Compensation and Rights In the Liberal Conception of Justice
1991 Nomos XXXIII
(1991) Compensatory Justice
Randy E. Barnett
John Chapman, ed. (New York University Press)
I. The Compensatory Justice Thesis
II. The Incompatibility Thesis
III. The Limits of Legal Competence
IV. Conclusion
In the previous chapter on "The Limits of Compensatory Justice," Cass Sunstein offers two distinct theses. First, he defines a conception of'
compensatory justice that he claims lies at the root of Anglo-American legal systems. Second, he contends that in a number of doctrinal areas this model of compensatory justice is incompatible with the best theories that underlie
the claims being made by plaintiffs. I will call the first of these arguments the "compensatory justice thesis" and the second the "incompatibility thesis.
In part I of my commentary, I question the accuracy of the compensatory justice thesis and suggest that Sunstein's account neglects a crucial characteristic
of the liberal conception of justice that underlies Anglo-American legal systems: the dimension of entitlement. Even so, rejection of the compensatory justice thesis does not entail rejection of the incompatibility thesis. For,
when properly understood, the underlying entitlements component of' the liberal conception of justice may be nearly as incompatible with some of the legal theories favored by Sunstein and others. These theories also run afoul of
the liberal conception of the rule of' law. In part II, 1 explain how, by acknowledging this basic incompatibility, Sunstein has made an unintended but significant concession to the liberal conception of justice and the rule of law.
Finally, any analysis of limits, whether of compensatory justice or anything else, must be comparative. As Sunstein implicitly acknowledges in his
presentation of substitutes for compensatory justice, the limits of a theory support its abandonment only when the superiority of an alternative is established. In part III, I suggest that, for his argument to succeed, Sunstein
must come to grips with the limits of legal competence created by the pervasive problems of knowledge, interest, and power. I conclude by briefly summarizing the liberal conception of justice and the rule of law and how they
address these problems.
Let me begin by restating Sunstein's five basic principles of compensatory justice. First, the event that produced the injury is both discrete and unitary.
Second, the injury is sharply defined. Third, a clear causal connection exists between the defendant's conduct and the plaintiff's injury. Fourth, a bilateral relation obtains between an identifiable plaintiff and an identifiable
defendant. Fifth, the purpose of the remedy and of the legal system is to restore the plaintiff to the position that he would have occupied if the unlawful conduct had not occurred. Presumably, the fifth of' these principles gives
this model the name of compensatory justice. If we accept this as a working definition of compensatory justice, it presents at best only a partial view of
the conception of justice that informs Anglo-American legal systems. Of course, the fifth principle describing the goal of compensation captures what for a long time has been the presumptive remedy at common law for breaches of
certain obligations. But other types of remedies, such as specific performance in contract law and injunctions in tort, were traditionally available at equity and even at common law. These other remedies, although sometimes seen as
compensatory, are more accurately viewed as vindicating the rights of the aggrieved party. What is missing from Sunstein's account is the rights-based nature of the common law of tort, contract, and property.
The conception of justice that informs the Anglo-American legal system is liberal. Compensation for objective wrongs rather than for subjective
"injuries" is fundamental to the liberal conception of justice. The concept of injury is simply too undiscriminating to provide the kind of social-ordering principle that a liberal society requires. Subjective injury is
in the eye of the beholder. Any act can subjectively injure another provided only that the performance of the act is bothersome or offensive. For example, the fact that one watches certain movies or listens to certain music may be
genuinely offensive to others; consequently knowledge that such conduct is occurring may inflict a real though psychological injury. Therefore, if subjective injury is the criterion for legal relief, any act may potentially be the
subject of a legal sanction and there is little way of knowing before one acts whether one is acting in a purely noninjurious way. Although the existence of
a subjective injury may be a necessary condition of prohibiting an action, it cannot be sufficient. Because we all must engage in a variety of acts any of which could inflict a subjective injury upon another, we need some more
objective criterion to distinguish permissible from impermissible action, compensable from uncompensable injuries. That the in sustained by the plaintiff is "discrete and unitary" or "clearly defined" is not.
enough. More than a "clear causal connection" between the defendant's conduct and the harm to the plaintiff is needed to obtain legal relief. All of these characteristics are usually present with subjective injuries
caused by offensive conduct. In a liberal conception of justice, rights are the concept by which the wrongful nature of an action is identified.
