Negligence, compensation, and the coherence of tort law
The ongoing debate over the appropriate purpose of tort liability has not adequately considered the long-held belief that the tort system is designed to compensate physical injuries. The neglect may seem surprising. After all, the compensation rationale for tort liability appears in the Restatement (Second) of Torts, which states that “the purposes for which actions of tort are maintainable . . . are: (a) to give compensation, indemnity or restitution for harms; (b) to determine rights; (c) to punish wrongdoers and deter wrongful conduct; and (d) to vindicate parties and deter retaliation or violent and unlawful self-help.”1
The neglect of the compensation rationale becomes understandable in light of the tort system’s general reliance on negligence liability. A compensatory objective would seem to justify a tort regime based on strict liability, yet tort compensation typically is limited to injuries caused by unreasonable or negligent behavior. Even negligence liability is limited in important ways, providing no compensation to numerous individuals who suffer emotional or economic harms caused by the negligent behavior of another. Negligence doctrine would seem to contradict the compensation rationale for tort liability.
The compensation rationale is not invoked by the draft Restatement (Third) of Torts: Basic Principles, which instead relies on the justification that negligence liability “remedies an injustice” and supplies an incentive to deter such unjust behavior.2 This justification, despite its apparent logic, may be inconsistent with important tort doctrines. Eliminating that inconsistency yields a compensatory justification for tort liability.
If the purpose of negligence liability is to remedy wrongful injuries, as the Restatement (Third) posits, then the justificatory force is supplied by the nature of the injury-causing conduct rather than the injury itself. Negligent behavior is wrongful-the actor unjustly injures another-and that wrong normatively grounds the duty to compensate. This rationale requires unjust or wrongful conduct, so the fact of injury does not provide a morally coherent justification for the duty to compensate.3 What, then, justifies the elements of strict liability in the tort system?
Strict liability could be justified in terms of the other rationale for tort liability adopted by the Restatement (Third)-the deterrence of unjust behavior-but the justificatory force then derives from the prevention of wrongful behavior rather than its consequences. Deterrence would seem to justify liability for wrongful behavior, whether or not that conduct caused injury. Tort liability, though, requires injury and not merely wrongful conduct.
The Restatement (Third) rationales for negligence liability therefore are hard to square with important tort doctrines governing strict liability and damages. Moreover, the two rationales may be incompatible with one another, providing yet another potential source of incoherence.4
The problem of coherence arises because a rationale for negligence liability applies to other tort doctrines. The justification for negligence liability in the Restatement (Third) proceeds from the compelling premise that one ought not engage in wrongful behavior and should be obligated to compensate others for the harmful consequences of such behavior. This justification depends on a conception of wrongful behavior. That conception is supplied by the negligence standard and has implications for all of tort law.
For purposes of negligence liability, tortious behavior is defined by the legal requirements of reasonable care or the negligence standard. The standard of care compares the conflicting interests of the parties involved in the injury-producing interaction. In the exercise of her liberty interest, the defendant engaged in risky behavior that threatened the physical security of others, including the plaintiff. This conflict between liberty and security interests is mediated by the negligence standard in a manner that defines wrongful conduct for purposes of negligence liability. The resultant conception of tortious conduct has consequences for all of tort law, because any tort rule mediates the conflicting interests of the parties. The coherence of tort law may require a negligence standard that mediates interests consistently with other tort doctrines, including the intentional torts and strict liability.
In mediating conflicting interests, both the intentional torts and strict liability give the security interests of potential victims legal priority over the competing liberty interests of potential injurers. Each form of tort liability involves a duty to compensate for physical injury, a duty that burdens the subordinate liberty interests to compensate any harms inflicted on the legally superior security interests. This priority of interests is not affected by the fundamental difference between the liberty interests involved in the intentional torts and those involved in the strict liability torts. Many intentional torts, such as assault, battery, and trespass, often involve criminal behavior.5 By contrast, the rule of strict liability for abnormally dangerous activities involves behavior that is not wrongful or unreasonable.6 Even though strictly liable behavior is not subject to moral approbation, such behavior nevertheless creates a duty to compensate, just as criminal behavior creates a duty to compensate for intentional harms. Both forms of behavior, whether morally problematic or not, create a duty to compensate, making the associated liberty interest legally subordinate to the security interest protected by the duty.
Consistency would seem to require that negligence liability reflect the same priority of interests, given that negligent behavior is either morally wrongful (like many intentional torts) or not (like strict liability torts). But if the security interests of potential victims have legal priority over the conflicting economic and liberty interests of potential injurers, the priority would seem to yield a tort system designed to compensate physical injuries. Such a system gives priority to the victim’s (harmed or threatened) security interest over the injurer’s liberty interest (burdened by the obligations of care and compensation). The compensation rationale for tort liability is much more plausible than it may seem.
In this Article, I argue that the compensation rationale coherently explains the important doctrines of tort law, including the significant limitations of liability. Part I begins by developing the tort compensatory norm, which is defined as the legal priority of one’s security interests over another’s liberty interests consistent with a thinly specified norm of equality. In light of that priority, the tort system must adequately protect physical security while allowing risky behavior. The priority implies that for any risky interaction, the potential “victim” is the party facing a threat to her physical security, whereas the potential “injurer” is the one whose exercise of liberty creates that threat. Ideally, the potential injurer would obtain the consent of the potential victim prior to the risky interaction. The potential victim would agree to assume the risk only if she were fully compensated for doing so. Consensual risks are ideal for compensatory purposes because the agreement ensures adequate protection of the security interest while allowing the risky behavior.
The ideal of consensual risky interactions is not ordinarily feasible. Consequently, tort law provides compensation for nonconsensual risks resulting in injury. As Part II shows, this nonideal mode of compensation yields a general preference for negligence liability. Tort damages do not compensate a dead person for the lost pleasures of living, so an individual exposed to a fatal risk is never assured of full compensation. To compensate for this limitation of the damages remedy, tort law provides additional protection to physical security by relying on a negligence standard that reduces risk below the level that would exist in the ideal, fully compensatory regime. The risk reduction benefits potential victims by making it less likely that they will be seriously injured or killed, providing a nonmonetary compensatory rationale for negligence liability consistent with the priority of security over liberty interests. Hence the compensation rationale can explain why tort awards typically are limited to negligently caused injuries.
Part III discusses a different practical problem posed by damages awards, a problem that explains why negligence liability must be limited for compensatory reasons. Imposing liability on a negligent defendant for all foreseeable harms would predictably bankrupt defendants in certain classes of cases, leaving the defendant unable to compensate fully the physically injured victims. To ensure that physical injuries are adequately compensated, in most cases tort law limits the duty of negligent defendants to exclude stand-alone emotional harms and economic loss. In some cases, these harms are compensable because the duty is formulated so that only a limited number of parties can make claims on the negligent defendant’s assets. By limiting the number of tort claimants, these duty rules substantially increase the likelihood that the negligent defendant ordinarily will have sufficient assets to compensate the physical injuries. Like negligence law in general, these particular limitations on negligence liability can also be justified in compensatory terms.
