Brandeis University Web Stite

Responsibility and Tort Law



Copyright (c) 2004 Fordham Law Review
Fordham Law Review

April, 2004

72 Fordham L. Rev. 1811

LENGTH: 17022 words

PANEL V: TORT: THE DIVISION OF RESPONSIBILITY AND THE LAW OF TORT

NAME: Arthur Ripstein*

BIO:
 
* Professor of Law and Philosophy, University of Toronto. I am grateful to Michael Blake, Ronald Dworkin, Gregory Keating, Stephen Perry, Ernest Weinrib, and Ben Zipursky for comments and discussion. Thanks also to Martin Hevia for research assistance.

SUMMARY:
... In A Theory of Justice, Rawls makes almost no mention of the issues of justice that animated philosophers in earlier centuries. ... My aim in this Article is to argue that implicit in Rawls's writing is a powerful and persuasive account of the normative significance of tort law and corrective justice. ... To paraphrase the point here in terms of Bernard Williams's familiar objection to utilitarianism, the problem with the contract argument is that the parties in the original position attach value only to states of affairs. ... ) Again, the Rawlsian answer is that as between private parties, the division of responsibility mandates that we treat each person as free to do as he or she sees fit with the means at his or her disposal, that is, to use both bodily powers and income and wealth in pursuit of his or her own conception of the good, in a way consistent with others having the same freedom to pursue their conception of the good. ...  

TEXT:
 [*1811] 

Introduction
 
In A Theory of Justice, Rawls makes almost no mention of the issues of justice that animated philosophers in earlier centuries. There is no discussion of justice between persons, issues that Aristotle sought to explain under the idea of "corrective justice." Nor is there discussion, except in passing, of punishment, another primary focus of the social contract approaches of Locke, Rousseau and Kant. 1 My aim in this Article is to argue that implicit in Rawls's writing is a powerful and persuasive account of the normative significance of tort law and corrective justice.

Tort law is initially puzzling from the point of view of distributive justice. It protects persons and property against injury and invasion without regard to the distributions it upholds. It focuses on misfortunes that one person brings on another, leaving equally devastating losses to lie where they fall. Why have an institution charged with figuring out whose problem it is when things go wrong - that answers its questions not by looking to some idea of distribution, but instead to norms of conduct governing relations between private parties, invoking duties and standards of care, questions of remoteness, proximity and causation? Much contemporary scholarship turns this puzzle into conflict: Either tort law must be made to shift losses on grounds that are at least indirectly sensitive to concerns about distributive justice, or, as some libertarians would have it, distributive justice must give way to the rights to person and property that tort law recognizes. 2 I argue that Rawls offers us the  [*1812]  basis of an account that enables us to understand the normative significance of ideas about private wrongdoing and, more importantly, to locate that significance in relation to the ideas of freedom and equality that more conspicuously animate A Theory of Justice. I argue that the key to understanding the relation between distributive and corrective justice can be found in an idea that Rawls introduces almost in passing, but which is of the first importance to his project as a whole. That is the idea of what he calls "the division of responsibility" 3 between society and the individual.

In focusing on this aspect of his thought, I make no direct use of the aspect of the theory of justice that has attracted the most attention among legal scholars, that is, the ideas of the "original position," "veil of ignorance," and the idea of the social contract that they articulate. I take Rawls at his word in his claim that these ideas must be understood primarily as expository devices, and that the contract is, in Rawls's words, "a device of representation." 4 It does a serious disservice to Rawls's many contributions to political philosophy to imagine that he means to be offering anything resembling an algorithm for determining how society's institutions should work. Instead, the contract argument is a way of articulating other arguments concerning the fundamental ideas of freedom and equality. The division of responsibility is essential to understanding the way the contract argument articulates those ideas.

In Social Unity and Primary Goods, Rawls introduces the idea of what he calls "the division of responsibility" between society and the individual. 5 Society as a whole has a responsibility - that is, an obligation - to see to it that citizens have adequate shares of primary goods, which they need in order to pursue and revise their own conceptions of the good life. Distributive shares must be not only adequate, but also fair. Fairness enters as the necessary condition for the other dimension of society's share of responsibility: society as a whole can discharge its responsibility in the sense of its obligation only if it has given people fair shares. If it has, then private citizens, rather than society as a whole, are responsible for how things turn out for them, for whether what Rawls calls their "life plans" succeed or fail. If society fails to discharge its responsibility, society might still say that a particular person's failure is his or her own problem, but it lacks the license to say that it is her responsibility. The other aspect of the division of responsibility is that each citizen has a special responsibility for how his or her own life goes. The responsibilities of private citizens are responsibilities in the same double sense in which  [*1813]  the responsibilities of society as a whole are double: Provided that society has given them fair shares, private citizens must make what they will of their own lives, and they alone are accountable for what they make of these lives.

Rawls presents the idea of the division of responsibility by focusing only on the relation between society and the individual. Since his concern is with constitutional essentials, this is an entirely appropriate focus for him to have. However, that focus may create the misleading impression that relationships between the individual and the state exhaust the normative space and significance of Rawlsian liberalism. I want to suggest that, in order for the idea of each person having a special responsibility for how his or her own life goes to have content, we need to understand it in terms of the responsibilities between private citizens. My concern with responsibility is, again, with the double sense of that term as it appears on both sides of the Rawlsian division: In order for each of us to be answerable for what we make of our own lives, we must also take responsibility - that is, accept obligations - for the ways that our actions change the life prospects of others. Conversely, we are responsible, that is, answerable, for the impact that our deeds have on others. It makes no sense to say that someone is responsible for what he or she makes of his or her own life if what becomes of that life depends in the wrong kinds of ways on the deeds of others, either because he or she is made to bear costs that properly lie with others, or because he or she is free of costs that are put onto others. In Political Liberalism, Rawls makes this point when he speaks of a "division of labor" between the principles governing the basic structure of society and those governing individual transactions. 6 Rules governing individual transactions see to it that "individuals and associations are then left free to advance their ends more effectively within the framework of the basic structure, secure in the knowledge that elsewhere in the social system the necessary corrections to preserve background justice are being made." 7 Rawls's example concerns the rules of contract, which govern voluntary transactions. I will argue that the same line of reasoning requires that we also conceive of tort law as governing transactions between private parties, albeit involuntary ones.

I will make these points in the context of a discussion of some broader issues in Rawlsian liberalism. In Justice as Fairness: A Restatement, Rawls defends an idea of society that he refers to as "property owning democracy." 8 This is an idea that Rawls introduces, without developing it in great detail. I will argue that it provides a fundamental part of the Rawlsian picture, indeed, that it is needed in order to complete that picture successfully. The topic of private  [*1814]  property in political philosophy has, together with the related topic of tort liability, fallen into the hands of those whom we might, following Kant, describe as dogmatists and skeptics. Libertarians are dogmatists about property - they take ownership rights to be basic and inviolable, and so view any redistribution as necessarily unjust. 9 They exclude distributive concerns by insisting that anything that cannot be deduced from some axiom of self-ownership is groundless. Contemporary egalitarians are skeptics, fearing, like Erasmus, that the only way they can justify their faith in redistribution is by calling any competing claims into question. Erasmus doubted science, insisting that it was only a useful way of making predictions to handle mundane matters, in the hope that he would thereby secure his faith against its encroachments; 10 contemporary egalitarians doubt the claims of property, seeing it instead as merely a useful way of creating incentives to be used by those charged with insuring distributive justice.

Rawls has been a primary target of both dogmatists and skeptics, attracting the ire of libertarians such as Nozick, who deny the state any powers to limit property. For reasons that will become clear in what follows, libertarians go wrong in supposing that if it is wrong for one person to do something to another, it must also be wrong for the state to do that thing.

Rawls has also been criticized by skeptics. G.A. Cohen and Liam Murphy have each criticized him for what they take to be his excessive focus on institutions. 11 Cohen and Murphy misunderstand Rawls's project, because they focus only on some of the institutions that are central to it. As a result, Murphy, in other places, refers to private property as a "myth" supposing that, on any defensible view of social justice, it must be nothing more than the plaything of public policies directed at achieving just outcomes. 12

My topic is not property as such, but tort law. Tort law protects property as well as persons, however, so a Rawlsian account of its normative structure must take account of property, and explain how it can merit protection as a matter of justice. Justice requires that private law - tort, contract, property and unjust enrichment - have a  [*1815]  certain kind of independence. It is only a certain kind of independence, since I do not mean to suggest that these private law regimes cannot, consistently with Rawlsian justice, be limited by the concerns of public justice. The aggregative effects of contractual transactions may lead to distributive injustice that needs to be addressed through public law. Nonetheless, particular transactions can be judged on their own terms, rather than being subordinated to distributive justice. 13 The same point applies to property, and to the involuntary transactions governed by tort law. They too must be understood in their own terms. A proper understanding of the place of private ordering, and so, the place of tort law, does not force us into the Charybdis of libertarianism. Private ordering regulates relations between private parties, and so need not preclude public limitation. But if we are to avoid the reef of libertarianism, we must also avoid being sucked into the idea that relations between private individuals must be subordinated to distributive concerns. Instead, the Rawlsian idea of a division of responsibility requires that there be separate institutions charged with the separate tasks required by that division. To overlook this division is to make what purports to be a liberal theory turn out to be illiberal in a way that exactly mirrors the way in which libertarians are rightfully charged with being illiberal. Libertarians allow no space for a public sphere because they deny that the state can have any powers against private citizens that one private citizen does not already have against another. 14 Egalitarian liberals make the converse mistake when they suppose that any norms of conduct governing private parties must be the instrumental delegation of the powers that the state claims over its citizens, 15 or when they suppose that self-seeking behavior must be inconsistent with the broader demands of justice. 16

Rawlsian liberalism provides an alternative to both of these extreme positions. Rather than trying to generate redistribution out of ideas of private liberty, or private liberty out of redistributive ideas, Rawls offers a synoptic vision able to accommodate both. These themes will remain in the background for most of my Article, but they will re-emerge at the end, once my discussion of tort law is filled out.