That the defendant caused injury by committing a wrongful act is what justifies the imposition of legal force to obtain a remedy, and the nature of the plaintiff's rights define the wrong. Without identifying the rights of
the parties, one cannot distinguish a legally cognizable harm from the many genuine subjective harms for which no legal remedy is available. Consider a woman charged with shooting a man. Only a proper understanding of
her rights permits us to distinguish a noncompensable injury inflicted by the woman in self-defense during the course of his attempt to rape her from a compensable battery that she inflicted wrongfully upon him. Regardless of which
description of rights is the more accurate, the injury to the man is quite the same. In sum, the liberal conception of' justice underlying the
Anglo-American common law is rights-based not injury based and Sunstein's definition of compensatory justice misses this crucial entitlements dimension. There are, however, two distinct ways of viewing the rights or
entitlements that galvanize the common law. One could adhere to the view of Oliver Wendell Holmes arid say that rights of persons are nothing more than the legal remedies they can receive in court. According to tire Holmesian
conception of rights, the availability of a legal remedy entirely and precisely defines the rights of the parties; the concept of right or entitlement acids nothing to the analysis. The Holmesian identification of rights with
remedies is sometimes referred to as the "bad man" theory of law since it views law from the perspective of the person who cares nothing for tire law and is influenced only by the prospective legal
sanction: If you want to know the law arid nothing else, you must look at it as a bad rnan, who cares only for the material consequences which such knowledge enables hirn to predict, not as a good one, who finds his
reasons for conduct whether inside the law or outside it, in the vaguer sanctions of conscience.
In contrast with the Holmesian conception of rights as defined by the availability of legal relief, one can view rights as justifying the availability of legal relief. According to the
liberal conception of rights, a legal remedy is not identical to the right that justifies its imposition. Rather, a legal remedy vindicates an antecedent right of the plaintiff' that was violated by the defendant. In this way, the
"legal right" or remedy available in a court is distinguished from the underlying or "background right" that morally justifies the court's imposition of legal force oil a defendant.2
Although this distinction between legal and background rights is correctly associated with natural rights thinking, the terra natural rights evokes a controversy concerning the origin or source of background rights that is not germane here. For present purposes we need only note that, in contrast with the Holmesian conception of rights, the liberal conception acknowledges background rights, however grounded, that are antecedent to legal rights. In the liberal conception of justice, violating these background rights is unjust and justifies rectification.
Despite their substantial differences, both of these conceptions of' rights undercut Sunstein's thesis that the principles of justice informing
Anglo-American legal systems are exclusively compensatory. Given the Holmesian view that recognizes only legal rights and equates these rights with available remedies, because other types of remedies besides compensation were
always and are still regularly awarded, rights are not exclusively compensatory. And viewed from a liberal perspective that acknowledges both background and legal rights, the common law is not strictly compensatory, since remedies other than compensation may sometimes appropriately vindicate a right.
Moreover, if the conception of justice that informs Anglo American legal systems is liberal, then Sunstein's omission of tire dimension of entitlement is
misleading. It inaccurately suggests that any injury satisfying the first four principles he lists is sufficient to justify legal relief, whether compensatory or injunctive. In fact, the liberal conception of justice also requires
that a right be violated before the legal system may justly rectify even a sharply defined injury produced by a discrete and unitary event clearly caused by a defendant's conduct.