Part IV provides a compensatory rationale for the rule of strict liability for abnormally dangerous activities. The difficulty of enforcing the negligence standard, usually stemming from problems of proof, can make it incapable of producing the amount of risk reduction required for compensatory purposes. In these instances, negligence liability loses its nonmonetary compensatory advantage over strict liability, making strict liability preferable on grounds of both risk reduction and injury compensation. In other contexts, the rules of negligence and strict liability would each attain the same risk level. When the two liability rules cannot be distinguished in terms of deterrence, considerations of reciprocity-long important to the tort system-provide a compensatory rationale for limiting strict liability to nonreciprocal or abnormally dangerous risks. These two, limited compensatory roles for strict liability are fully reflected in the Restatement (Second)’s rule of strict liability for abnormally dangerous activities.7
Part V turns to the defenses of assumption of risk and contributory negligence, tort doctrines that mitigate or eliminate the liability of a negligent defendant. When the plaintiff assumes the risk, the act of knowing consent implies that exposure to the risky activity makes her better off than she otherwise would be. Liability is not required in these cases because the compensatory norm merely requires that the risky interaction not make the plaintiff worse off than she would otherwise be. In cases of contributory negligence, the plaintiff becomes an injurer for compensatory purposes. The compensatory norm merely requires that an injurer compensate the victim for the injury. Whether the plaintiff (as injurer), defendant (also an injurer), or both bear liability are all outcomes consistent with the compensation rationale.
Finally, Part VI addresses the question whether the compensation rationale adequately explains why the tort system adopted the general rule of negligence rather than strict liability. The modern tort system emerged in the nineteenth century. At the same time as courts were expressly adopting negligence liability, the case law exhibited widespread concern about protecting accident victims and controlling risk. On one view, the judicial concern for potential victims was merely an insincere attempt to mask the subsidization that a negligence regime afforded to business during the industrial revolution.8 The compensation rationale, by contrast, explains why judges could sincerely express a concern for protecting potential victims while adopting negligence as the general rule of liability. The period witnessed a staggering toll of accidental physical injuries and death, giving the superior risk-reducing capabilities of the negligence standard a decided compensatory advantage over the monetary damages afforded by strict liability. However, the compensatory rationale remained immanent until the Legal Realist movement in the twentieth century. The Realists developed negligence theory in a manner that more clearly revealed the compensatory rationale for tort liability, an outcome ultimately reflected in the Restatement (Second) position that compensation is the primary purpose of tort law.
The Article concludes by discussing how the compensation rationale is reflected in the current understanding of tort law. At first glance, it looks like the compensation rationale is rejected by the Restatement (Third), which justifies negligence liability in terms of fairness and deterrence. However, fairness and deterrence express the two important compensatory features of negligence liability. An important aspect of nonideal compensation involves the reduction of risk below the level attained in the ideal, fully compensatory regime. One’s failure to exercise reasonable care therefore is an appropriate locus of normative concern because the risk reduction was required to make the risky interaction adequately compensatory. The consequent injury, particularly in the case of death, cannot always be fully remedied by the tortfeasor’s payment of damages. Negligence liability accordingly involves wrongful behavior, as per the Restatement (Third). Thus negligence liability can be justified both in terms of fairness-it helps to remedy the consequences of wrongful behavior-and deterrence (or risk reduction), even if tort liability is based on a norm of compensation.
I. THE TORT NORM OF FULL COMPENSATION
The tort norm of compensation must be derived from the practice of tort law. Various compensatory norms could be imagined, but without a sufficient correspondence to important tort doctrines they could not properly be called tort norms. My development of the compensatory norm accordingly accepts the relevant tort practices without justifying them. A compensatory norm lacking any moral appeal, however, is unlikely to be compelling, no matter how consistent it may be with tort law. Hence I will derive the tort norm from the relevant tort practices while also indicating its apparent appeal. What ultimately justifies this compensatory norm, or indeed what justifies all of tort law, is not the issue under consideration.
Nevertheless, the analysis does have implications for the purpose of tort law. The compensatory norm is not consistent with all justifications for tort liability. A norm of allocative efficiency, for example, justifies tort liability as a means of minimizing costs. Some corrective-justice norms, on the other hand, justify tort liability as a remedy for injuries wrongfully caused by another. These norms are not compensatory for the simple reason that they limit tort compensation to inefficient or wrongful injuries. For a compensatory norm, by contrast, the fact of injury or harm can be an appropriate basis for tort liability, even if the defendant did not act inefficiently or wrongly. Consequently, the compensatory norm implies some underlying rationale for tort law that differs from rationales based on allocative efficiency or wrongful conduct.
Although the compensatory norm justifies more compensation than the norm of allocative efficiency and norms requiring wrongful conduct, the compensatory norm does not justify total compensation of all injuries. A norm of total compensation has been posited by others.9 Such a tort norm is easily rejected due to its apparent inconsistency with tort law.10
Even in its idealized form, the tort compensatory norm does not require compensation of all injuries. A well-specified norm also shows why full compensation is not feasible. In essence, compensation must be limited due to the practical problem of implementing a compensatory norm in a world of scarce resources. The limitation stems from the need to compensate injuries, which is necessarily imperfect as compared to the ideal form of consensual compensation prior to the risky interaction.
A. THE TORT IDEAL OF FULL COMPENSATION
Initially the tort norm of full compensation can be developed as an ideal, an objective that may not be practically attainable but is worth pursuing.11 What might that ideal look like?
To answer this question, we must first determine what injuries require compensation. What makes someone a “victim” for compensatory purposes? Relatedly, what makes someone an “injurer”? Having made these determinations, other questions then arise. Do the categories of injurer and victim imply any limitations on compensation? And for those injuries governed by the compensatory norm, what is the ideal mode of compensation?
As this line of inquiry suggests, a compensatory norm requires much more elaboration than others have recognized.
1. The Categories of “Victim” and “Injurer”
The concept of compensation presupposes injury. An automobile driver may accidentally injure a pedestrian. A tort rule that obligates the driver to compensate the pedestrian merely shifts the loss from one party to the other. Only one party can be compensated. But which one? The answer cannot be supplied by the concept of compensation.
This indeterminacy in the compensatory norm reflects the Coasean conception of risky interactions as involving two parties, neither of which necessarily falls into the category of injurer or victim.12 Two parties have interacted, the interaction caused injury to one party, and shifting the loss to the other merely makes that party the accident victim. Without further development, the categories of injurer and victim are unhelpful. By implication, a norm of compensating “victims” is incomplete.
A compensatory norm necessarily depends on a defensible conception of injurer and victim. Tort law has based that conception on the respective individual interests at stake. Frequently, the individuals in a risky interaction have different types of interests that conflict, a difference that can be used to define the injurer and victim. The automobile driver typically desires the transportation to pursue various economic or liberty interests; the pedestrian also transports herself to pursue such interests. In the event that the driver physically injures the pedestrian, by definition the pedestrian’s interest in physical security has been harmed. The pedestrian also suffers emotional harms (pain and suffering) and intangible economic harm (like medical expenses). If the driver is obligated to compensate those harms, the monetary damages would be detrimental to her economic interests. Any precautionary obligations tort law imposes on the driver would also be detrimental to her liberty interests. The way in which tort law regulates the risky interaction therefore means that at least one party’s interests will be burdened or harmed: either the pedestrian’s interests in physical, emotional, and economic security; or the driver’s liberty interests, including the economic interest. Insofar as the conflicting interests of the two parties differ, it becomes possible to define the concepts of injurer and victim in terms of those interests.
