owner. 272 For defenders of the traditional conception of private law, this represents the source of the hostility toward the doctrine under discussion. See Ripstein, Force and Freedom, supra note 26, at 274–75, 277. Yet insisting on upholding formal equality between the parties in circumstances of an unexpected emergency amounts to empty formal­ism—it is implausible to disregard the disadvantaged position of the dis­tressed individual relative to this owner.

The doctrine of private necessity contends with this inequality in a way that goes beyond the familiar contract law doctrines of duress and unconscionability. It sets aside the basic requirement for the owner’s con­sent to the use of her property and permits a person in distress to make unilateral use of that property to save her own person or property; 273 See Hanoch Dagan, Unjust Enrichment: A Study of Private Law and Public Values 81–82 (1997). the law even holds a nonconsenting owner liable for interfering with such use of her property. 274 The leading authority on this point is Ploof, 71 A. at 189. But to ensure against excessive liability, the person in distress bears a duty to the property owner to compensate for any damage caused during property use. 275 See Francis H. Bohlen, Incomplete Privilege to Inflict Intentional Invasions of Interest of Property and Personality, 39 Harv. L. Rev. 307, 313 (1926). This latter rule corrects for the imbalance in autonomy that would arise were the nonconsenting owner left completely uncompensated for the unilateral use of her prop­erty. Again, the actual workings of the law manifest Hart’s observation on the need to distinguish “between the gravity of the different restrictions on different specific liberties and their importance for the conduct of a meaningful life.” 276 Hart, Between Utility and Rights, supra note 102, at 834–35; see also Glanville Williams, The Defence of Necessity, 6 Current Legal Probs. 216, 224 (1953) (“[T]he de­fence of necessity involves a choice of the lesser evil. It requires a judgment of value, an adjudication between competing ‘goods’ and a sacrifice of one to the other. The language of necessity disguises the selection of values that is really involved.”).

3. Responsibility of Property-Right Holders. — Our theory of just re­lationships can also illuminate another dimension of private law respon­sibilities: the burdens, rather than duties, borne by virtue of occupying the position of property-right holder (including, in particular, owner). The conventional wisdom, nicely captured by the maxim “a per­son acts at her own peril,” suggests that nonowners bear the entire risk of making mistakes with respect to property use. 277 See, e.g., Merrill, Property as Modularity, supra note 25, at 151, 157 (defending a minimalist private law regime). By implication, owners are said to assume no responsibility to guide nonowners in fulfilling their tort duty (such as the duty against committing trespass to land or chat­tels). 278 See, e.g., Stevens, supra note 174, at 205–06. Certainly, a commitment to formal freedom and equality renders this wisdom perfectly coherent, but our account rejects it, thus vindicat­ing these prevalent burdens.

As established above, a principled objection to owner responsibility is inconsistent with the accommodative structure of relationally just terms of interaction among substantively free and equal persons. 279 See supra section III.A (discussing accidental harm to life and limb). Moreover, property-right holders should not be exempted from making reasonable effort (such as giving reasonable, clear notice) to reduce some accidental mistakes made by nonowners with respect to the property in question. Happily, there is ample doctrinal evidence to this effect—for example, doctrines of consent, mistake, and proprietary estoppel as well as bur­dens arising from registration or recordation law. 280 See, e.g., Dagan, Values and Institutions, supra note 232, at 18–26; Dorfman & Jacob, supra note 76, at 75.


For more than a century, most approaches to the study of private law have been divided, broadly speaking, into two categories. On the one side are the traditionalists, who argue that private law expresses an apolitical idea of ordering horizontal interactions between formally free and equal persons. On the other side are critical thinkers and lawyer-economists, who take private law to be nothing more than an offshoot of public law that hides well its fundamentally regulatory orientation. The two opposing positions, which have developed respectively favorable and dismissive approaches to the idea of private law as a distinctively valuable institution, nevertheless share the view that private law treats its subjects as formally free and equal.

This Article challenges that shared understanding and lays the groundwork for a novel approach to private law. We have developed an account of the justice that can and should serve as the normative foun­dation to the law of horizontal interactions among private individuals in a liberal state. Rather than adhering to the unappealing ideal of formal freedom and equality, private law can—and to some extent already does—“move beyond formal freedom to real-world justice.” 281 3 Bruce Ackerman, We the People: The Civil Rights Revolution 211 (2014); see also id. at 154, 215.

In this respect, private law is indispensable. Only such a legal order can establish frameworks of interaction among free and equal individuals who respect each other for the persons they actually are. Indeed, it is one thing for the state to respect its constituents as genuinely free and equal persons; it is quite another to live in a society that expects individuals themselves to comply with the ideal of just relationships between free and equal agents. 282 In other words, the demands of relational justice support all pair-wise relations and all legitimate social structures (both statist and others). Accordingly, we have discussed the implications of this account of private law for understanding and assessing a variety of doctrinal areas. The theory of just relationships, developed here, clarifies core aspects of private law that traditionalists and critics (including lawyer-economists) fail to render intelligible. It also provides a critical frame­work upon which we can further incorporate the ideal of relational jus­tice into private law.