 [*1816] 

I. The Contract Argument and the Division of Responsibility
 
Rawls is probably best known for introducing the idea of a social contract into contemporary legal and political philosophy. The contract is supposed to provide a model for the institutional design of the basic structure of society, by showing that rational persons, concerned with advancing their own interests, but knowing neither what the particular content of those interests would be nor what powers they would have in order to achieve them, would agree to his two principles of justice - the principle of maximum equal liberty, and the "difference principle" - so as to best secure their chances of success in pursuit of their interests. Rawls is explicit that the contract is an expository device that generates the results that it does because of the constraints that are built into it. He is also explicit that the choice of constraints underlying the contract is itself to be justified not in terms of any sort of agreement, real or hypothetical, but rather in terms of a normative conception of the fundamental ideas of freedom and equality. In A Theory of Justice, Rawls also suggests that the contract argument might be extended so as to cover all of morality, leading to an idea that he characterizes as "rightness as fairness." 17 Many scholars have supposed that this suggestion is both the most important innovation of Rawlsian political philosophy, and also that it provides the appropriate tools for thinking about a wide variety of legal problems that Rawls himself never entertains, including not only the best understandings of the rules of contract law 18 and the Takings Clause of the Fifth Amendment, 19 but also my current topic, the problem of the appropriate tort regime. I want to suggest, however, that, taken on its own, the contract argument is poorly suited to understanding the doctrinal details of private law.

I make the argument of this part in two stages. The first stage is straightforward: as Rawls himself reminds us, the contract argument itself is merely an expository device, and the specification of the interests and concerns of the parties cannot be derived from it, but must instead be brought to it, and defended on independent grounds. What people would agree to must depend on what they know and what they care about. Rawls acknowledges this when he writes:


 
The hypothetical nature of the original position invites the question: why should we take any interest in it, moral or otherwise? Recall the answer: the conditions embodied in the description of this situation are ones that we do in fact accept. Or if we do not, then we can be persuaded to do so by philosophical considerations of  [*1817]  the sort occasionally introduced. Each aspect of the original position can be given a supporting explanation. 20
 
Those who disagree about doctrinal issues in tort may well find themselves also disagreeing about the appropriate way to pose the question of what parties would agree to particularly because, given the expository nature of the contract device, everyone will be able to see just where the argument is heading. Defenders of the negligence standard prominent in tort law will insist that their contractors attach weight to liberty as well as security, and couch their arguments in terms of risk; 21 defenders of strict or enterprise liability demand that their contractors focus on actual injury, rather than the risk of it, and decide only who should bear the costs. As a device of representation, the contract may perhaps make these disagreements perspicuous, but it cannot resolve them.

My second, more serious objection to this strategy is that, for reasons first pointed out in detail by Thomas Pogge, the contract argument by its very nature lacks the resources to draw certain distinctions that are fundamental to an institution such as the law of tort. 22 The law of tort always asks a highly structured question: Is this plaintiff entitled to recover from this defendant? If the plaintiff recovers, she recovers in her own right, for what Cardozo called "a wrong personal to her, and not as the vicarious beneficiary of a breach of duty to another." 23 That is, she does not recover because she is in need and her injurer was bad. Nor does she recover so as to advance some broader set of social purposes. 24 Like the person who seeks redress for breach of contract, the plaintiff in a tort action must establish the right kind of relationship between her and her injurer. I will argue that such relationships must be invisible from the standpoint through which the Rawlsian social contract is usually thought to proceed.

II. Expository Devices and What They Expose
 
In treating the contract as an expository device, Rawls seeks to underscore the way in which the argument is part of a more general strategy of justification. In A Theory of Justice, Rawls characterizes  [*1818]  the broader strategy as "reflective equilibrium," the reconciliation of our normative commitments to both general principles and particular considered judgments. 25 Where judgment and principle diverge, one must give way; through the reflective back and forth of mutual adjustment, we arrive at principles that are defensible both at the level of generality at which they are posed, and also at the more specific level of particularity that they demand when applied to particular examples. No principle is immune to revision on the grounds that it leads to unacceptable consequences in a particular case; and no particular considered judgment is sacrosanct if it is at odds with general principles with which we find ourselves no choice but to endorse. In light of this general strategy, it is no surprise that Rawls construes the parties to the original position as strongly averse to risk. Those who wish to use the contract argument to yield a result closer to utilitarianism might point out that people are, as a matter of fact, variable in their willingness to take on risks, and that, again, as a general matter, the "maximin" strategy 26 that Rawlsian choosers employ is irrational in most of the contexts of ordinary life. The strategy of reflective equilibrium provides a ready answer: The parties are not choosing anything ordinary, and so, their principles of choice should be guided by their highest-order interest in protecting their own ability to pursue and revise their own conception of a good. Again, other critics have suggested that parties in the original position would choose material equality rather than the difference principle, 27 or that they would choose to extend the first principle to include economic as well as political liberties. Again, Rawls has a ready answer in terms of reflective equilibrium: Parties in the original position, concerned with protecting their own ability to pursue their own conception of the good life, whatever that turns out to be, would want both as much freedom as possible - hence the first principle - and also as many means at their disposal as possible - hence the difference principle, which allows inequalities where they work to the benefit of all, that is, where they increase the means that each person has at his or her disposal.

I regard these Rawlsian rejoinders to the utilitarian, egalitarian, and libertarian objectors as convincing. I think that Rawls is right to say that the capacity to set and pursue one's own conception of the good is a more basic moral concern than the derivative concern with maximizing one's expected welfare (which yields utilitarianism), equalizing one's comparative share of resources (which yields egalitarianism), or minimizing the legal limits on one's range of  [*1819]  pursuits (which yields libertarianism). From Rawls's point of view, each of these interests must be regarded as derivative. But I am not surprised that his critics have found them unconvincing, because the thing that ultimately distinguishes Rawlsian liberalism from its opponents is the conception of a person's higher-order interests with which it works. The same line of thought has even greater force when Rawls puts it in the language of constructivism in Political Liberalism. Constructivism starts with a normative idea of persons as free and equal, and articulates its conception of justice in light of that idea, so that the two moral powers of political liberalism, the ability to form a conception of the good and a capacity for justice, reflect freedom and equality respectively. 28 Whatever its advantages, however, it is the point at which the debate with utilitarians, libertarians and their ilk must be joined, because it is the locus of real disagreement. The contract argument provides a device to use in representing those disagreements, but it is not their source. As a result, it is of no real help in settling that sort of dispute. Rawls provides a detailed discussion of the sort of reasoning from which parties in the original position would reach utilitarianism, because "contractarianism" is not an alternative to utilitarianism at that level. It is a way of thinking about competing views, and revealing the differences between them.

In the same way, if we hope to use a version of the Rawlsian contract argument to adjudicate the long-standing dispute (in American tort doctrine, at any rate) between negligence and strict liability, 29 it seems that we need to resolve the question of the relation between liberty and security that lies at the heart of that dispute itself.

The problem is deeper, however, because the rationale for the contract approach in the situation in which Rawls employs it - choosing the basic structure of society - depends upon the all-pervasive nature of the choice being made. Rawls offers arguments to explain both why the basic structure is especially significant, and also why it forms a self-contained topic for choice from behind the veil of ignorance. If we are dealing with a topic that is not self-contained in the requisite sense, parties behind the veil of ignorance might well ask why it is that they are being asked to make such a choice, or why they  [*1820]  are to be bound by the choice they may have made, given that there are other factors that might be equally significant to their liberty and security, factors which must be addressed independently if they are to decide what the answer will be.

The problem here is not just one about what interests parties behind the veil of ignorance would have, nor is it a problem about how they would weigh those interests against each other, whether, for example, they would attach greater weight to security than to lower prices for consumer products. If the debate between strict liability and negligence is a debate about what is fair and just, it must be about what is fair and just between the parties to the tort action, not about what is fair and just globally. It must therefore take place within the context of an idea of private persons having rights as against each other. For reasons I will now explain, that is a context that the contract itself cannot generate.

III. The Structure of Tort Law
 
The second problem for anyone who would approach the questions of tort doctrine through the lens of the contract argument runs deeper. The difficulty is that the tort doctrine is deontological in its structure. 30 One of its central concepts is the concept of duty; another is the parallel concept of proximity. However we might disagree about the precise contours of these concepts there is no room for controversy about their deontological structure: the law of tort draws a distinction between misfortunes based upon the way in which they come about. As Oliver Wendell Holmes Jr. reminded us over a century ago, the law of tort takes no interest in misfortune as such, nor even an interest in misfortune that is occasioned by the deeds of others, considered as such. 31 Neither the law of tort nor its liberal interpreters imposes or advocates a regime of absolute liability. 32 Instead, the law of tort focuses on the relationship between the parties: Plaintiff's complaint is not that she suffered, nor that defendant caused her suffering, nor even that her suffering was the result of wrongdoing on the part of the defendant, but rather that her suffering was wrongful because defendant was not supposed to cause her injury in that way. This deontological structure cuts across any disputes between strict liability and negligence. Friends of strict  [*1821]  liability may wish to gloss "that way" in terms of foreseeability, 33 or of the risks characteristic of a particular activity in which defendant was engaged; 34 defenders of negligence will talk about the particular risk imposed by defendant, and the steps that defendant should have taken to ameliorate them. 35 Either way, tort law focuses on what one person has done to another, on the relation between what defendant has done and plaintiff has suffered, rather than focusing on either what defendant has done, or what plaintiff has suffered, considered in isolation.

This deontological structure of tort doctrine poses an immediate difficulty for any explication or adjudication of it within the structure of a Rawlsian contract argument. The problem is that this structure will always be invisible from the point of view of the contract argument. Parties behind the veil of ignorance have an interest in being free of injury, especially bodily injury, and an interest in being free to engage in activities that have the potential to injure others. Parallel to these two interests, but distinct from them, is a pair of interests in receiving compensation if injured, and in spending as little as possible to compensate others. In addition to these four interests, parties will also have an interest in avoiding what economists call transaction costs - they will want to use no more resources than are necessary to achieve the desired levels of liberty, security, compensation, and economy. Indeed, as rational choosers concerned to protect their own means with which to pursue their own purposes, they must be prepared to forego liberty or security if the price is too high. It is not clear how the parties could adopt a maximin strategy with respect to any of these interests, because they are in tension with each other, and the best outcome with respect to one may be disastrous with respect to another. Considered in isolation, safety sounds more important than anything else. But rational beings concerned with protecting their ability to choose would not give it unlimited priority if doing so would narrow the range of their choices too radically.