Tort law traditionally has distinguished between liberty and security interests, giving “peculiar importance” to the nature of the interests and their social value.13 Numerous moral philosophers also distinguish liberty and security interests.14 Perhaps surprisingly, the difference between physical security and economic interests is recognized by economic analysis.15
For economic purposes, the risky interaction between the driver and pedestrian is modeled as a transaction. As in any transaction, there must be a buyer and seller. Must the pedestrian purchase the protection of her physical security from the driver, or must the driver purchase from the pedestrian the right to impose the risk? The answer implies a legal priority of the respective interests, which then can be used to define the injurer and victim.
Consider a tort rule that gives the entitlement to the pedestrian, making her the seller. By forcing the driver to purchase the right to impose the risk, the rule burdens her economic interest. By requiring that the pedestrian be paid for facing the risk, the rule prioritizes her interest in physical security. That priority accordingly defines the categories of injurer and victim for compensatory purposes, as the seller (pedestrian or potential victim) is the party receiving compensation from the buyer (driver or potential injurer).
Economic analysis cannot determine whether the driver or pedestrian should hold the initial entitlement.16 Instead, normative justification determines initial entitlements, which then determine the categories of buyer (injurer) and seller (victim) for purposes of economic analysis.
The tort norm of compensation gives the entitlement to the pedestrian because tort law gives priority to the security interest over other interests. According to a leading torts treatise, “the law has always placed a higher value upon human safety than upon mere rights in property.”17 Other laws regulating health and safety embody the same priority,18 which has also been accepted by many philosophers and economists.19 One rationale for this priority is that liberty and security are important to individual flourishing, but physical security is of fundamental importance.20 A related rationale for the priority is that physical injury is more disruptive for individuals than the expenditure of money. Insofar as fairness requires that risky interactions should be structured to minimize the burden faced by the most disadvantaged party, the security interest receives priority.21
This reasoning merely identifies the victim for purposes of compensation; the category of injurer is still undefined. Even if the driver’s car did hit the pedestrian and was a physical cause of the injury, other causes can be identified. The state, by permitting driving, could also be considered a “cause” of the accident, making it a candidate to be the injurer for compensatory purposes. The resultant conception of the state as injurer would lead to a social insurance fund. Indeed, the parents of the driver are also physical causes of the accident. Of the various causes of an injury, which should become obligated to compensate as an injurer?
Just as the security interest defines the category of “victim,” liberty interests define the category of “injurer” for purposes of tort law. A tort obligation to compensate the plaintiff’s injuries depends on the existence of duty, which ordinarily requires that the defendant’s actions created foreseeable risks to the plaintiff.22 These duty requirements define those aspects of liberty that make someone an injurer for tort purposes. By definition, the exercise of these liberty interests foreseeably threaten the security interests of another, obligating the actor or potential injurer to take care towards the other.23 If a breach of that precautionary duty causes physical harm, the defendant-injurer is obligated to compensate the plaintiff-victim.24
Defining the injurer in terms of the liberty interest does not imply that any exercise of liberty necessarily turns the victim into an injurer as well. Had the pedestrian stayed at home, she would not have been run over by the driver. Nevertheless, this exercise of the pedestrian’s liberty ordinarily does not make her an injurer for compensatory purposes.
The compensatory norm does not give the pedestrian’s security interest legal priority over her own liberty interest. If the tort norm were to require an intrapersonal priority of interests, it would govern all human conduct and not be limited to risky interactions. The norm, in other words, would cease to be compensatory and instead become the guide for how one must engage risk under any circumstances. Hence a compensatory tort norm necessarily is limited to an interpersonal priority of security and liberty interests. That priority creates a compensatory obligation for the driver as an injurer. Although the pedestrian is also a cause of the accident, no such priority necessarily attaches to her security and liberty interests, ordinarily eliminating her as an injurer for compensatory purposes.25
2. Compensatory Implications of the Victim and Injurer Categories
By defining the injurer in terms of certain liberty interests, tort law has adopted a conception of compensation that is limited in important respects. The requirement that the injury be caused foreseeably by some action of the defendant reflects the conception of tort law as a legal solution to the normative problem posed by risky interactions. That conception has compensatory implications. It means that tort law does not purport to compensate all injuries, such as those caused by natural disasters. And a tort norm of compensation for injuries caused by individuals’ interactions rules out compensation for injuries caused by a failure to rescue. An individual who fails to act (nonfeasance) may have been able to save another from injury, but the harm stems from the lack of interaction. The absence of tort liability for such injuries implies that any tort norm of compensation is limited to risky interactions.26
Once the tort norm is conceptualized in terms of risky interactions, compensation becomes well defined. Risky interactions involve conflicting individual interests, and the qualitative difference in those interests defines the categories of injurer and victim. Any given physical injury can be causally connected to many individuals, but the requirements of feasance and foreseeability define an injurer as one who knew or should have known that the exercise of her liberty would expose others to an identified risk of injury.27 The potential victim is someone like the pedestrian whose physical security is threatened. Further prioritization of interests is possible. The individual interest in physical security can, for example, be distinguished from, and given priority over, the interest in emotional tranquility.28
This conception of compensation implies yet another limit. Indistinguishable interests cannot be governed by the compensatory norm, because a distinction of interests underlies the categories of injurer and victim. In cases of self-defense, for example, the security interest of the defender conflicts with the security interest of another. Under established tort principles, the reasonable use of force in self-defense does not create liability for any resultant bodily injuries, even if suffered by an innocent bystander.29 The absence of liability stems from the absence of a normative difference between the two sets of security interests. One set of security interests has been harmed; the victim can be identified. But what distinguishes the victim’s security interest from the other’s? Causation does not suffice.30 Without any normative distinction between the conflicting security interests, the injurer cannot be defined. In these circumstances, tort liability is not justified by the compensatory norm as the liability would merely shift the loss from one victim while creating another victim.
Similarly, the interaction of indistinguishable economic interests in market competition creates “winners” and “losers” rather than “injurers” and “victims.” A tort norm of compensation is no more relevant for contexts involving winners and losers than it is for contexts involving only victims. To become compensatory, tort liability must shift the loss between distinguishable interests captured by the categories of injurer and victim.
Whereas a simplistic interpretation of compensation suggests that tort damages should be available for all injuries, closer analysis shows otherwise. A compensatory norm depends upon the concepts of injurer and victim, and those concepts imply limitations on the reach of compensation. Even in its ideal form, the tort norm of compensation would not justify damage awards for all injury victims.
3. The Mode of Compensation
To complete the tort norm of full compensation, we must consider the ideal mode of compensation. The potential victim could be compensated either for risk or injury, each of which reduces individual welfare. In addition, the mode of compensation could depend on consent or the neutral determination of a third party, such as a court.
The tort compensatory norm most plausibly involves the compensation of injuries rather than risk. Physical injuries harm the security interest. Facing the risk of physical injury is upsetting and also detrimental to individuals, but the interest in emotional tranquility differs from the interest in physical security. Compensation defined in terms of the security interest thus indicates that the compensation is for injury rather than risk.
A norm of injury compensation does not render risk irrelevant. Tort damages for injuries depend on risk. Compensatory tort damages cannot be determined by asking how much money the victim would require to accept the certainty of suffering the injury in question, implying that tort compensation for injuries somehow depends on risk.31
A norm of injury compensation depends on risk when the norm applies to premature death. Prior to a fatal accident, ordinarily the potential victim would view her premature death as a grievous loss. From that perspective, a norm of injury compensation should apply to fatal injuries. By contrast, if the norm depends on the victim’s perspective at the time of injury, there may be no loss requiring compensation. Any loss attributable to death must be understood from the perspective of the living, so a dead person may suffer no injury requiring compensation.32 A tort compensatory norm encompassing premature death therefore must adopt the perspective of the potential victim prior to the fatal accident. From that perspective, the fatal injury is only a possibility rather than a certainty (otherwise the potential victim would avoid the interaction and there never would be any accidental deaths requiring compensation). The compensation of premature death accordingly depends on the contingency of the fatal accident, explaining why tort compensation for injuries depends on risk.