By severing the goal of rectifying injuries from the conceptual framework that distinguishes compensable from uncompensable injuries, Anglo-American legal
systems may be made' to appear unjust by their own standards for failing to rectify many genuine injuries that cannot plausibly be cast in terms of entitlements. Yet the entitlements dimension of the liberal conception of justice
functions precisely to limit legal relief to certain injuries and not others. In part III, I shall say something about why this distinction among kinds of injurious action is necessary.
Although the treatment of imposed risk is controversial among rights theorists, no one denies that self-defense may rightfully be employed when a person is
confronted with a credible threat of a rights violation. The liberal conception of justice does not require that one suffer a battery and then collect compensation afterwards; one can also defend oneself, by force if necessary,
when confronted with the imminent prospect or risk of a battery. The imminent threat of receiving a battery is itself defined as a rights violation and is called an "assault"
On the other hand, although the entitlements dimension of justice may deprive the incompatibility thesis of some examples, enough examples of incompatibility
remain to give it considerable descriptive merit. Sunstein's argument is that, in some important cases, compensatory justice simply misses the basic point of the plaintiff's claim and leads to a regime of legal rules that is
"positively perverse" from the perspective of the plaintiff's theory of relief. His incompatibility thesis could easily be recast to say that, in many of tire same cases, tine rhetoric of vindicating individual background
rights also misses the basic point of the plaintiff's claim. For example, consider the case of a plaintiff who has not been discriminated against by a
particular employer, but who is a member of a group who has suffered pervasive discrimination. Suppose this plaintiff sues a particular employer alleging that, because of the long history of discrimination, the group of which she
is a member is represented in certain occupations at a substantially lower rate than other groups that had not been similarly subordinated. Sunstein would correctly argue that to characterize this claim as alleging that the conduct
of' (.iris particular defendant violated the individual background rights of this particular plaintiff would surely be to distort the plaintiff's theory.
Moreover, recasting such claims in terms of a "right to be free from wrongful discrimination on the basis of race or gender" paves the way for the type of reverse discrimination claims that Sunstein and others lament.
This sort of rights analysis is simply too "color-blind" to address effectively the injury suffered by members of subordinated groups. Affirmative action plans that use racial quotas to end group subordination may
unavoidably entail that some white males are intentionally discriminated against on the basis of their race and gender. To protect these white males from wrongful discrimination would direly be incompatible with the
nonsubordination theory underlying the original complaint. My purpose is not to express an opinion on the merits of these controversial issues,
any more than it was Sunstein's purpose to do so. Rather, this is his example of a claim that is incompatible with principles of compensatory justice. My purpose is only to note a similar incompatibility between this claim and
entitlements, provided that Anglo-American legal systems implicitly accept the liberal, rather than a Holmesian, conception of justice.
Holmesian legal rights are defined entirely in terms of remedies. If' a "nonsubordination" legal remedy is deemed to be available, then this is the
plaintiff's right. Therefore, no incompatibility arises between Holrnesian legal rights and any judicially enforceable remedy favored by Sunstein. Nor does incompatibility with a Holmesian conception of rights arise when persons
are denied judicial review of the "risk management" schemes of administrative agencies. If no remedy is deemed to be available, under the Holmesian conception. There is simply no right to such judicial relief. In this
sense, Sunstein's theories are entirely compatible with a Holmesian conception of rights because the Holmesian conception is empty. To sustain
the incompatibility thesis, we must therefore assume that Anglo-American legal systems adopt a liberal conception of justice that assesses legal rights against a nonnative framework of individual background rights. Only if
justice requires the protection background rights that are antecedent to legislative or judicial action can there be a serious problem of incompatibility. The liberal conception of justice requires (flat the use of legal
force against a defendant be justifiable oil the grounds that the defendant has violated a particular background right of the plaintiff's. A claim by a plaintiff seeking an end to the subordination of a particular
group, irrespective of whether the defendant has acted wrongfully, cannot be translated into this sort of rights claim without losing something in the translation. In this sense, these two types of claims are genuinely incompatible.