Thomas Pogge underscores this point by describing the contractarian approach as "consequentialist," because the contract argument shares with familiar forms of consequentialism its exclusive focus on outcomes. 36 Indeed, as Pogge points out, in the one place that Rawls imagines the parties considering a case that Locke, Rousseau, or Kant might have taken up, that is, the topic of punishment, he presents them as reasoning in just this sort of way.  [*1822]  Rawls considers the possibility that, were the proliferation of firearms a sufficiently serious threat to personal security and social stability, citizens might adopt a regime of strict liability for mere possession of a firearm. Anyone caught in violation of the prohibition would be subject to criminal punishment, even though they presumably intended to do no harm, and perhaps cause no harm and, moreover, may not have even been aware of their possession of it. 37 A number of commentators, notably Thomas Pogge and George Fletcher, have pointed out that this is among the least satisfactory aspects of Rawls's argument. 38 And it is unsatisfactory for a completely straightforward reason: The parties are concerned only with outcomes, and, because they are concerned only with outcomes, must be prepared to trade liberty off against security in whatever way will best protect their interests. Rawls treats this example as one of extreme circumstances in which citizens choose "the lesser of two evils" by abandoning the idea that the punishment should be conditional on responsibility. 39 The difficulty comes with the claim that "the principle of responsibility ... is simply the consequence of regarding a legal system as an order of public rules addressed to rational persons in order to regulate their cooperation, and of giving the appropriate weight to liberty." 40 If the idea of responsibility is subordinated to the contract argument in this way, it is unsurprising that it should give way in particular circumstances. 41

The point of this example, and the problem that I want to use it to draw attention to, is not that it is a reductio ad absurdum of the contract approach that it yields strict criminal liability. Perhaps such liability is sometimes defensible. 42 The problem instead is that the conclusion, palatable or otherwise, seems to be reached in the wrong way. None of the concerns ordinarily thought to be proper to questions of just criminal punishment - particularly issues of responsibility and fault, but also the related issue of proportionality - is in any way visible in this account, because the contract argument is centrally concerned with outcomes. To paraphrase the point here in terms of Bernard Williams's familiar objection to utilitarianism, the problem with the contract argument is that the parties in the original  [*1823]  position attach value only to states of affairs. 43 We could solve this problem by stipulating that parties in the original position are not willing to have their freedom compromised by being held to account for things for which they are not responsible - a reasonable enough stipulation, actually - but the cost of doing so is parallel to the problem of attaching infinite disutility to certain types of wrongdoing on a utilitarian analysis: It deprives the contract argument of its apparent analytical advantage, and makes it less an expository device than a format through which conclusions that have been arrived at independently can be stated.

Parallel difficulties arise if the original position is deployed to deal with issues of tort. Parties must be concerned to best protect their interest in being able to pursue (or pursue and revise) their own conceptions of the good life. From the point of view of that concern, however, the distinction between harms that I suffer in general and those harms that are brought about through the wrongdoing of others is invisible. If I am concerned with protecting my ability to pursue my own conception of the good, then I am just as concerned about the ravages of disease and the effects of my own poor judgment as I am with the effects of wrongdoing by others. I must be indifferent to the same amount of money whether I need to spend it on first party insurance or on the implicit "insurance" that a regime of enterprise liability adds to the prices of consumer goods I might purchase. If I suffer an injury, such as a gash in my leg, from the point of view of my ability to pursue and revise my own conception of the good, it does not matter whether it was broken through your carelessness, through the risks characteristic of an activity in which you were engaged, or through my own clumsiness, or by an act of God. Any way you cut it, the interference is the same, and, it would seem that I would be just as concerned about being indemnified against my loss, regardless of how it comes about. The contract argument also has room for the opposite concern: I presumably have some interest in avoiding liability to others, since any liability payments I might make deprive me of means with which I might have been pursuing my own conception of the good. This countervailing interest cannot be made to do very much work, if only because it seems that, if we focus only on the domain of accidental injury, the two competing interests are fully commensurable: I would want whatever liability regime would lead to the lowest overall level of cost. On a Rawlsian view, not all costs must be commensurable: Parties in the original position would perhaps attach greater priority to safety than convenience.

Yet any priority that is attached to safety is at odds with the familiar legal distinction between nonfeasance and misfeasance. If parties in the  [*1824]  original position give much higher priority to safety than to the inconveniences imposed by a concern for the safety of others, it is difficult to see a basis on which they could attach any weight to the difference between the failure to take precautions to prevent injury to others and the failure to take steps of an equal magnitude to aid those who are in distress. Morally speaking, there may be no basis for distinguishing easy rescues from the taking of minor precautions - so much is at stake on one side of the balance, so little on the other. 44 The original position captures this moral idea quite nicely, but it does so in a way that makes its concerns very different from those of tort law, which denies that there is a civil duty to rescue.

So much the worse for tort law, some might say - perhaps the correct conclusion is that there should be a tort duty to rescue. The problem is much more pervasive, however. If appeal to a contract is supposed to make any contact with tort doctrine, or with ordinary understandings of its scope, it must have something to say about accidental damage to property. Parties concerned with security would also presumably be concerned with protecting whatever property they had against accidental injury. Having one's home destroyed through another's action is much less serious than suffering serious injury or death. Nonetheless, it is still very serious. The contract argument would seem to require a duty to rescue property as well as persons when the cost of doing so is small enough in comparison to the expected loss.

The inability of the contract argument to exclude duties to aid is only a single example of a much more general problem. The distinction between nonfeasance and misfeasance is more general and pervasive; if I wrongfully injure somebody else and your interests are set back as a result - for example, I carelessly destroy the bridge that connects your island to the mainland, but you do not own that bridge - you don't recover because I owe you no duty, even though my wrong caused your injury. 45 This is the case even if my wrongdoing means that your medicine doesn't arrive, and your injury is serious. If parties in the original position attach greater weight to security than to the payment of damages, and so would depart from a strict cost-benefit analysis, then it is difficult to see why no liability would lie in this case. If the concern that injuries, or at least personal injuries, must be compensated animates reasoning in the original position, it will sweep such cases into its net. I do not mean to deny  [*1825]  that there could be "extrinsic" reasons, sometimes called reasons of policy, to limit liability in such cases. My point is only that if they are extrinsic, they are also extrinsic to the original position.

In talking about the doctrinal structure of tort law, I do not mean to claim that its doctrinal features are beyond the reach of criticism from the standpoint of justice. My point is rather that, engaging in the full exercise of "reflective equilibrium," the direct appeal to what people would agree to, concerned to advance their interests in both liberty and security, seems to lead to the wrong normative results. Whatever we think about the proper application of the distinction between wronging a person and failing to aid that person, some such distinction must at least show up in any adequate account of the enforceable obligations that private citizens owe to each other. So too must the distinction between wronging one person and wronging another. Consequentialists are famous in their resistance to such distinctions; the consequentialist is, after all, the one who has a vendetta against common sense. 46 I take it that one of the hallmarks of Rawlsian liberalism in general, and the appeal to both Constructivism and the idea of reflective equilibrium, is that Rawls shares with Kant a certain kind of philosophical humility. In rejecting the approach to justification that he characterizes as "Cartesian," Rawls declines to cast doubt on the most familiar distinctions of moral life. As Kant says in The Critique of Practical Reason, he would not take it upon himself to discover a new moral principle. 47

The distinction between nonfeasance and misfeasance is important for another reason as well. Unless we have some way of determining the structure of the obligations between private persons, any attempt to fix the burdens each of us can be made to bear on behalf of others will collapse into cost-benefit analysis, even if it is modified to attach extra weight to bodily safety. The debate between negligence and strict liability can only be joined if we have the distinction between nonfeasance and misfeasance in hand. So we need some account of what it is to owe something to each particular person in order to figure out what we owe to each other.

I will illustrate this point by considering Gregory Keating's influential appropriation of Rawlsian social contract theory in his argument in favor of enterprise liability. 48 Keating situates himself squarely in the Rawlsian tradition when he writes:


 
The particular justice conception that I am concerned to explicate has its roots in the social contract tradition; it therefore adopts this internal perspective in an explicit and self-conscious way. It asks:  [*1826]  "What terms would free and equal persons, concerned to cooperate fairly with each other, agree upon to govern the risks of accidental injury created by beneficial activities?" This is the basic question of social contract theory, applied to the tort law of accidents, and it is a question of justice. 49
 
Although he does not explicitly employ the device of an original position, Keating says that issues of justice are to be understood in Rawlsian contractual terms. 50 He argues that if we conceive of society as made up of free and equal moral persons, then we must also suppose that those persons would attach special priority to their own safety, so as to protect their ability to pursue their own conception of the good over a complete life. Keating argues forcefully that if persons who conceived of themselves in this way were given the choice between a fault-based regime of negligence for tort law, and a regime of strict, or enterprise liability, they would opt for the latter, as it would provide them with greater protection in the event of injury. Keating notes that parties in the original position are, on Rawls's account, aware of "general facts about society," and, points out that in the modern world, one of the most familiar and least disputable facts about society is that when activities, even ones with very small risks, are carried out on a large scale, it is inevitable that some accidents will occur. 51 Given this information, Keating shows that parties concerned to protect their security would opt for a system that protected them against such misfortunes. 52 Elsewhere, Keating notes the analogy between such reasoning and the reasoning that leads Rawlsian parties to the difference principle: Those who benefit least from such dangerous activities as the transport of gasoline - that is, the small handful of people who are inevitably injured as a result of the ineliminable risks attending that activity - will be prepared to accept the balance of the benefits of modern transport made possible by gasoline, against the costs, provided that they themselves do not end up bearing all of the costs. 53 Instead, the worst off do as well as they can, consistent with the interest everyone has in gasoline, if they receive compensation should they be injured. 54

 [*1827]  Keating puts his argument in distinctively Rawlsian terms, speaking of interests in security, of interests each of us have over a complete life, and also of the inherently conservative nature of such choices. Others, less sympathetic to Rawls, have made strikingly similar arguments. Louis Kaplow and Stephen Shavell, in their book Fairness Versus Welfare, argue for welfare-maximizing rules in tort and contract, by considering what parties concerned to advance their welfare would choose ex ante. 55 In another recent book, Charles Fried and David Rosenberg apply similar reasoning to the selection of tort rules. 56 These scholars treat the choice of tort regime as a special case of what Guido Calabresi once called "the decision for accidents," the decision about how much we, either as a society or as individuals designing the society in which we will live, are willing to pay for safety. 57 At some point, the priority of safety must give way to the costs of everything else. Ex ante choice is supposed to enable us to identify that point.