The tort compensatory norm clearly applies to premature death. The compensatory norm is defined by the priority of the security interest, and surely the most important aspect of physical security involves protection from premature death. Not surprisingly, such protection is provided by tort law.33 To explain tort law, the compensatory norm must apply to premature death. What, then, is the best way to compensate physical injuries, including fatalities?
Of the various possibilities, the tort ideal involves the potential victim’s well-informed consent prior to the risk exposure. According to long-established tort doctrine, “no one suffers a legal wrong as the result of an act to which, unaffected by fraud, mistake or duress, he freely consents or to which he manifests apparent consent.”34 An individual presumably consents to the risk exposure to derive some benefit. That benefit is adequately compensatory for tort purposes, explaining why consent eliminates the tort duty that one otherwise would owe another.35
Compensation based on well-informed consent prior to the risk exposure has both normative and economic appeal. Tort law is private law, making it appropriate for the parties to resolve privately the compensatory issue by well-informed mutual consent.36 Such consent promotes individual autonomy and resolves the issue of responsibility in a normatively defensible manner: Rights and duties do not depend on the arbitrary consequences of whether the risk materializes in injury.37 A consensual transaction of this type also would be efficient.38
To see why well-informed consent prior to the risky interaction satisfies the compensatory ideal, consider an exchange between the driver and pedestrian. For the parties to be well informed, each must know about the risk in question. To simplify, suppose each party knows that the interaction poses a given probability that the pedestrian will suffer an identified physical injury. By definition, the compensation norm merely requires that the risky interaction not make the pedestrian worse off than would otherwise be the case. That amount of compensation can be precisely defined. For a given probability of suffering an identified injury, the pedestrian generally would be willing to accept some amount of money to face the risk. This amount of money will be called the “Willing-to-Accept” (WTA) risk proceeds. That amount makes the pedestrian indifferent between (1) the state of the world in which the pedestrian does not face the risk and consequently receives no money, and (2) the state of the world in which the pedestrian does face the risk and receives the WTA risk proceeds. For example, if the interaction were certain to kill the pedestrian, then she would not accept any amount of money and would not face the risk: The WTA measure equals infinity. For smaller risks, the WTA measure is a finite number that depends on the probability and severity of injury.39 In effect, the WTA measure is the monetary equivalent of the benefit the pedestrian must receive before assuming the risk.
By ensuring that the pedestrian is not made worse off by the risky interaction, the consensual exchange of the WTA risk proceeds satisfies the compensatory norm. Consent promotes the pedestrian’s autonomy and satisfies the demands of compensation in a normatively defensible and efficient manner, while enabling the driver to engage in the risky behavior.
B. NONIDEAL IMPLEMENTATION OF THE COMPENSATORY NORM
Defining the tort norm of full compensation in ideal terms immediately reveals its impracticality. Rarely will drivers be able to gain the consent of all those who would be exposed to the risk of driving. Without such consent, how might the tort system proceed? How should the ideal of compensation be implemented under nonideal conditions?
An extreme solution would stick to the ideal. All nonconsensual risky interactions could be banned, which would have the practical effect of banning a wide range of social behavior. Transportation, necessary for the functioning of any community, has been responsible for large numbers of nonconsensual physical injuries, from the railroads of the nineteenth century to the automobile today. Banning these forms of transportation and other types of nonconsensual risks would be extraordinarily disruptive for society and all its members.
Tort law, of course, has rejected this extreme solution. The mere imposition of a nonconsensual risk on another does not violate tort law. The question, then, is whether the important tort practice of permitting some nonconsensual risks is consistent with a compensatory norm.
A legal ban of nonconsensual risks would be hard to square with any defensible compensatory norm. Presumably such a norm regulates risky interactions in a manner that does not unfairly disadvantage potential victims. An individual is a “potential victim” when the individual’s security interests have legal priority over the liberty or economic interests of the “potential injurer.” The priority merely reflects the relative importance of these conflicting interests for purposes of compensation. Giving physical security priority for compensatory purposes does not deny the fundamental importance of liberty for each person. Indeed, the priority required for compensatory purposes cannot ignore the subordinate economic or liberty interests, both as a matter of equality and individual welfare. The subordinate liberty interests would be effectively negated by tort rules banning nonconsensual risks, contrary to the normative requirement that a legal rule adequately respect the interests of both the right-holder and the duty-holder.40 Moreover, the welfare of each person would be reduced by a legal ban of nonconsensual risky interactions because such a ban would severely disrupt the community. Unthinkingly applied, a compensatory norm could harm everyone, whether denominated “potential victims” or otherwise. Adequate concern for the interests of potential injurers and potential victims accordingly requires tort rules that permit nonconsensual risks.41
However, the tort system need not permit all types of nonconsensual risks. The ideal of full compensation described above is not a fully specified justification for tort liability. Instead, the compensatory norm is assumed to be an implication of some normative theory of tort law. Such a normative theory could proscribe some risky interactions because they involve normatively problematic behavior, like a driver who speeds for the thrill of scaring others.42 By prohibiting such behavior or requiring consent, tort law would protect potential victims from such harms consistent with the underlying justification for tort law and the norm of compensation implied by that justification. But since most nonconsensual risky interactions further normatively defensible interests, the tort system as a general rule permits such interactions.
Tort law regulates nonconsensual risky interactions consistently with the ideal of compensation by relying on injury compensation as a substitute for the ideal compensatory exchange. Ideally the pedestrian consents to face the risk and in exchange receives the WTA risk proceeds. Ordinarily the exchange is not feasible, requiring tort rules that permit the driver to create the risk despite the lack of consent. Even though the driver and pedestrian have not transacted, tort rules could award the pedestrian the WTA risk proceeds in the form of damages for injuries. Such compensation is afforded by a damages award equal to the pedestrian’s WTA measure divided by the underlying probability of injury.
To see why, suppose the individual were repeatedly exposed to a risk of 1 in 10,000 and has a WTA measure of $5. For each risk exposure, the individual is entitled to the WTA risk proceeds of $5 as full compensation. Over the course of 10,000 risk exposures, the individual is entitled to WTA payments totaling $50,000. Over the course of 10,000 risk exposures, the individual is likely to be injured once. Rather than receiving a $5 payment prior to each of the 10,000 risk exposures, upon injury the individual could collect in a tort suit the entire WTA risk proceeds of $50,000 as damages or compensation for the injury. Tort damages equal to the WTA risk measure of $5 divided by the probability of injury (or multiplied by 10,000) yields tort damages of $50,000. Properly formulated, tort damages provide a method for giving individuals the WTA risk proceeds to which they are entitled.43
This method for computing tort damages satisfies the requirements of the case law.44 Such tort damages plausibly satisfy the norm of full compensation, because giving potential victims a guarantee of such tort damages is functionally equivalent to the exchange in which potential victims receive their WTA risk proceeds from potential injurers prior to the risk exposure.