If anything, Sunstein has understated the incompatibility of his legal theories with fundamental notions underlying Anglo American legal
systems. These theories are not only incompatible with a substantive view of justice based on background rights, they are also incompatible with the procedural protections associated with the liberal conception of
the rule of law. In addition, because coercively imposed remedies create serious risks of enforcement error and enforcement abuse, the liberal conception of the rule of law
also requires that sufficient evidence of liability must exist and be presented in a neutral forum before a remedy may be imposed. Civil defendants must be proved liable beyond a preponderance of the evidence; criminal defendants
must be proved guilty beyond a reasonable doubt. In this way, both the rate of enforcement errors and the potential for enforcement abuse will be reduced.
This requirement of proof will seriously inhibit the feasibility of risk management strategies. Requiring that sufficient evidence of risk imposition be
presented in a neutral forum before a legal remedy is imposed, means that hunches or guesses or speculalion that a defendant's activities may someday, in some way, cause harm to a plaintiff are insufficient to justify a finding
that the defendant's conduct is actually risky and subject to coercive regulation. Presumably, even -a regulatory regime of risk management would refrain from banning conduct that cannot be shown to create a risk of harm-provided,
of course, that this sort of a regime respected a liberal conception of the rule of law. The background rights of the parties that define the
liberal conception of justice, coupled with procedural protections associated with the rule of law, give rise to a heavy presumption in favor of liberty. Even when conduct is shown to create a serious risk of 'a rights violation,
we require those who advocate restricting such conduct to prove that such a restriction is really necessary. The importance of a presumption of liberty is well-accepted when other kinds of genuinely risky conduct not discussed by
Sunstein are considered. For example, even when speech creates a risk of serious emotional and even physical injury to others, a heavy burden falls oil
those advocating prior restraint to show that this sort of remedy is really warranted. And, despite the demonstrable risks that many accused criminals pose to others, we require a high standard of proof be met before imposing
punishment. Proposing "risk management" strategies to handle these kinds of risks of injury has a decidedly different ring from the sorts of risks Sunstein proposes to manage, but no difference in principle exists between
them. The very same dangers of managing risk so familiar to us in the areas of speech and crimes exist to an every greater degree with the other more amorphous and speculative kinds of risks he wants to address.
We may summarize the analysis to this point as follows: The compensatory justice thesis misses the dimension of entitlement that is a crucial part of the
conception of justice accepted by the Anglo-American systems of justice. To sustain the incompatibility thesis requires that this dimension of entitlement be liberal, rather than Holmesian. For, while a Holmesian conception of
rights is perfectly consistent with any of the theories discussed by Sunstein, a liberal conception of justice based on individual background rights is incompatible with at least some of these theories. Moreover, these theories are
also incompatible with important aspects of the liberal conception of the rule of law. Consequently, if Sunstein is correct to claim that the fundamental principles of Anglo-American legal systems are incompatible with the theories
he describes, then he must concede that Anglo-American legal systems implicitly accept a liberal conception of justice, a liberal conception of the rule of law, or both.
Of course, given that the reason for formulating the incompatibility thesis was to highlight the normative shortcomings of Anglo-American legal systems, this
may seem more like an in dictment of the liberal conception of justice than a "concession." Yet a concession it surely is. For decades we have been told by those who accept a Holmesian view of rights that because any
claim can be cast in terms of a legal right, the concept of rights in no way assists in deciding among conflicting claims. The incompatibility thesis undermines this argument. It concedes that, given the individual background
rights that inform the liberal conception of justice, at least some claims simply cannot be couched in terms of a rights violation without either absurd or even perverse doctrinal results. To implement legal theories such as risk
management and nonsubordination means abandoning important aspects of' the liberal conception of justice and the rule of law. The concession implicit in the
incompatibility thesis is important in at least two respects. First, the thesis concedes that these legal theories are indeed iconoclastic, insofar as they cannot comfortably be fit within long-standing conceptions of corrective
justice and long-accepted precepts of the rule of law that inform Anglo-American legal systems. Of course, to call a legal theory iconoclastic is in no way to refute it; nothing is intrinsically wrong with being an iconoclast.