Keating differs from both of these approaches in important ways. He focuses on security over a complete life, because he supposes that the parties are concerned to protect their ability to pursue their own purposes, rather than their welfare. As a result, the prospect of a shorter, but happier life cannot be offered to them. He also differs from them in his explicit recognition that questions about what people would agree to are always questions of fairness, 58 and in his presentation of the idea of agreement in terms of deliberation. 59 Keating's choosers ask, "Which system of liability will best protect me, consistent with my interest in freedom over a complete life?" while Kaplow, Shavell, Fried and Rosenberg's choosers ask, "Which system will benefit me most?"

Despite these differences, Keating's approach shares the difficulties of the other accounts. It begins with the idea that the something like ex ante choice behind the veil of ignorance is the appropriate way of determining which regime we should have for dealing with accidental injury. Any such approach needs to explain why parties concerned to protect their freedom would take the topic of accidental injury as the  [*1828]  subject matter of agreement. Someone seeking to explain tort doctrine might take accidents as their starting point (though to do so would be to overlook torts ranging from trespass to misappropriation). 60 No such strategy is available for those who appeal to ideas of agreement to justify a choice of tort rules. The distributive idea that motivates the presentation of the alternatives as competing regimes of liability lacks the resources to restrict its operation to circumstances in which one party injures another. Parties behind the veil of ignorance or any analogue of it may well be concerned to protect their security. It is not at all clear why they would be more concerned to protect their security against accidental injury occasioned by the activities of others, rather than against accidental injury more generally, including injuries that they bring upon themselves. Why would they not choose a scheme of comprehensive social insurance, or at least a scheme of social insurance that protected them against disasters, whatever their source? The only way to distinguish between those misfortunes due to human choices and those due to nature is to start with an idea of responsibility, because the social contract cannot generate such an idea. Parties "choosing" liability rules must strike some balance between their interest in receiving compensation if injured and their interest in avoiding excessive premiums incorporated into consumer prices. The same reasoning applies whether the injuries are natural or the result of human agency, and whether the cost of providing compensation takes the form of taxes or of higher prices for consumer goods.

These points are not new. I have only applied familiar objections to the context of tort law. As I mentioned, Thomas Pogge has raised the structural point before, characterizing the indiscriminate use of the social contract device as indistinguishable from consequentialism. That is perhaps why it has such surprising and unpalatable results in the example of punishment. Unlike Pogge, I do not think this undermines Rawls's own use of the contract, because Rawls uses it to look to the implications of various competing ways in which a state could think about its citizens. Rawls shows the implications for constitutional law of thinking of citizens as free and equal, and as responsible for their choice of ends. Moreover, with the exception of the unfortunate discussion of punishment, Rawls makes it clear that the primary goods sought by parties in the original position are essential to their ability to pursue and revise their conceptions of the  [*1829]  good, thus making his focus on the basic structure principled. Nor do I accept Pogge's claim that his arguments cast doubt on the idea that the focus of justice should be on institutions. I believe that Rawls is right to say that institutions are the starting point for thinking about justice. The process of identifying the demands of just institutions must avoid the inevitable difficulties of cost-benefit analysis, and begin instead, as Rawls himself does, with the idea that individuals are responsible for their choices.

Indeed, with the exception of the brief discussion of gun control, Rawls explicitly tells the reader that his account is limited to what he calls "constitutional essentials." 61 It is unsurprising that he limits it in this way, especially in light of his appeal to Kant, Locke, and Rousseau. Rousseau raises special problems of interpretation, but both Kant and Locke are clear that the contractarian aspect of their argument enters only after the basic structure of rights between private parties is already in place. 62 In the same way, the ideas of agency and responsibility that animate legal and philosophical thinking about tort are ideas that we must bring to the contract argument, because they rest on considerations that cannot be derived from it. That is why Rawls is explicit that the principles chosen in the original position are not meant to apply to individual transactions, 63 including involuntary ones in which one person injures another. Individual transactions must be governed by principles of individual responsibility, not by distributive principles.

As Rawls puts it in Social Unity and Primary Goods, the contract argument begins with a conception of the person. Part of that conception, I will now argue, is a conception of a certain kind of independence and the conception of interpersonal responsibility that follows from it. That conception supposes that people are not willing to have their freedom compromised by being held to account for things for which they are not responsible, and the converse idea that they are also not willing to have it compromised by being left to bear burdens for which others are responsible. As I will now show, it must occupy that place through an argument that is more basic than the idea of the social contract.

IV. The Division of Responsibility
 
In Social Unity and Primary Goods, Rawls introduces the idea of a "division of responsibility" between society and the individual. Rawls writes:


 
Society, the citizens as a collective body, accepts responsibility for maintaining the equal basic liberties and fair equality  [*1830]  of opportunity ... while citizens (as individuals) ... accept the responsibility for revising and adjusting their ends and aspirations in view of the all-purpose means they can expect, given their present and foreseeable situation. This division of responsibility relies on the capacity of persons to assume responsibility for their ends and to moderate the claims they make on their social institutions in accordance with the use of primary goods. Citizens' claims to liberties, opportunities, and all-purpose means are made secure from the unreasonable demands of others. 64
 
Rawls formulates this idea in terms of the need for individuals to moderate the claims they make on social institutions: having received my fair share, it is up to me to make what I will of it, and to make what I will of my life, using it as my fair share. This is certainly sound advice, since it reminds us that, on the Rawlsian understanding of equality and social justice, the concern with distributive shares is meant to be an expression of an underlying idea of equal freedom, and that latter idea attaches no independent value to the relative sizes of different people's holdings. Each of us must moderate the claims we make on social institutions, because the shares that we have are ours to do with as we please; it would also be contrary to the freedom-based rationale for redistribution if people were then denied the freedom to do as they please with their shares. It would be contrary to the idea of equal freedom if a person were entitled to use his or her share and then demand another. Each of us has a special responsibility to make what we will of our own lives inasmuch as it is up to us to use the shares we would receive as we see fit. It is not up to any one of us to demand more than our fair share if we come to regret a choice we have made, nor, conversely, is it up to society to determine for us what conception of the good our lives should be organized around. People have their own lives to live, which means that society has no responsibility for how those lives go, provided that each has sufficient opportunities and resources.

I want to suggest that this idea of the division of responsibility has a further dimension to it as well. Rawls draws a distinction between society and the individual. We might rephrase this distinction as a distinction between the public and private, noting that some such distinction is obviously constitutive of liberalism, but at the same time, allowing for the fact that this precise way in which the distinction is drawn will be a matter about which reasonable people might disagree without thereby abandoning their claim to be liberals.

If we think of the realm in which each person is responsible for what he or she makes of his or her own life, there is some temptation to suppose that our special responsibility is a responsibility that each of us has, as against the state, which has the other responsibility for  [*1831]  seeing to it that we have adequate resources and liberties to do so. To view the division of responsibility in this way is to model the special responsibility that each person has for his or her own life on the sort of special relation to God and one's own salvation that figures in some Protestant understandings of religion. This way of framing the special responsibility treats it as a generalization of the idea of freedom of religion. Something like this idea arguably underlies John Locke's argument in A Letter Concerning Toleration 65 in which he argues that the state has no business interfering with religious matters, because religion is a matter of achieving salvation, concerning only the individual believer and God. However influential this view may have been in the formative period of liberal thought, it does not seem to me to be a promising, let alone compelling, model for the idea that Rawls is putting forward. The responsibilities of society as a whole, acting through the state, are responsibilities to enable people to make what they will of their own lives, providing them, among other things, with a fair share of resources to use in pursuit of their chosen conception of the good. Citizens are not left with the space within which to decide what matters to them most, or even a space in which to express what matters to them most, but rather, they are provided with both the space and resources, that they can use to decide on and carry out a plan of life.

A better way of thinking about the special responsibility that each of us is said to have for how our own life goes begins with the thought that, although the private realm is private in the sense that it contrasts with the public realm of communal decision making and provision, there is another sense in which the private realm is inescapably public, because each of us does not engage in our private task of self-creation in the absence of or abstraction from others. Indeed, often the only way to realize a conception of the good is to do things with others - lifting something too heavy for one person to carry, building a building, or having and raising children are all things that private persons must do together if they do them at all. But even when one is engaged in a quintessentially private pursuit, tending one's own garden, there is a sense in which the activity is also essentially public. As I engage in my activity, and you engage in yours, if each of us has a special responsibility for what comes of our respective pursuits, the separateness of those pursuits must be reconciled in a way that preserves their independence. 66

 [*1832]  If we think of the private realm as essentially public in this way, it becomes immediately clear why something with the broad doctrinal structure of tort law would be required by the division of responsibility. Each of us has a special responsibility for how his or her own life goes; that responsibility can only be understood in terms of each of us being free to set and pursue our own plan of life in a way that is consistent with others being able to do the same, all of which must take place within the limits set by the just basic structure, including both political and civil liberties and such economic redistribution as is necessary. We do not each want as much freedom as possible, if that is conceived in abstraction from the requirements of a just basic structure; we each would want, instead, as much freedom to set and pursue our own conception of the good as we can have in a way that is consistent with others having the same. Now having the basic civil liberties, such as freedom of expression and freedom of thought, built in as constitutional essentials is one aspect of the interest that we have in that liberty. But it is only one aspect, because we do not merely wish to be able to conceive our conceptions of the good. Nor do we merely wish to be able to discuss those conceptions. Without minimizing the importance of freedom of thought and discussion, from the point of view of agents concerned with their ability to set and pursue their conception of the good, these things must not be the end of the story. We need to be able to actually pursue whatever life plan we choose. Having a fair share of resources is important. But so too is the respect in which we have those resources, and indeed, the sense in which we have the other "resources" important to setting and pursuing a conception of the good, namely our own bodily powers. My use of those bodily powers is, once again, subject to the demands of justice, and so my claim over them does not entitle me to refuse to yield up some of their fruits towards the needs of others, via the tax system. For that matter, it does not even entitle me to refuse to yield them to some pressing public purpose - perhaps I can be pressed into service to build a dike to prevent a flood, or even to fight a war against an enemy invader. All of this falls under the public side of the division of responsibility; society can demand that I do my fair share, just as I can demand that I receive my fair share from society. Moreover, both the individual and the society can make these demands on an ongoing basis. I do not receive my fair share only once, or discharge all of my obligations to society through a single payment.

What my claim to my bodily powers (and property) does entail, is that once these fair public arrangements are in place, no other  [*1833]  individual has any claim on my person or resources, unless I have somehow undertaken to give him or her such a claim. Absent contractual arrangements, or some peculiar set of private arrangements deemed by law, such as those between parents and children, others can make no demands on my person or property. Conversely, I can make no demands on theirs. To allow me to make further demands on others - to allow me, for example, to injure them, or damage their goods as I go about my pursuit of my own plan of life - would be inconsistent with the idea that each of us has a special responsibility for his or her own life, because it would allow me to displace the costs of my own pursuit of my own plan of life on to others. I also cannot make further demands on others by using what is theirs in pursuit of purposes that they do not share, borrowing their property without their consent, or tricking them into doing something on my behalf. In the same way, to allow others to make further demands on me would saddle me with the responsibility for how well those other people find that their lives go.