Although the two forms of compensation are functionally equivalent, tort damages do not perfectly substitute for the WTA risk proceeds. In an actual exchange, the pedestrian receives the money while healthy; in the damages exchange, the pedestrian receives the money while injured. An individual’s ability to derive satisfaction from money can depend on health. Dead persons, after all, derive no satisfaction from a damages award, even if their heirs do.
Consequently, implementing the compensatory ideal via damages awards necessarily leads to nonideal outcomes: Monetary damages for injury do not perfectly substitute for the fully compensatory monetary payment prior to the risky interaction. This compensatory limitation does not mean the tort system rejects the objective of full compensation. The ideal is not attainable, and damages for injury may be the most feasible compensatory alternative.
For these reasons, the tort system can adhere to a norm of full compensation, even though tort compensation is not perfectly compensatory. The compensatory limitations of tort law stem from the reality of a world of scarce resources. Absent transaction costs, drivers and pedestrians could always transact in a manner that satisfies the compensatory ideal. The inability to engage in such transactions makes nonconsensual risks inevitable. Instead of compensating perfectly for consensual risks, tort rules compensate imperfectly for injuries. To be persuasive, however, this compensatory rationale must also explain other tort practices that are less than fully compensatory for accident victims.
II. NEGLIGENCE AS NONIDEAL COMPENSATION
In noncontractual settings, “accident law is still on the whole the law of negligence.”45 A negligence rule monetarily compensates accident victims only for injuries caused by unreasonable (negligent) risks. Giving potential victims a guarantee of tort damages for injuries caused by unreasonable risks is not functionally equivalent to an exchange in which potential victims are compensated for facing both reasonable and unreasonable risks. The tort system’s general reliance on negligence liability seems to cripple the claim that the primary purpose of tort law is to compensate physical injuries. Instead, the compensatory norm would seem to justify strict liability, which provides accident victims with a damages remedy for injuries caused by both reasonable and unreasonable risks.
The compensatory properties of negligence and strict liability require further development, however. Someone killed by a tortious risk typically does not get tort damages for the lost value of life, popularly known as “hedonic damages.”46 In the vast majority of states, damages in wrongful death actions are measured by the pecuniary loss rule, which limits damages to the monetary value of the benefits the decedent could have been expected to contribute to her survivors (or estate in some jurisdictions) had she lived.47 This rule means that individuals who are killed by tortious risks receive no compensation for the loss of life’s pleasures. What the individual loses by premature death-the benefits, joys, and pleasures of continued life-is not compensated by tort damages. The lack of compensation occurs under both strict liability and negligence.
The lack of damages compensation for the loss of life’s pleasures explains the otherwise anomalous tort awards in personal injury cases. For example, during the period 1984-93, the average jury verdict in New York City in a wrongful death suit ranged from $1.2 million to $1.9 million, depending on the borough.48 Verdicts for cases in which the most serious injury involved brain damage were significantly higher, averaging from $3.6 million to $6.2 million.49 The highest verdicts for cases in which the most serious injury involved herniated discs, knee injuries, or leg injuries were typically more than double the average award in wrongful death cases.50
The most serious knee or leg injury is not twice as injurious as the average case of premature death. Yet these tort awards make sense insofar as money cannot compensate a dead person for the lost pleasures of life. Giving damages to a dead person is not functionally equivalent to giving her the WTA risk proceeds prior to the risky interaction. But if potential victims do not receive tort damages for the loss of life’s pleasures, then they do not receive their full WTA risk proceeds for nonconsensual, fatal risks. These risks pose the greatest threat to the security interests of potential victims, who, however, are then not fully compensated for them. The problem arises under both strict liability and negligence. The question, then, is whether one of these rules is more capable of ameliorating that problem. That rule could be justified by the compensatory norm.
To address this issue, we must consider in greater detail the idealized compensatory exchange between the driver and pedestrian. Suppose the driver unilaterally imposes a risk on the pedestrian and is the only party capable of reducing that risk by taking precautions. (Contexts involving bilateral risks and the possibility of victim care will be examined later.) For each precaution, there is a cost or burden B that would be incurred by the driver. If the driver does not take the precaution and instead creates the risk, she owes the associated WTA amount to the pedestrian. The driver would not incur the burden B if it would be less expensive to pay the WTA amount to the pedestrian. By definition, the pedestrian would agree to face the risk in exchange for the WTA risk proceeds. Hence the two parties would agree that in exchange for the WTA proceeds, the driver can impose risks whenever the cost of eliminating the risk B exceeds the safety benefit expressed by the WTA measure: B > WTA. The parties also would agree that the driver must take precautions satisfying the cost-benefit test (precautions for which B
The idealized compensatory agreement seems to correspond to the outcome that occurs under strict liability. A strictly liable driver rationally takes only those precautions satisfying the cost-benefit test, as it would be less costly to pay damages for the residual risks that produce injury. The apparent correspondence between the idealized compensatory agreement and strict liability explains why the compensatory norm seems to justify strict liability. The compensatory advantage of strict liability, however, may be more apparent than real. Tort damages do not fully compensate potential victims for nonconsensual risks threatening death and serious physical injury. That compensatory gap under strict liability creates a compensatory role for negligence liability, making it possible to justify a negligence regime in compensatory terms.
In a regime of strict liability, the potential injurer can choose the standard of care, resulting in precautions that satisfy the cost-benefit test. A negligence regime, by contrast, establishes the standard of care. The standard of care has important compensatory implications.
If, for example, the standard of care is defined in cost-benefit terms, it would require the same safety precautions as would arise in a regime of strict liability. In these circumstances, strict liability would be preferable for compensatory purposes because it provides tort damages for injuries caused by both reasonable and unreasonable risks, whereas negligence liability provides tort damages only for injuries caused by unreasonable risks. The negligence regime, however, need not rely upon such a standard of care.
By requiring more precautions than would be justified by a cost-benefit test, a negligence standard would reduce risk below the level that occurs under strict liability.52 The risk reduction benefits potential victims by making it less likely that they will be seriously injured or killed. That important safety advantage plausibly makes negligence liability a better compensatory rule for potential victims, even though strict liability provides a damages remedy in more instances. Monetary damages for injury cannot fully restore the accident victim’s physical security, whereas risk reduction directly protects the security interest.
The additional safety expenditures required by such a negligence standard need not be unfair to potential injurers. The standard of care can be set so that potential injurers incur the same total burdens they would otherwise bear in a perfectly compensatory regime. A potential injurer’s total burden in a perfectly compensatory regime can be determined from the idealized compensatory exchange prior to the risky interaction. That burden can then be compared to the lower burden incurred by the potential injurer in the nonideal damages regime, which effectively permits the potential injurer to avoid paying potential victims the full WTA risk proceeds for fatal risks. The difference between the ideal and nonideal burdens-the amount of money that the potential injurer avoids paying for fatal risks in a damages regime-can be added to the precautionary requirements otherwise faced by the potential injurer in a perfectly compensatory regime (that is, requirements satisfying the cost-benefit test for reasons given above). The net result is a well-defined negligence standard requiring precautions above the cost-benefit amount that leaves the potential injurer no worse off than she would be in a perfectly compensatory regime.53
The precautions required above the cost-benefit amount are attributable to the expenditure of the compensatory funds-the WTA risk proceeds-on risk reduction. That risk reduction provides potential victims with an important form of nonmonetary compensation that is unavailable in a regime of strict liability with compensatory damages. The greater risk-reducing capabilities of negligence liability therefore can explain why a compensatory tort regime would generally rely on negligence liability as a nonideal solution to the compensatory problems posed by serious physical injuries or death.