Still, candor has its rhetor as proponents of these theories are wont to do, as mere modest and inevitable extensions of already accepted principles. Second, and more significantly, legal theories that are candidly incompatible
with an entitlements approach are vulnerable to a potent line of criticism: that the "limits" of the liberal conception of justice and the rule of law, not coincidentally, correspond to the limits of legal competence.
Fully to explain the
limits of legal competence requires a lengthy analysis of the liberal conception of justice and the rule of law and their relation to the pervasive social problems of knowledge, interest, and power. For obvious
reasons, I cannot present a comprehensive analysis here.6
Instead, to give some flavor of why legal competence is limited, I briefly summarize just a few of the many aspects of the social problems that the liberal conception of justice and the rule of law address. Although I shall not repeat this caveat, it is, in effect, for the balance of this commentary.
Some legal philosophers have argued that institutional features of' the judicial system make it an appropriate place to adjudicate claims of
right and generally an inappropriate place to settle polycentric disputes that do not involve rights violations. Nonsubordination and risk management theories involve controversies that cannot
be couched in terms of liberal rights, in part, because the problems they address involve complex interactions among and wide-ranging effects upon countless people. Likewise, the unforeseen consequences of a judicial decision in
applying such a theory to an actual controversy will swamp the immediately perceivable benefits to be gained by this decision. That the outcome of any adjudication of such controversies will be for the better is pure chance. Moreover, forcing parties to couch their legal claims as claims of right provides a conceptual connection between the legal rights recognized and
enforced by a legal system and the back ground rights specified by the liberal conception of .justice. This connection increases the correspondence between the outcomes reached by a legal system and the background rights
that define the liberal conception of justice, making for a more just legal system. It also enables the adjudication of legal rights claims constructively to contribute to the evolving understanding of background rights.
If judicial competence is truly limited in this way, to concede that a particular legal theory cannot be couched in terms of a claim of right is to concede
that such a claim does not belong in a court of law. To the extent that the incompatibility thesis is correct, then, the goals of nonsubordination or risk management may be inappropriate judicial concerns. In sum, the "limits
of compensatory justice" that Sunstein stresses may simply reflect the limits of judicial competence. The mere fact that the liberal conception of justice and the rule of law is limited in this way does not, therefore, support
his conclusion that it is normatively objectionable. But doesn't this limit of judicial competence bolster Sunstein's claim that courts should not interfere
when regulatory agencies such as the Environmental Protection Agency or the Occupational Safety and Health Administration adopt risk managctnetrt policies or when school boards adopt. plans to end the subordination of certain
groups? In some cases, the limits of judicial competence may indeed justify judicial deference to other institutions. Still, adoption of risk management or nonsubordination strategies by government agencies does not cause the
background rights of citizens to evaporate, any more than the adoption of censorship strategies to minimize the harmful risks of speech causes the First Amendment rights, of citizens to evaporate.