The division of responsibility thus requires an account of the sense in which we have what is our own. My share of primary goods must be mine, in the sense that it is at my disposal to use in setting and pursuing my own conception of the good. But I don't have those things at all if others can subordinate them to their pursuit of their own conceptions of the good, either by damaging them as a by-product of their pursuit of their purposes or by using them without my consent. If someone wrongs me in the first of these ways, they interfere with my ability to take responsibility for my own life in a familiar way, because they deprive me of the means I had to do so. The person who wrongs me in the second way, by using what is mine - that is, my person or property - without my consent violates my special responsibility in a different way, by making use of powers or goods that are mine in pursuit of something that is not part of my conception of the good. In so doing, such a person both advances his or her own conception of the good with more than his or her just share of resources and also violates my freedom by subordinating my power to decide which purposes to advance. I am pressed into service in pursuing a conception of the good that I do not accept.

The division of responsibility thus requires that I have what is mine as against all others, where the phrase "all others" is understood severally rather than jointly. Each person owes each other person duties of forbearance, duties to avoid using or damaging what belongs to another. At the same time, the other, public side of the division of responsibility provides the space within which the private part makes sense, so there is no objection to requiring people to do things, or surrender goods, for public purposes.

The idea that a liberal vision that allows for redistribution must allow people to have what is their own, and to use it as they see fit  [*1834]  may seem puzzling, if we suppose that the underlying motivation for the Rawlsian project is to find a way to divide up a pot of common goods on which nobody has any prior claim. Although Rawls does discuss how goods can be justly divided, the underlying motivation for setting the problem is much more Kantian than that description would suggest. Distributive shares are important because they enable choice. Rawls's formulation in A Theory of Justice is often read as resting on the empirical claim that people want as much as possible, and just as often criticized for assuming that people have exclusively or excessively individualistic and possessive conceptions of the good. 67 In Political Liberalism, Rawls preempts this misunderstanding by formulating the idea of primary goods in the language of constructivism. Primary goods are not the things that selfish people expect they will want prior to knowing the specific ways in which they are selfish. Instead, they are the things wanted by parties concerned to protect their own ability to decide for themselves how to live their lives. Like Kant, Rawls recognizes that to choose something, rather than merely wish for it, I must take myself to be in a position to act in pursuit of it. 68 Parties in the original position want income and wealth so that they are free to choose. 69 As such, they must have what is theirs in a way that enables them to choose, rather than subjecting them to the choices of others.

Thinking of my powers as enabling me to make choices does not depend on any idea about unencumbered or disembodied selves making choices in abstraction from everything. To the contrary, my body is that through which I act, and my bodily powers determine what I am able to do in pursuit of my own conception of the good. As we have seen, talk about a "conception of the good" has the potential to create the impression that all I need to do is conceive of my good. But my claim to my bodily powers in particular is my claim to be able to do things, to choose them rather than just wish for them, and to do so, as Rawls puts it, "over a complete life." 70

 [*1835]  The difference principle applies to the structure governing the results of social cooperation, and as such limits my claims on the fruits of my powers, and the use of my property. In correcting for the aggregate effects of legitimate private pursuits the difference principle does not presuppose any ideas about property being merely conventional or serving as a tool to promote appropriate distributions. 71

Understanding primary goods in this way enables the Rawlsian to provide a principled answer to both the libertarian dogmatist and the egalitarian skeptic. The libertarian dogmatist falls prey to Rawls's argument that we cannot attach sense to questions about what a person deserves, or what a person has produced, unless we have already determined fair distributive shares. So the libertarian dogmatist cannot talk about who is entitled to what in terms of what people have produced, without an independent account of what people are entitled to do with, or keep, of what is theirs. The dogmatist confuses principles governing particular private transactions with ones governing the other side of the division of responsibility. 72 My entitlement to my share of primary goods flows from principles of justice in distribution; so too does any entitlement I have to the fruits of my share of goods.

The egalitarian skeptic insists that ideas of tort and property are merely matters of positive law that raise no issue of justice, and so are to be decided at a later, legislative stage. He may hope to take comfort in Rawls's response to the libertarian dogmatist. But the  [*1836]  response to the libertarian dogmatist also defeats the skeptic. The root of the libertarian dogmatist's confusion is an illicit inference from the perfectly sound idea that it would be inconsistent with my freedom to set and pursue my own conception of the good to allow any other private person to force me to pursue his or her own purposes, by depriving me of what is rightfully mine, to the very different, and unsound idea that what is rightfully mine does not depend upon more general claims of justice, so that if no private party can take my property for his or her own purpose, then the state cannot tax me for its purposes. The egalitarian skeptic sees that the conclusion of the inference is false, and so concludes that the premise must also be false, and so insists that property rights, and private obligations more generally, must be mere instruments in pursuit of some distributive goal. The Rawlsian liberal is the one who sees that although the dogmatist's conclusion is false, its falsity does nothing to impugn the premise, because the premise does not entail the conclusion. That just is the division of responsibility: Different rules regulate the justice of the basic structure and the justice of individual transactions, both voluntary and involuntary.

Moreover, when it comes to tort law, the Rawlsian does not need to accept the Lockean picture of the relation between rights to the person and rights to property that animates dogmatists and skeptics alike. The law of tort treats persons and property in parallel fashion, despite the obvious differences between them. There are both trespasses against the person and trespasses against property, as well as injuries to both person and property, and the set of obligations governing each are strikingly similar. The dogmatist celebrates this parallel, imagining it to show that property just is an extension of personhood. The skeptic is more likely to be embarrassed by it, if only because the claim that property rights are merely conventional and instrumental is so much easier to advance than the parallel claim that rights to personal security is. If both are in the service of egalitarian aims, why do they receive the same protection? (Of course, the egalitarian might say that they should not.) Again, the Rawlsian answer is that as between private parties, the division of responsibility mandates that we treat each person as free to do as he or she sees fit with the means at his or her disposal, that is, to use both bodily powers and income and wealth in pursuit of his or her own conception of the good, in a way consistent with others having the same freedom to pursue their conception of the good. All of this is done within a broader framework of redistributive institutions, the other half of the division of responsibility. Personal powers and property are alike in their significance to the claims of other private parties, and so subject to parallel treatment in tort.

To sum up, parties choose principles governing the distribution of primary goods because they are concerned with their ability to set and  [*1837]  pursue their own conception of the good. They do not ask what each of them is prepared to do for all the others; they ask what coercive structures they should select so that each may have freedom to pursue his or her own conception of the good in a way that is consistent with everyone else being able to do the same. But they can only ask that if they have entitlements to use their shares, whatever they are, in pursuit of their own conceptions of the good, that is, without needing to subordinate those pursuits to the particular pursuits of others, to do so would be to give up on the idea of independence. 73

V. Tort and Contract
 
I want to highlight the structure of the argument so far by focusing on the example that Rawls considers of the institutional significance of the division of responsibility. In Political Liberalism, Rawls notes that there is a division of labor between two kinds of rules, those of distribution and those of transfer. 74 This division of labor follows from the division of responsibility. His example of the latter sort of rule is the law of contract. We can see why it is required by the idea of each person's responsibility for his or her own life. My interest in having both my own powers and my own fair share of primary goods is an interest in having those things at my disposal, that is, to have them available to me, so as to pursue and revise my own conception of the good. Rawls describes primary goods as "all-purpose means" precisely because they are things that may be used in pursuit of a wide variety of purposes, and that make it possible for me to take up various particular purposes. 75 But one of the most important ways in which I will be able to pursue my own conception of the good is by entering into private arrangements with other people, including the joint pursuit of common ends and economic transactions that enable each of us to pursue our own separate ends in a way that is to the advantage of both. I can only do such things provided that I am able to give what is mine to another person, either by giving them my  [*1838]  goods or by performing services for them. If I and those with whom I enter into arrangements are separate persons, with separate purposes and, more to the immediate point, separate schedules, then one of the primary ways in which we could enter into mutually advantageous transactions is by each of us being able to transfer our entitlements to things - both goods and services - separately from the actual delivery of those things. The ability to subordinate our share of our primary goods to our own private conceptions of the good thus requires that we be able to enter into forward contracts. Moreover, it requires that those contracts be enforceable, just in the sense that the law must deem the obligation to transfer to have taken effect at the moment of contract formation. If I promise to cut your lawn next Wednesday, I thereby give you the right to have me cut your lawn. If I fail to do so at the specified time, you still have the right that I do so, and so are entitled either to performance or to damages in lieu of performance.

The power to enter into contracts makes distributive shares valuable to people in pursuing their own conceptions of the good. The power to have those contracts enforced does not merely provide an incentive to compliance. More significantly, it makes voluntary exchanges possible, by enabling people who exchange things to get what they bargained for, in the sense of what they expected when they entered into those arrangements.

Matters are no doubt much more complicated than this initial sketch suggests: there surely are some powers that cannot be transferred, and there are limits on the contexts within which such a transfer is binding (although one would hope that the fair distribution of primary goods in the society would do much to see to it that those limits were generally satisfied). Nonetheless, the point of the example is to illustrate the more general ways in which the division of responsibility presupposes some idea of private ordering. The sketch also suggests that we do not need to ask whether this set of enforceable institutions would be acceptable to parties in the original position. Unless we assume some such set of arrangements, it is difficult to see why parties in the original position would be interested in their shares of primary goods at all. Absent the power to transfer, including the power to transfer in advance, through promise, the distributive shares would own the people as much as the people owned the distributive shares. 76

None of this is inconsistent with the idea Rawls defends in Political Liberalism, according to which the aggregate effects of individual transactions may lead to injustices in holdings that need to be addressed. The point is simply that the need to address those  [*1839]  inequalities in no way undermines the importance of particular transactions to the relation between primary goods and freedom.

Contract is necessary to the ability to use distributive shares because it governs voluntary transactions. Tort is just as important because it governs involuntary transactions, that is, those transactions in which one person suffers a loss (or is used for gain) at the hands of another. If what is mine is not subject to my choice, but to yours - which happens when you use what is mine without my consent - or vulnerable to the effects of your choice - which happens when you injure me or my goods - then it is not mine to use in setting and pursuing my own conception of the good.