This compensatory rationale for negligence liability is consistent with the negligence standard actually used by the tort system. If the negligence standard is altered for compensatory reasons, then the applicable standard of care should change as the compensatory problem changes across contexts. Such a variable standard of care is consistent with tort practice.
The negligence standard ordinarily is defined in terms of how a reasonably prudent person would have acted under the circumstances confronted by the defendant.54 A survey of 100 judges found that most applied the negligence standard in a more exacting manner when the conduct threatened serious bodily injury rather than property damage, even though the cases were otherwise identical in cost-benefit terms.55 Lay individuals and jurors tend to emphasize safety to an even greater degree, requiring more than the cost-benefit amount of safety for nonconsensual risks threatening serious bodily injury.56 Safety requirements in excess of the cost-benefit amount serve a compensatory role for serious bodily injuries. The compensatory problem is less severe for property damage, explaining why judges and jurors resort to a lower standard of care in cases involving such harm.57
A compensatory rationale for negligence liability also explains the centrality of the jury in applying the open-ended negligence standard. Even if no fact is in doubt, the jury decides on the reasonableness of the defendant’s conduct.58 The jury therefore makes some sort of normative decision in negligence cases. The nature of that decision can be clearly defined in compensatory terms.
Although the negligence standard can be precisely defined to ensure that potential injurers incur the same total burdens they would face in a perfectly compensatory regime, the standard nevertheless is not perfectly compensatory for potential victims. Some individuals will be injured by the nonconsensual, reasonable risks permitted by the negligence standard, an outcome without full tort compensation. The amount of undercompensation depends on the amount of nonconsensual risk permitted by the tort system. How much undercompensation is fair-that is, how much nonconsensual risk should be permitted by the negligence standard-is a distributive problem between potential injurers and potential victims. Once viewed as a distributive problem, it becomes apparent why the problem is appropriately resolved by members of the community on a case-by-case basis, the type of resolution effectuated by current tort practice involving the negligence standard. Even if the facts are not in dispute, the distributive problem requires resolution by the jury.59
Consistently with the compensatory rationale for negligence liability, tort law alters the negligence standard for product cases, relying on a cost-benefit test rather than the reasonable-person standard.60 The change reflects a difference in the compensatory problem. In the vast majority of product cases, the potential victim can be conceptualized as a consumer who both pays for and benefits from safety investments or guarantees of injury compensation.61 The costs and benefits of tort liability are largely internalized by the consumer, and the cost-benefit (or risk-utility) test maximizes the welfare of consumers (potential victims).62 Defining compensation in terms of both the security and economic interests of the potential victim is appropriate in these circumstances.63 In the driver-pedestrian interaction, by contrast, the costs and benefits of tort liability are not internalized by potential victims (pedestrians). Compensation of potential victims in such cases requires consideration of the security interest alone, resulting in the more exacting reasonable-person standard of care.
The compensatory norm therefore is consistent with a general rule of negligence and a standard of care that varies across cases. The compensatory ideal could be satisfied by consensual assumption of risk in exchange for monetary compensation, an outcome that readily shows why the ideal is not attainable. The nonideal solution of relying on monetary damages for injuries creates a compensatory problem for serious physical injuries and death. To ameliorate that problem, tort law tries to reduce risk below the level that would exist in a fully compensatory regime. That risk reduction can be attained only by an appropriately formulated negligence standard. Tort law apparently alters that standard in the appropriately compensatory fashion across the range of cases.64
III. COMPENSATION AND DUTY LIMITS ON NEGLIGENCE LIABILITY
Historically, the duty of reasonable care for purposes of negligence liability did not encompass risks threatening only emotional distress or economic loss. This omission absolved a negligent injurer of the obligation to compensate those foreseeable harms. Today potential injurers face a limited duty regarding such risks, but most stand-alone emotional and economic injuries remain uncompensated. These limits on compensation are consistent with the compensation rationale for tort liability. In a world of scarce resources, negligent defendants usually will not have the assets to pay for all injuries they foreseeably cause. Unless liability is limited, most negligent defendants will be unable to compensate fully the physically injured victims. Compensation of the interest in physical security therefore requires tort rules limiting the compensation of other injuries.65
A. EMOTIONAL DISTRESS
Consider again the car driver who negligently hits a pedestrian, and now consider the bystanders and family members who suffer emotional distress as a result. These third parties are foreseeable victims of the negligence, yet the defendant may have no duty to compensate their stand-alone emotional harms.
In the past, courts denied recovery for stand-alone emotional harms “unless the defendant’s act [was not merely negligent but] amounted to some other tort such as libel or slander.”66 This bar to recovery has eroded over time, and today most courts allow recovery for negligently inflicted emotional harm.67 Such recovery, though, is limited in various ways, typically by the requirement that the plaintiff must have been within the “zone of danger.”68 The limitations usually take the form of limited duty rules: The negligent defendant is not obligated to compensate the injury because she had no duty with respect to that type of harm.69
These limitations on duty seem arbitrary, and their judicial rationales do not clarify matters. According to courts, the reasons for limiting duty include the difficulty of determining monetary damages for purely emotional, nonmonetary harms; the difficulty of assessing the severity of emotional injuries; the difficulty of knowing whether monetary damages meaningfully compensate emotional injuries; and the difficulty of limiting the number of emotional distress claims that can be brought as the result of a single tort.70
The concerns regarding the computation and compensatory nature of emotional distress damages are hard to defend given that such damages are available in the ordinary negligence case involving physical injury.71 The separate concern about fraud is valid, due to the possibility that stand-alone emotional distress claims are more easily faked than other claims. That concern, though, does not require a limitation of duty, as fraudulent claims could be addressed by more stringent standards of proof.72 Indeed, these concerns “have been answered many times, and it is threshing old straw to deal with them.”73 The only valid concern for limiting duty in these cases pertains to the floodgates problem of unlimited liability.
This justification for limiting duty, which by now has become conventional, emphasizes a concern about unfairness for defendants and similarly situated potential injurers:
It would be an entirely unreasonable burden on all human activity if the defendant who has endangered one person were to be compelled to pay for the lacerated feelings of every other person disturbed by reason of it, including every bystander shocked at an accident, and every distant relative of the person injured, as well as all his friends.74
Justified in this manner, the limit on duty is hard to defend. The conventional justification implies that a negligent defendant’s interests have legal priority over the emotional interests foreseeably harmed by her conduct, thereby indicating that tort law devalues emotional interests.75 The conventional justification is undermined by other problems as well.
Presumably, the added compensatory burdens for stand-alone emotional harms are “entirely unreasonable” because the total liability of the negligent defendant would then be disproportionate to the wrongdoing. However, proportionality does not provide a compelling justification for limiting duty based on the type of harm.
Proportionality certainly matters for purposes of retribution-the punishment should “fit” the crime-but the tort obligation in question is one of compensation. The defendant’s negligence foreseeably caused the emotional harm, and the damages merely compensate the plaintiff’s injury. As a matter of compensation, the tort obligation is not disproportionate.
Indeed, if the total compensatory obligations of a negligent defendant can be limited out of concern for proportional fairness to the defendant, then a mitigation defense should be available to those unlucky defendants whose negligence consisted of a slight inadvertence that caused severe physical injuries to another. Such a defendant could face ruinous liability for an act that does not even merit personal reproach.76 Despite the apparent disproportionate liability in such cases, tort law does not reduce the defendant’s liability, presumably because the damages merely compensate the plaintiff’s physical injuries. Tort law’s failure to provide a mitigation defense indicates that duty is not limited out of some concern that the obligation to pay compensatory damages would be disproportionately burdensome for a negligent defendant.