When the First Amendment is implicated, courts scrutinize legislation or regulations to ensure that the risks alleged by the regulators to be posed by the
activity sought to be restricted are serious enough to warrant legal restraint. The courts are not only the best places to adjudicate claims of' right, they are virtually the only place where these claims may be vindicated. The
liberal conception of justice and the rule of law create a presumption of liberty that places the burden on regulatory agencies of all varieties to justify in a neutral forum any restrictions they may seek to impose on the
otherwise rightful actions of critics. Courts are that neutral forum. To pursue the goals of risk management and nonsubordination without judicial oversight
requires that such efforts be confined to nongovernmental institutions which are not subject to the share degree of scrutiny as government agencies. Of course, the more that government becomes intertwined with other institutions
the more difficult it is to permit these institutions to pursue important social goals without judicial review of their actions. This argues for less rather than more legislative activism in these areas, permitting other powerful
social institutions to carry the ball. Incompetence to deal with the polycentric issues raised by Sunstein's theories is not limited to courts. Legislatures and administrative agencies may be incompetent as well. The limits of
their competence stein from the pervasive problems of knowledge, interest, and power. Every society faces a pervasive problem of knowledge that has two
dimensions. On the one hand, we must somehow let persons make use of the knowledge that they possess; on the other hand, persons who act on their own knowledge must somehow take into account the knowledge of others of which they
are and must be profoundly ignorant. The knowledge problem, then, involves both the radical dispersion of knowledge and the pervasiveness of ignorance. The problem of knowledge may be analogized to a "knowledge glass"
that is only partially full. Persons must be able to make use of the part that is full, while somehow taking into account the part that is empty.
We solve the knowledge problem in practice by respecting the particular background rights specified by the liberal conception of justice. The principle of several property solves the first dimension of the knowledge problem
by allocating jurisdiction over physical resources to the individuals and associations most likely to have access to the dispersed personal and local knowledge of how such resources may best be used. Similarly, the principle of
freedom to contract enables transfers of these property rights to reflect the personal and local knowledge of current right holders. The principle of
freedom from contract addresses the second dimension of the knowledge problem by preventing transfers of property rights without the consent of the right holder. In this way, anyone seeking to displace a current right holder must
offer a benefit sufficient to induce a transfer. This forces one who seeks to obtain a given right to take into account the current right holder's knowledge of flow this jurisdiction may best be exercised. In the aggregate,
consensual transfers of property rights give rise to an irreplaceable set of market prices for physical resources that consolidate in a usable form dispersed worldwide personal and local knowledge of how these resources may be
used. Requiring people to respect these market prices forces them to take into account indirectly the knowledge of others to which they could never have direct access.
Theories of risk management and nonsubordination that operate independent of these entitlements face all insuperable problem of knowledge. They require a
knowledge of (last, present, and future circumstances that may simply be unavailable to any person or group, including legislatures and administrative agencies. Both of these approaches require that we know more about past events
and causes, more about everyone's present knowledge and preferences, and more about the future consequences of. actions-including the consequences of any lawsuit-than we call possibly know.
Of equal importance, behavior that risks harm may also yield considerable benefits. Indeed, the potential for gain motivates almost all risky action. Few of
the benefits we currently enjoy and take for granted carne without considerable risks. Moreover, an unappreciated benefit of engaging in risky activity is discovering which courses of actions are beneficial and which are mistaken.
Often, until risky conduct is undertaken we cannot know whether it is a mistake or not; only after all action is performed will its mistaken character be discovered. Experiencing mistakes is crucial to the evolution of knowledge of
which actions to take and which to avoid. The very idea of risk management underestimates the value of mistakes in the discovery of knowledge. If central planners responsible for risk management could know in advance what they
would need to know to balance the cost of these mistakes against the benefits yielded by discovery, there would he nothing to discover. In a
utopia of perfect information where all of this knowledge is freely available, we probably would be wise to reconsider our traditional insistence on rectifying only proven rights violations. In our world, however, empowering
anyone-whether judges, legislators, or regulators-to use force to rectify conditions about which we are largely or wholly ignorant virtually ensures the violation of background rights that are needed to solve the pervasive problem
of knowledge. With this in mind, Sustein's call "to develop a careful system of priorities; to coordinate regulatory systems now diffused over many statutes and agencies; and to recognize the high potential costs of both action and inaction ill the face of inadequate information,"
Moreover, both risk management and nonsubordination policies create grave and possibly insurmountable problems of power. Using force to achieve compliance
with justice raises the severity of erroneous judgments, giving rise to a serious problem of enforcement error. The background rights of the liberal conception of justice limits the severity of these errors by
limiting the scope of enforcement. By extending the use of legal coercion beyond the protection of individual background rights, however, theories such as Sunstein's would expand the severity of enforcement error. The liberal
conception of the rule of law reduces the rate of enforcement error by allocating burdens of proof and by varying the degree of proof needed to satisfy these burdens. By reversing the burden of proof and by reducing the
degree of proof required, these theories risk a serious increase in the rate of enforcement errors. The problem of enforcement error applies to
enforcement decisions made in good faith. The problem of enforcement abuse, in contrast, arises when the power to regulate the actions ofothers by force is used, not to pursue justice, but to serve the interest of the
enforcers or others. James Madison called this "the problem of faction": By a faction understand a number of citizens, whether amounting to a majority or minority of the whole, who are united
and actuated by some common impulse of passion, or of interest,adverse to the rights of other citizens, or to the permanent and aggregate interests of the community.