By focusing on tort and contract this way, I have attributed to Rawls a broadly Kantian understanding of the nature of private rights. To do so seems to me to be the best way of understanding his commitment to constructivism, as well as the way in which parties in the original position are concerned to maximize their expected shares of primary goods, rather than, for example, welfare. The Kantian conception has a further advantage as well, in that it is congruent with Rawls's focus, especially in Political Liberalism, on the idea of the centrality of the coercive structure to fundamental issues of justice. Like Rawls, Kant sees the fundamental question of political philosophy to be one about the legitimate use of coercion, rather than seeing it as primarily about moral obligations, which may be enforced coercively when it is effective to do so. By focusing on the idea of freedom, and what it is to interfere with the freedom of another, we get an account of tort law that is concerned with the basis and limits of enforceable obligations. Rawls does not need to say that there is a moral obligation to repair the wrongful injuries I cause, and that at some later, legislative stage, the state decides whether to enforce it. Instead, he can say that when private citizens have their rightful shares of primary goods, anyone who violates their private rights wrongfully interferes with their freedom, and so the aggrieved party has an enforceable entitlement to be put back into the situation he or she was in. 77

VI. Nonfeasance and Misfeasance
 
We are now in a position to restore the distinction between nonfeasance and misfeasance to its proper place. When choosing, free and equal persons are interested in their distributive shares because they want to be able to do things with them; they can only be interested in shares of primary goods if they have them in the requisite sense. 78 That sense of having is prior to any considerations  [*1840]  about what parties would agree to, because there is no basis for agreement unless people already know what it is to have things as their own, as against all others. I only have things as my own to use in forming, pursuing and revising my conception of the good inasmuch others are under an obligation to avoid interfering with them. So they must be under an obligation to avoid using or injuring my person or property. As private parties, they cannot be under any corresponding obligation to confer any benefit on me, no matter how significant the benefit, and no matter how easy it is to confer. Such an obligation would undermine the sense in which what they have is their own. That is just the distinction between nonfeasance and misfeasance.

This brings us back to the law of tort. Almost anything I do will have some effects on other people. The way in which I pursue my own conception of the good may lead me to take up opportunities that, as a result, will no longer be available to others, whether by acquiring things that others might have wanted to pursue their purposes, by entering into various arrangements, personal and otherwise, with others that preclude some other person from entering into the same relation with the same person, and by exercising choices in a way that makes some popular things more expensive for others. All of these effects that one person might have on another are consistent with each of us having a special responsibility for how our own life goes, because they are simply the inevitable side effects of separate persons making separate decisions in the presence of others. But there are other ways in which we have effects on others that are different. If I use what is yours without your consent, then I subordinate your pursuit of your purposes to my pursuit of mine. If I injure you, or damage your goods, I prevent you from using your powers to set and pursue your own conception of the good. So while the former class of side effects must simply be accepted 79 as inevitable, the latter set is inconsistent with each of us having a special responsibility for our own life. However, to say that they are inconsistent is not to say that they will never happen, and here too, the division of responsibility sheds considerable light on the doctrinal structure of tort law. If I wrongfully injure you, I am liable to you in damages, just because the payment of damages aims to "make you whole," that is, to restore to you, as much as it is possible to do so, means equivalent to those of yours that I have injured. To put you back in the same place is to put you back in the same place with  [*1841]  respect to your ability to set and pursue your own conception of the good.

It is worth noting that on this understanding of the division of responsibility, the payment of damages does not serve to restore a previous distribution that is judged to be just on the basis of, for example, the difference principle. The pattern of holdings in society will change whenever one person wrongs another, and, when damages are set out to make up that wrong, to right it, the pattern of holdings will change yet again, but they will not be returned to the pattern before the wrong took place. If I injure you through my carelessness, for example, you will end up with less at your disposal. If I am compelled to repair the wrong, so that you are restored to the situation in which you are in a position to decide what to make of your life, I will end up in a worse position than I was in beforehand. But this is just what we should expect from the point of view of the division of responsibility, because my ending up worse off is a consequence of my own special responsibility for how my own life goes. It is as though I had injured myself, or damaged my own property. Of course, I end up with less than I would have had there been no such injury, but, because I have a special responsibility for how my own life goes (but not for how your life goes, nor you for mine) it is consistent with that responsibility that, through my voluntary conduct, should I behave carelessly, I bear the costs of that carelessness. The person who uses what is mine without my consent, either by borrowing my property, or duping me into doing something for her, can also be made to repair any loss I suffered, or alternatively to surrender any gain she has acquired, because I must be the one who decides the purposes for which my goods will be used. 80

The division of responsibility also solves the other problem that faced the contractarian/consequentialist approach. The  [*1842]  contractarian/consequentialist approach had no room for the distinction between nonfeasance and misfeasance, because it seemed to focus all of the attention on how much of a burden one person may be required to bear for the sake of another. But that question must be, both as a matter of common sense morality and as a matter of tort doctrine, a question that only comes up secondarily, that is, a question that only comes up when we have already established that one person owes another a duty to look out for some interest or other. The division of responsibility tells us that as private citizens, we must look out for the interests in person and property of others; as such, it does not tell us exactly how much we should do with respect to those interests. But without the division of responsibility, if we look only at how much one person's interest burdens the freedom of another, it seems that the only answer can be the sort of aggregative one that we saw above.

With the division of responsibility in place, we still have a question left about how much care one person can demand of another. The answer, both legal and Rawlsian, must of course be "reasonable" care. Reasonable persons, Rawls tells us in Political Liberalism, "are not moved by the general good as such but desire for its own sake a social world in which they, as free and equal, can cooperate with others on terms all can accept." 81 How much care is reasonable is fixed by asking what is consistent with the interests that all have in liberty and security. 82 But we can only get to that question once we have already demarcated the boundaries of duty, as the division of responsibility enables us to.

It is perhaps worth mentioning that a further advantage of thinking about the law of tort in terms of the division of responsibility is that it gives a ready answer to the question of whether, why, and most of all, when we should take an interest in the effects that one person's deeds have upon another. The Rawlsian answer is that we take such an interest in circumstances in which we are more or less satisfied with the justice of the background circumstances against which they interact. Where those background circumstances are not just, we still have no difficulty in saying just who did what to whom - I broke your vase, you broke your promise, and so on - but we may wonder about the wisdom of enforcing any of the rights that are so violated. The division of responsibility enables us to articulate the reason: The special responsibility that each of us has for how our own life goes is a responsibility that we can only be held to in conditions of justice; we cannot hold somebody liable in damages except as a way of recognizing that person's special responsibility for his or her own life,  [*1843]  in the context of a plurality of separate persons with separate purposes.

VII.
 
"Non-Ideal Theory"
 
The Rawlsian account of tort law I have outlined gets its normative impetus from the background justice of the society as a whole. If a society is unjust in its liberties and opportunities, or in the distribution of primary goods, we must work with what Rawls calls "non-ideal theory." 83 There is an interesting and difficult question about whether, or to what extent, private rights are legitimately enforceable where background justice is not satisfied. I can only gesture at the beginnings of an answer here. If private holdings are inconsistent with justice, there are bound to be misgivings about treating those holdings as enforceable. At the same time, it is far from clear that those misgivings generate an entitlement for private parties to disregard the safety of others with impunity. Certainly few would wish to argue that because disparities in wealth are so great in modern societies, wealthy people should be free to disregard the person and property of poor people. We might further doubt whether any particular person can rightly appeal to the overall structure of society as a whole in response to a tort action, if only because those who are in the best position to raise such an issue will be the very people who are too poor to be sued anyway.

The problem here is a special case of a much more general kind - the claim to enforceable rights, including enforceable private rights, depends on the existence of a largely just basic structure. Where the basic structure is entirely unjust, Rawlsian theory may well say that there are no enforceable rights at all. 84 Where the structure is largely just, but not entirely so, Rawlsian theory must be silent about how to classify particular cases.

Even if the lack of a sufficiently just basic structure raises questions about the extent to which the division of responsibility can justify existing tort practice, that division does enable us to understand the kind of justice to which tort law must aspire if it is to be consistent with the ideas of freedom and equality at the root of the Rawlsian conception of justice. The law of tort rightly concerns itself only with discrete transactions involving two private parties; its claim to do so as a matter of justice must be understood as a claim to be doing justice between those parties, because, as between the two of them, neither has a claim on the powers or resources of the other. On this understanding, tort doctrine is only secondarily about who pays; the primary focus is on how people are allowed to treat each other. 85

 [*1844] 

Conclusion
 
The division of responsibility animates Rawls's use of the social contract in a way that avoids the objection that he is engaged in merely consequentialist reasoning. The parties in the original position are not simply out to advance their purposes; they are concerned to protect their ability to take responsibility for their choices as they set, pursue, and revise their own conceptions of the good. The ability to do so, consistent with the equal freedom of others, requires a just constitutional structure, specifying their rights against society as a whole, and adequate shares of other primary goods, including income and wealth. It thus provides a powerful account of distributive justice. I have argued that it also provides a powerful account of corrective justice, one that explains why it is neither the master nor the servant of distributive concerns. Each person has a special responsibility for his or her own life, not only as against the state, but also as against any other private party. This idea of independence explains the place in Rawlsian doctrine for relational ideas of duty, proximity, and causation in ordering relationships between private parties. Our separate pursuits must be rendered consistent if each of us is to be responsible for what we make of our own lives.

Legal Topics:

For related research and practice materials, see the following legal topics:
Constitutional LawBill of RightsFundamental RightsEminent Domain & TakingsCriminal Law & ProcedureSentencingAdjustmentsGovernmentsLegislationTypes of Statutes

FOOTNOTES:


Click here to return to the footnote reference.n1. John Rawls, A Theory of Justice 3, at 10 (rev. ed. 1999).



Click here to return to the footnote reference.n2. For examples of the first, see Gregory C. Keating, Reasonableness and Rationality in Negligence Theory, 48 Stan. L. Rev. 311 (1996), and Hanoch Dagan, Qualitative Judgments and Social Criticism in Private Law: A Comment on Professor Keating, 4 Theoretical Inquiries L. 89 (2003). For the second, see Richard A. Epstein, A Theory of Strict Liability, 2 J. Legal Stud. 151 (1973).



Click here to return to the footnote reference.n3. John Rawls, Social Unity and Primary Goods, in Utilitarianism and Beyond 159 (Amartya Sen & Bernard Williams eds., 1982) [hereinafter Rawls, Social Unity].