Tort law also burdens defendants with precautionary obligations. Expanding duty to include stand-alone emotional harms could result in precautionary obligations that would be unfairly burdensome for defendants and similarly situated potential injurers, because duty determines the risks encompassed by the standard of reasonable care.77 But this problem could be overcome by altering the standard of care to yield precautionary obligations that would be fair to potential injurers in light of the expanded duty.78 Consequently, any unfairness associated with precautionary obligations does not justify a limitation of duty based on the type of harm.
Contrary to the conventional understanding of this issue, proportionality does not persuasively justify the limitation of duty. Moreover, the conventional justification is based on an erroneous assumption. By justifying the duty limitation in terms of the negligent defendant’s interests, the conventional rationale assumes that tort law gives a negligent defendant’s interests legal priority over the emotional interests foreseeably harmed by such conduct. That assumption is inconsistent with tort law. The obligation to compensate emotional harms in the typical case of physical injury implies that the negligent defendant’s interests are legally subordinate to the emotional interests foreseeably harmed by the negligence. The priority of emotional interests does not depend on whether the individual suffered physical injury as well. Tort law allows recovery for at least some stand-alone emotional harms. The compensation of such harms reflects the legal priority of the plaintiff’s emotional interests over the interests of a negligent defendant, and that priority does not depend on physical injury. As a matter of tort law, the negligent defendant’s interests are subordinate.
But if the emotional interests of a negligently harmed victim have priority over the interests of the negligent injurer, why would recovery for stand-alone emotional harms ever be limited?79 Any limit on recovery for those who suffer stand-alone emotional harms must mean that the associated emotional interests of foreseeable victims are subordinate to some set of interests with legal priority. The interests most likely to satisfy this condition are the security interests of those individuals who were also harmed by the negligence. In comparing the emotional interests of distressed individuals with the security interest of a physically injured victim, the security interest would seem to have priority as the core concern of tort law.80 Consequently, the security interest could provide a defensible basis for limiting the duty of a negligent defendant with respect to stand-alone emotional harms.
To determine whether compensation of the security interest justifies such a limitation of duty, consider the duty that would exist if negligent drivers were liable for all foreseeable emotional injuries. Suppose the risk threatens a total of n individuals: the pedestrian who faces the risk of bodily injury and (n – 1) foreseeable victims such as family, friends and bystanders who would suffer emotional distress in the event that the pedestrian was physically injured.
The duty encompassing stand-alone emotional harms, like any tort duty, does not give priority to any individual harmed by a breach of the duty. All n individuals can seek full compensation for their injuries. Frequently, the sum total of these judgments will exceed the assets and insurance of the defendant. Each plaintiff, including the physically injured victim, will get less than full recovery.
Unless the number of claimants arising from any given negligent act is limited, bankruptcy becomes a concern. Just as bankruptcy law prioritizes among the claimants to an insufficient pool of assets, so must tort law. That priority is accomplished by a limitation of duty. Tort law gives highest priority to physical security, implying that any concern about bankruptcy should be resolved in favor of physically injured victims. Tort law accordingly limits duty to exclude stand-alone claims for emotional harm to give physically injured plaintiffs a much higher chance of receiving full compensation for their injuries.81
This reasoning is reflected in the prevailing conception of duty, which deems duty to be important only when a negligent defendant proximately caused injury to the plaintiff.82 To have acted negligently, the defendant must have breached an established duty of care-the duty encompassing risks that threaten physical injury. That breach makes it possible to ask whether the negligent defendant’s duty should be expanded to include stand-alone emotional harms.83 The duty question regarding emotional harms therefore arises in the category of cases involving an unreasonable risk of physical injury.84 As a categorical matter, the duty question must consider how the more expansive duty and increased number of claimants would affect compensation of the security interests also threatened by the defendant’s negligent conduct. This inquiry, which is required by the compensatory norm, is a direct implication of the prevailing conception of duty.
Hence limitations of duty for stand-alone emotional harms are plausibly understood in compensatory terms, explaining why these limits seem so arbitrary (as with the “zone of danger” rule). The limitations serve the purpose of limiting duty to a group of claimants (defined categorically) small enough to be capable of receiving full compensation from the defendant (again defined categorically) in most cases. How the line is drawn will inevitably appear unfair in some cases. A general concern about bankruptcy need not be present in a particular case involving a wealthy defendant. The seemingly arbitrary result in a particular case does not undermine the compensatory rationale for limitations of liability that must be generally applied to categories of cases.
B. ECONOMIC LOSS
Consider again the negligent driver who physically injures a pedestrian. Suppose the injured pedestrian is a sole proprietor who employs five individuals, and those five workers lose their jobs because the injured proprietor can no longer run the business. Although the injured proprietor (as pedestrian) can recover for the lost profits caused by the physical injury, the workers could not recover their lost wages from the negligent driver.85
These cases pose many of the problems present in the cases involving stand-alone emotional harms. “The common thread running through the limitations on recovery for emotional distress, consortium, and economic loss is not difficult to identify. . . . [I]t is an age-old concern about extending liability ad infinitum for the consequences of a negligent act.”86 To see why, assume now that (n – 1) individuals suffer foreseeable stand-alone economic harm rather than emotional distress. The resultant duty of the potential injurer again encompasses n individuals: the pedestrian and the (n – 1) individuals who suffer pure economic loss as a result of the negligence. As compared to the number of individuals encompassed by the duty for emotional distress, the duty for economic harm is likely to include even more potential victims given the ripple effect of commercial or economic harms.
As in the case of stand-alone emotional harms, courts and commentators believe that a duty for pure economic loss would be overly burdensome for negligent defendants.87 Justifying the limitation of duty in terms of the negligent defendant’s interests, though, is subject to the same critique that undermines the justification as applied to stand-alone emotional harms. In claims involving only economic loss, the economic interests of the plaintiffs and defendant are implicated. The defendant has acted negligently, so her relevant economic interests should be subordinate to the harmed economic interests of the nonculpable plaintiffs. Such a priority of interests is inherent in the duty of care owed to physically injured victims, which enables them to recover for economic losses. That priority is also inherent in the tort rules allowing some plaintiffs to recover for pure economic loss, showing once again that the negligent defendant’s interests are subordinate and do not persuasively justify the duty limitation.