That the pursuit of certain worthwhile ends, such as the goals of risk management and nonsubordination, requires that precepts of justice and the rule of law
be set aside creates enormous opportunities for enforcement abuse. We are all too familiar with regimes of' "public interest" being used as a veneer for oppression and injustice. The conceptual and institutional structure
that has evolved to cope with both the problems of enforcement error and abuse are the very conceptions of justice and the rule of law that are incompatible with the legal theories Sunstein favors. We abandon them at our peril.
Notice that the liberal conception
of justice addresses the problem of knowledge, not with the principle of compensation, but with the concepts of several property rights and freedom of contract. Well-defined rights of' several property and freedom of contract are
sufficient to solve the knowledge problem without any need for compensation. A principle of compensatory justice is needed only to address two serious problems of interest: the incentive problem and the compliance problem.
Generally, the "problem of interest" refers to a gap that can arise between the requirements of justice and the rule of law and people's perception
of their interests. Without a right to compensation for forced transfers of property, persons may lack sufficient incentive to use the liberty that the liberal conception of justice makes possible. Without this incentive the
resulting gap between rights and interest will undermine the ability of the strategy of decentralized jurisdiction to address the knowledge problem. And the compliance problem is the need to close the gap between the rights that
define the requirements of justice and the interest of persons who would violate these rights. Compensation that deprives rights violators of the gains derived front their wrongful conduct provides an incentive to comply with the
requirements of justice. By addressing these pervasive problems of interest, a right to compensation for injustice is an important aspect of the
liberal conception of justice, but the injustice that is to be rectified is defined by the rights of several property and freedom of contract that are needed to address the knowledge problem. Seeing flow these other background
rights perform a function distinct from that performed by the right to compensation makes it easier to appreciate why rights precede and justify remedies in the liberal conception of justice. Each of these rights reflects different
dimensions of justice addressing different social problems. Proponents of legal theories that are incompatible with the liberal conception of justice and
the rule of law cannot rest content with observing the limitations of compensatory justice or of the entire liberal conception of justice and the rule of law. They must confront seriously the pervasive problems of knowedge,
interest, and power that impose real limits on legal competence and that require for their solution the very conception of justice that conflicts with these legal theories. Before we may safely, jettison the liberal conception of
justice and the rule of law that addresses these problems, we must be shown persuasively that alternative conceptual and institutional arrangements are available that can transcend these limits.
Yet Sunstein is remarkably casual about urging the abandonment modification of these principles because they stand in the way of these legal theories
while leaving the "large task" of elaborating principles of' risk management and nonsubordination to the next generation. But surely this expression of preference is premature, if not supremely risky. Risk management in
the realm of political and legal theory requires that one first establishes the ability of a legal system enforcing these as yet unelaboraded principles of risk management and nonsubordination to deal with the pervasive
social problems that are currently being addressed by the conception of justice and the rule of law it would supplant. Only then would the observation that the conception of justice underlying Anglo-American legal systems is
limited be of normative significance. Wishing does not make it so.