Click here to return to the footnote reference.n4. John Rawls, Justice as Fairness: A Restatement 17 (Erin Kelly ed., 2001).



Click here to return to the footnote reference.n5. Rawls, Social Unity, supra note 3, at 159.



Click here to return to the footnote reference.n6. John Rawls, Political Liberalism 268 (1993).



Click here to return to the footnote reference.n7. Id. at 269.



Click here to return to the footnote reference.n8. Rawls, Justice as Fairness: A Restatement, supra note 4, at 138 & nn.



Click here to return to the footnote reference.n9. It may be more than a passing coincidence, then, that many contemporary libertarians adopt a conception of freedom and responsibility that is fundamentally Leibnizian!



Click here to return to the footnote reference.n10. See Desiderius Erasmus, De Libero Arbitrio (1524), reprinted in Desiderius Erasmus, Controversies (Peter Macardle & Clarence H. Miller trans., Charles Trinkaus ed., 1999); Richard H. Popkin, The History of Skepticism from Erasmus to Spinoza 5-8 (1979).



Click here to return to the footnote reference.n11. See Gerald A. Cohen, If You're An Egalitarian, How Come You're So Rich? 134-47 (2000); Liam B. Murphy, Institutions and the Demands of Justice, 27 Phil. & Pub. Aff. 251 (1998).



Click here to return to the footnote reference.n12. See generally Liam Murphy & Thomas Nagel, The Myth of Ownership: Taxes and Justice (2002).



Click here to return to the footnote reference.n13. See Rawls, Political Liberalism, supra note 6, at 267.



Click here to return to the footnote reference.n14. See Samuel Freeman, Illiberal Libertarians: Why Libertarianism Is Not a Liberal View, 30 Phil. & Pub. Aff. 105 (2001).



Click here to return to the footnote reference.n15. This view appears to have its origins in the work of Hans Kelsen, but has been widely influential among those influenced by H.L.A. Hart's legal positivism.



Click here to return to the footnote reference.n16. I am sure that there are some examples of powers that are merely delegated. Perhaps the right to make a citizen's arrest is one such example. My point is only to claim that the ordinary obligations of the citizens - the obligation to respect the person and property of others, as well as the obligation to repair wrongs one has committed against others - are not justified by their contribution to the state's legitimate purpose of providing fair equality of opportunity, or any other legitimate distributive purpose that the state might be thought to have.



Click here to return to the footnote reference.n17. Rawls, A Theory of Justice, supra note 1, 4, at 15.



Click here to return to the footnote reference.n18. See Anthony T. Kronman, Contract Law and Distributive Justice, 89 Yale L.J. 472 (1980).



Click here to return to the footnote reference.n19. See Frank I. Michelman, Property, Utility, and Fairness: Comments on the Ethical Foundations of "Just Compensation" Law, 80 Harv. L. Rev. 1165, 1233 (1967).



Click here to return to the footnote reference.n20. Rawls, A Theory of Justice, supra note 1, 87, at 514.



Click here to return to the footnote reference.n21. See George P. Fletcher, Fairness and Utility in Tort Theory, 85 Harv. L. Rev. 537, 555 (1972).



Click here to return to the footnote reference.n22. See Thomas Pogge, Three Problems with Contractarian-Consequentialist Ways of Assessing Social Institutions, 12 Soc. Phil. & Pol'y 241 (1995), reprinted in The Just Society 241 (Ellen Frankel Paul et al. eds., 1995).



Click here to return to the footnote reference.n23. Palsgraf v. Long Island R.R., 162 N.E. 99, 100 (N.Y. 1928).



Click here to return to the footnote reference.n24. I do not mean to deny that a legitimate state can set up institutions to advance other social purposes. My point is only that there can be no argument of Rawlsian justice for creating such institutions, the operation of which is triggered by an injury one person accidentally causes another. I discuss this point in more detail below. See infra note 74 and accompanying text.



Click here to return to the footnote reference.n25. Rawls, A Theory of Justice, supra note 1, 4, at 18.



Click here to return to the footnote reference.n26. "The Maximin rule tells us to rank alternatives by their worst possible outcomes; we are to adopt the alternative the worst outcome of which is superior to the worst outcomes of the others." See id. 26, at 133.



Click here to return to the footnote reference.n27. See Cohen, supra note 11, at 134-47.



Click here to return to the footnote reference.n28. See Rawls, Political Liberalism, supra note 6, at 19.



Click here to return to the footnote reference.n29. Strict liability is a prominent feature of the American tort landscape, particularly with respect to products liability. Yet the leading case for understanding the basis of strict liability turns out to be the nineteenth century English case of Fletcher v. Rylands, 1 L.R.-Ex. 265 (1866), aff'd sub. nom. Rylands v. Fletcher, 3 L.R.-E.&I. App. 330 (H.L. 1868). Rylands has pretty much lost its steam as a precedent within the rest of the common law world, including England. See Cambridge Water Co. v. Eastern Counties Leather plc., 2 A.C. 264 (H.L. 1993) (Lord Goff, J.); Burnie Port Auth. v. Gen. Jones Pty Ltd., 179 C.L.R. 520 (H.C. 1994) (Mason, C.J.). More recently, the House of Lords has concluded that Rylands was a nuisance case, and that its ruling applies only to uses of land. See Transco Plc v. Stockport Metro. Borough Council, 2003 WL 22656452 (H.L. Nov. 19, 2003).



Click here to return to the footnote reference.n30. As Peter Glassen and Michael Thompson have each independently pointed out, strictly speaking, it is dikaialogical rather than deontological, from the Greek dike for right, rather than deon for duty. See Peter Glassen, The Classes of Moral Terms, 11 Methodos 223 (1959); Michael Thompson, What Is It to Wrong Someone? Reason and Value Themes from the Moral Philosophy of Joseph Raz 333, 336 (R.J. Wallace et al. eds., 2004).



Click here to return to the footnote reference.n31. Oliver Wendell Holmes, Jr., The Common Law 77 (1881) (Lecture III).



Click here to return to the footnote reference.n32. See Epstein, supra note 2; Stephen R. Perry, The Impossibility of General Strict Liability, 1 Can. J.L. & Jurisprudence 147 (1988).



Click here to return to the footnote reference.n33. See Perry, supra note 32.



Click here to return to the footnote reference.n34. See, e.g., Gregory C. Keating, A Social Contract Conception of the Tort Law of Accidents, in Philosophy and the Law of Torts 22 (Gerald J. Postema ed., 2001).



Click here to return to the footnote reference.n35. See, e.g., Richard W. Wright, The Standards of Care in Negligence Law, in Philosophical Foundations of Tort Law 249 (David G. Owen ed., 1995).



Click here to return to the footnote reference.n36. See Pogge, supra note 22.



Click here to return to the footnote reference.n37. See Rawls, A Theory of Justice, supra note 1, 38, at 212-13.



Click here to return to the footnote reference.n38. See Pogge, supra note 22. I am familiar with George Fletcher's thoughts on this matter through personal communication I have had with him.



Click here to return to the footnote reference.n39. See Rawls, A Theory of Justice, supra note 1, 38, at 213.



Click here to return to the footnote reference.n40. Id. at 212.



Click here to return to the footnote reference.n41. I believe that the root of the difficulty here is Rawls's uncharacteristically hasty acceptance of H.L.A. Hart's account of responsibility in criminal law. Hart's more general positivism and conventionalism about the rule of law sit uneasily with Rawls's more Kantian commitment to that idea.



Click here to return to the footnote reference.n42. See Andrew Simester, Is Strict Liability Really Wrong? (unpublished manuscript on file with the author).



Click here to return to the footnote reference.n43. Bernard Williams, A Critique of Utilitarianism, in Utilitarianism: For and Against 82 (J.J.C. Smart & Bernard Williams eds., 1973).



Click here to return to the footnote reference.n44. See Joel Feinberg, The Moral and Legal Responsibility of the Bad Samaritan, in Freedom and Fulfillment 175 (1992); James Barr Ames, Law and Morals, 22 Harv. L. Rev. 97 (1908-09); Ernest J. Weinrib, The Case for a Duty to Rescue, 90 Yale L.J. 247 (1980). I examine the issue of duties to rescue from the standpoint of the division of responsibility in Three Duties to Rescue: Moral, Civil, and Criminal, 19 L. & Phil. 751 (2000).



Click here to return to the footnote reference.n45. See Rickards v. Sun Oil Co., 41 A.2d 267, 269 (N.J. 1945).



Click here to return to the footnote reference.n46. I am grateful to Sergio Tenenbaum for this formulation.



Click here to return to the footnote reference.n47. See Immanuel Kant, The Critique of Practical Reason 5:9 (Mary Gregor ed., Cambridge Univ. Press 1997) (1788).



Click here to return to the footnote reference.n48. See Gregory C. Keating, The Idea of Fairness in the Law of Enterprise Liability, 95 Mich. L. Rev. 1266 (1997).



Click here to return to the footnote reference.n49. Gregory C. Keating, Distributive and Corrective Justice in the Tort Law of Accidents, 74 S. Cal. L. Rev. 193, 196 (2000).



Click here to return to the footnote reference.n50. "The relevance of Rawls' account of the equal basic liberties is thus not direct, but general and analogical: The concepts and categories of equal basic liberties provide an incisive framework for analyzing the problems of accidental injury and death. Those problems pit two liberties - the freedom of injurers and the security of victims - against one another." Keating, supra note 2, at 321.



Click here to return to the footnote reference.n51. Keating draws on Holmes's distinction between "the world of acts" and the "world of activities." See Oliver Wendell Holmes, The Path of the Law, in Collected Legal Papers 167, 183 (Peter Smith ed., 1952) (1920).



Click here to return to the footnote reference.n52. See Keating, supra note 2.



Click here to return to the footnote reference.n53. See Gregory C. Keating, Pressing Precaution Beyond the Point of Cost-Justification, 56 Vand. L. Rev. 653, 679 (2003).



Click here to return to the footnote reference.n54. Id.



Click here to return to the footnote reference.n55. See generally Louis Kaplow & Steven Shavell, Fairness Versus Welfare (2002). I criticize Kaplow and Shavell's combination of a refusal to engage with tort doctrine and readiness to suppose that accidental injury by others provides a stable subject matter for inquiry in Too Much Invested to Quit, 20 Econ. & Phil. (forthcoming 2004) [hereinafter Ripstein, Too Much Invested to Quit].



Click here to return to the footnote reference.n56. See generally Charles Fried & David Rosenberg, Making Tort Law (2003).



Click here to return to the footnote reference.n57. Guido Calabresi, The Costs of Accidents: A Legal and Economic Analysis (1970).



Click here to return to the footnote reference.n58. I demonstrate the difficulties of using arguments about what people would agree to in attempting to undermine claims of fairness in Ripstein, Too Much Invested to Quit, supra note 55 (manuscript at 17, on file with author).



Click here to return to the footnote reference.n59. Keating, supra note 2, at 321 (citing T.M. Scanlon, Jr., The Significance of Choice, in 8 The Tanner Lectures on Human Values 149, 174 (Sterling M. McMurrin ed., 1988)).



Click here to return to the footnote reference.n60. At various points Keating characterizes his account as providing an interpretation of existing tort doctrine. See, e.g., Keating, supra note 48. It is not entirely clear how this claim is supposed to be consistent with the idea, also central to his analysis, that tort gives expression to "natural duties" prohibiting injury to the person of others. Existing tort practice treats personal injury and property damages in parallel ways.



Click here to return to the footnote reference.n61. Rawls, Justice as Fairness: A Restatement, supra note 4, 9, at 28.



Click here to return to the footnote reference.n62. Rawls, A Theory of Justice, supra note 1, 3, at 10.



Click here to return to the footnote reference.n63. Rawls, Political Liberalism, supra note 6, at 265-69.



Click here to return to the footnote reference.n64. Rawls, Social Unity, supra note 3, at 170.



Click here to return to the footnote reference.n65. (James H. Tully ed., Hackett Publishing Co. 1983) (1689).



Click here to return to the footnote reference.n66. This way of conceiving of the division of responsibility has the further advantage of providing a principled response to Pogge's objection that the contract argument is consequentialist, and to Ronald Dworkin's argument that Rawls's difference principle, with its focus on the worst off, without any attention to how they became worst off, has no room for ideas of individual responsibility. See Ronald Dworkin, Sovereign Virtue: The Theory and Practice of Equality 113-18 (2000). Rawls's focus on the expectation of the worst off presupposes a structure of interaction in which what happens to a particular person depends on that person's choices. Those who decide not to develop their talents are not entitled to a set welfare or income floor, but to a social structure in which they can make what they choose to of their abilities and fair shares.



Click here to return to the footnote reference.n67. See the discussion of this issue in Will Kymlicka, Liberal Individualism and Liberal Neutrality, 99 Ethics 883 (1989).



Click here to return to the footnote reference.n68. This point is explicit in Kant. See Immanuel Kant, The Metaphysics of Morals 3 (Mary Gregor trans., Cambridge U. Press 1996) (1797).



Click here to return to the footnote reference.n69. This way of understanding the significance of primary goods underwrites the assumption that parties in the original position do not know their own conception of the good. Their interest in freedom demands that they choose their own conception of the good, so as to enable themselves to do so. The idea of freedom precludes the possibility that a person's conception of the good is simply a brute fact about them. As Rawls puts it in Social Unity and Primary Goods, "the use of primary goods, however, relies on a capacity to assume responsibility for our ends." See Rawls, Social Unity, supra note 3, at 169. He makes the same point when he describes means as necessary to the "the capacity to form, to revise, and rationally to pursue ... a conception of what we regard for us as a worthwhile human life." See Rawls, Political Liberalism, supra note 6, at 302.



Click here to return to the footnote reference.n70. Rawls, Political Liberalism, supra note 6, at 335.



Click here to return to the footnote reference.n71. Critics of private property sometimes point to the "bundle" analysis of property introduced by Hohfeld, see Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning, 26 Yale L.J. 710, 719 (1917), and made prominent by Tony Honore, see A.M. Honore, Ownership, in Oxford Essays in Jurisprudence 107 (A.G. Guest ed., 1961). Its strengths as an analysis of the concept of property remain controversial. See, e.g., J.W. Harris, Property and Justice (1996); James Penner, The Idea of Property in Law (1996). Whatever its merits, however, it is of no help in addressing issues of the appropriate normative conception of property. The rights that individuals have to their person and powers can also be represented as a bundle, but few who claim a concern with justice would want to conclude that individual rights against personal injury and exploitation are merely tools to be used in pursuit of some socially desirable outcome. The state does have the power to limit freedom of the person, not only through taxation and the criminal law, but also through quarantine in the case of an epidemic. But the legitimacy of quarantine does not show that the justification of the right to security of the person is instrumental. In the case of property, Hohfeld's own analysis confirms this point. See Hohfeld, supra, at 719. It starts from the idea that property is a matter of relations between persons with respect to things, and the related idea that a single person can stand in different relations to different persons with respect to the same thing. Thus it does not exclude the idea being explored here, according to which the justification of property rights against other citizens does not preclude a public power to tax property or regulate its uses. Conversely, the public power to regulate or tax does not preclude private rights.



Click here to return to the footnote reference.n72. At the level of doctrinal detail, most libertarians appear to misconstrue the principles governing private transactions as well, favoring a causation-based account of strict liability. See, e.g., Epstein, supra note 2.



Click here to return to the footnote reference.n73. What of the possibility of something like the system of social insurance for accidents that briefly replaced tort in New Zealand? The arguments that led to its adoption are all at odds with the Rawlsian view under consideration here. See P.S. Atiyah, Accidents, Compensation and the Law (1970); P.S. Atiyah, The Damages Lottery (1997). I examine these in Some Recent Obituaries of Tort Law, 48 U. Toronto L.J. 561 (1998). Nonetheless, we can imagine a society legitimately deciding to require that everyone carry liability insurance, and, for administrative reasons, dispensing with any requirement of proof of negligence. If such a system were operated by the state, it would approximate such a system. What would be unacceptable from the point of view of Rawlsian justice, however, is a system that displaced private rights in favor of a standardized schedule of payments.



Click here to return to the footnote reference.n74. Rawls, Political Liberalism, supra note 6, at 268-69.



Click here to return to the footnote reference.n75. John Rawls, A Theory of Justice 93 (1971). Rawls abandons this vocabulary in later writings, but only because he worries that it appears to be psychological in a way that it is not meant to be. Preface to the French Edition of A Theory of Justice, reprinted in Rawls Collected Papers 417 (Samuel Freeman ed., 1999).



Click here to return to the footnote reference.n76. Kant, supra note 68, at 125-38. Discussing perpetuities, Kant notes that their result would be that the land inherited the people instead of vice-versa. See id.



Click here to return to the footnote reference.n77. I develop these Kantian themes in more detail in Authority & Coercion, 32 Phil. & Pub. Aff. 2 (2004) [hereinafter Ripstein, Authority & Coercion].



Click here to return to the footnote reference.n78. The situation thus contrasts with that of parties we might imagine choosing a distribution of welfare. Welfare (if there is such a thing that can be distributed) must be a final end, so that there is no further question of what parties will do with it. Primary goods can only be wanted if we know how we are allowed to use them. This point has been lost on many readers of Rawls, perhaps because they have been misled by his generosity in attributing the idea of an original position to Harsanyi into the conclusion that the parties in the original position simply want as much as possible without knowing what they might be entitled to do with their shares.



Click here to return to the footnote reference.n79. See Bolton v. Stone, [1951] App. Cas. 850 (appeal taken from Eng.).



Click here to return to the footnote reference.n80. For the same reason, the law of damages is not simply a matter of enforcing a prior moral obligation to avoid injury to others. People do owe each other such natural moral duties, but to suppose that the law does simply give effect to them is to make a mistake parallel to that of supposing that the difference principle is enforceable in a just society as a way of giving effect to a natural duty of charity to those in need. In each case, the confusion comes from rejecting the Rawlsian insight that justice is distinctive because claims of justice are enforceable as reciprocal limits of human freedom. The use of force is consistent with freedom only when it is part of a set of reciprocal limits on freedom. In the case of the difference principle, ignoring this aspect of Rawlsian justice leads to objections about the limited causal impact of the coercive structure when compared with other aspects of social life. See, e.g., Cohen, supra note 11, at 134-47. In the case of the law of tort, ignoring this aspect of justice leads to the confusion of the idea of one person wronging another with that of one person behaving badly, and so to familiar "moral luck" objections to tort liability. See, e.g., Jeremy Waldron, Moments of Carelessness and Massive Loss, in Philosophical Foundations of Tort Law 387 (David G. Owen ed., 1995). In each case, the objection regards coercion as an instrument for inducing virtuous behavior, rather than in light of its relationship to freedom. I examine these issues in more detail in Ripstein, Authority & Coercion, supra note 77.



Click here to return to the footnote reference.n81. Rawls, Political Liberalism, supra note 6, at 50.



Click here to return to the footnote reference.n82. I develop this Rawlsian account of the reasonable in the law of negligence in my Equality, Responsibility, and the Law ch. 1-4 (1999).



Click here to return to the footnote reference.n83. Rawls, A Theory of Justice, supra note 1, 39, at 216.



Click here to return to the footnote reference.n84. See Kant, supra note 68, at [delta] 41.



Click here to return to the footnote reference.n85. Talk about each person having a special responsibility for his or her own life may seem artificial, when measured against the actual practices of tort litigation. Like the medieval and early modern fact patterns that are the stuff of the torts classroom - separating fighting dogs with a stick, carelessly helping a passenger onto a departing train, runaway horses and safe harbors in a storm - the analysis offered here may seem to overlook the ways in which money determines which precautions people take, and who makes it to court, both in the form of expensive procedure and in the form of defendants who are "judgment proof." Again, the defendant in most tort litigation is not some private party making his or her own way in the world, but rather some large corporation, that produced a dangerous or harmful product. Since tort liability does not typically make a significant difference to defendant's ability to make his, her, or its way in the world, it might be thought that it makes no real contact with the division of responsibility. Far from being irrelevant, the division of responsibility provides the lens through which we can articulate the precise nature and dimensions of the problem. Massive inequalities of wealth and power threaten the ability of private parties to take responsibility for their own lives, free from the interferences of others. That is why redistributive institutions are necessary: to protect independence, even if the threat to it arises through the aggregative effects of particular transactions, none of which is objectionable in its own right.




[PHIL 1A] [Syllabus] [Handouts] [Home] [Bio] [CV] [PHIL DEPT.] [E-MAIL]

Send comments to: Andreas Teuber
URL:   http://phils7.dce.harvard.edu/
Last Modified: 01/16/05
Instructor's Toolkit
Copyright © The President and Fellows of Harvard College