Until the 1950s, it was “virtually impossible” for a plaintiff to recover for pure economic loss in a negligence action absent privity with the defendant.88 Since then, the bar to recovery has eroded somewhat, so that today plaintiffs may be able to recover for negligently inflicted stand-alone economic harms in situations lacking privity. In most of these cases, the duty of care is based on the defendant’s special relationship with the plaintiff or the defendant’s undertaking of such a duty directly to the plaintiff.89 So defined, the duty involves a limited number of claimants and involves situations, like accounting or legal malpractice, that usually do not involve physical injuries. By allowing a limited number of plaintiffs to recover for pure economic loss, these tort rules give the economic interests of those plaintiffs legal priority over the economic interests of negligent defendants. That priority is not unique to plaintiffs of this type, however, because “economic loss cases lacking [the specter of widespread tort liability] do not receive distinctive treatment” from the courts.90
The legally subordinate economic interests of a negligent defendant do not justify limiting the recovery of those with superior interests, including individuals who foreseeably suffer pure economic loss as a result of the negligence. Any justification for limiting the defendant’s obligations should rely on interests with legal priority over those of the general class of victims who would be denied compensation for pure economic harms. The security interest of physically injured individuals has the greatest legal priority, creating the compensatory question of whether those interests justify a limitation of duty for cases only involving economic loss. Without that limit, there is a significant likelihood that the negligent defendant will be unable to pay compensatory damages to the physically injured plaintiff. The priority of security therefore justifies the limitation of duty. For cases in which the security interest is not implicated, as in accountant or legal malpractice cases, the compensatory norm requires limitation of duty to protect the most important set of economic interests. Tort law thus limits duty to the third-party beneficiary of an audit or will, ameliorating the concern that more widespread liability would bankrupt the defendant and undercompensate the most important set of economic interests.91
As in the cases of pure emotional distress, the tort rules limiting compensation for economic loss reflect the fundamental concern for compensating the security interest.92 In a world of scarce resources, negligent defendants typically will be unable to pay for all injuries they cause. To help ensure that physically injured victims are adequately compensated, tort law must limit the claims that can be levied against the assets of a negligent defendant.
Thus, the compensation rationale plausibly explains the tort system’s general reliance on negligence liability and the important duty limitations of such liability. The explanation is incomplete, though, unless it extends to the relation between the rules of negligence and strict liability.
IV. STRICT LIABILITY
Outside of the products liability context, the most important form of strict liability is the rule of strict liability for abnormally dangerous activities.93 According to the draft Restatement (Third) of Torts: Basic Principles, this rule can be justified in terms of reciprocity.94 The reciprocity rationale is best understood in compensatory terms. This rule of strict liability also can be justified by deterrence considerations, a justification that again conforms to the compensatory norm.
Reciprocity involves an important aspect of the risky interaction we have not yet considered. The analysis so far has addressed only those risky interactions in which the potential injurer (the driver) unilaterally imposes a risk of physical injury on the potential victim (pedestrian). Not all risky interactions are of this type. Two drivers, for example, each impose a risk of physical injury on the other, and each face the risk of being physically injured. In the extreme case of perfect reciprocity, the interacting individuals are identical in all relevant respects, including the degree of risk that each imposes on the other, the severity of injury threatened by the risk, and the liberty interests advanced by the risky behavior.
In a highly influential account of reciprocity, George Fletcher argued that reciprocal risks should be governed by negligence because “a rule of strict liability does no more than substitute one form of risk for another-the risk of liability for the risk of personal loss.”95 Without further elaboration, this argument is unpersuasive. Its criticism of strict liability for inappropriately substituting one form of reciprocal risk for another can also be leveled at negligence liability. More precisely, negligence liability could involve substituting the reciprocal risk of personal loss for the reciprocal risk of liability that otherwise would obtain under strict liability. Why is this substitution inappropriate for strict liability but not for negligence liability? The question can only be answered by reference to the appropriate background rule, but reciprocity does not justify negligence liability or strict liability.96 Rather than justifying either rule, reciprocity arguments assume that either negligence or strict liability is the appropriate background rule. What, then, is the significance of reciprocity?
Reciprocity becomes meaningful if understood in compensatory terms. As courts have long recognized, a negligence regime can be adequately compensatory even though accident victims do not receive tort damages for all their injuries. The victim of a reasonable risk does not recover under the negligence rule, but he “receives his compensation for such damage[s] by the general good, in which he shares, and the right which he has to [engage in similarly risky conduct].”97 The underlying idea is that reasonable risks typically are quite small and merge into the background risk of society assumed by each of its members.98 More significant risks merit special consideration. Highly significant risks created by common activities, such as automobile driving, are reciprocal and merge into the acceptable level of social risk. Those individuals injured by such risks are not compensated under a negligence rule, but the negligence regime works equally in their long-run favor (reciprocity) if they also engage in the activity.
This same reasoning explains the rule of strict liability. Highly significant risks created by uncommon activities are not reciprocal and do not merge into the background level of social risk. For these risks, the negligence rule does not provide an adequate reciprocal benefit for potential victims. Absent such a benefit, potential victims must be compensated for their injuries, the result attained by the rule of strict liability for abnormally dangerous activities.
Reciprocity therefore becomes meaningful for tort purposes if understood in compensatory terms. But how should reciprocity be conceptualized as a compensatory matter? In any given risky interaction, like that involving the driver and pedestrian, the risk is decidedly nonreciprocal. Yet the interaction can be conceptualized as being reciprocal insofar as the pedestrian over time also engages in a like amount of driving. The pedestrian’s right to drive, and her exercise of that right over time, may mean that driving confers an adequate benefit on her, making a particular risky interaction with a driver reciprocal. Should reciprocity as compensation be conceptualized in long-run terms or relative to each risky interaction?
The tort compensatory norm defines reciprocity in terms of risky interactions rather than long-run interactions. Only that conception fully explains how reciprocity affects the compensatory relation between the rules of negligence and strict liability.
In the case of a perfectly reciprocal risky interaction, there are no relevant differences between the two interacting parties. The two can be conceptualized as one entity. Whatever safety precautions are required of one individual will be required of the other. Whatever safety benefits accrue to one person will accrue to the other. Because the costs and benefits of tort liability inure equally to both individuals, the interests of each are best protected by the cost-benefit or allocatively efficient tort rule.” Negligence liability is more efficient than strict liability, all else being equal, due to the higher cost of injury compensation via the tort system as compared to other insurance mechanisms.100 Negligence liability for reciprocal risks accordingly provides the best protection for the interests of potential victims, thereby satisfying the compensatory norm.101
Nonreciprocal risks, by contrast, do not have this important compensatory feature. For such risks, if negligence liability and strict liability would each attain identical risk levels, then the guarantee of injury compensation gives strict liability a decisive compensatory advantage. The compensatory rationale therefore explains why the rule of strict liability for abnormally dangerous activities is limited to highly significant, nonreciprocal risks whenever the rules of negligence and strict liability would attain identical risk levels.102
This compensatory account requires consideration of the risk levels attainable by each liability rule. Those risk levels must be determined by reference to risky interactions. How does negligence liability affect interactions between drivers and pedestrians as compared to strict liability? The answer does not depend on whether the pedestrian is walking to the garage to pick up her car. Yet that is the sort of answer implicitly provided by the long-run accounts of reciprocity. For compensatory purposes, reciprocity must be conceptualized in terms of risky interactions.103
To be sure, courts have described reciprocity in long-run terms, concluding that strict liability is not appropriate for matters of common usage.104 The choice between the rules of negligence and strict liability does turn on this factor, but only when strict liability would reduce risk and become more compensatory than negligence liability.
The compensatory rationale identifies another role for strict liability not dependent on reciprocity. This role is immanent in the division between the rules of negligence and strict liability. The rule of strict liability for abnormally dangerous activities implies that negligence liability inadequately compensates such risks. A related implication is that negligence liability adequately compensates other risks, even though accident victims do not always receive damages when injured. One form of such compensation is identified by the reciprocity rationale, which explains how potential victims can benefit from a negligence rule in ways unrelated to the receipt of damages for injury. Another important form of such compensation stems from the ability of a negligence standard to reduce risk below the level attainable by strict liability.105 Whenever negligence liability loses its deterrence advantage, there may be a compensatory rationale for strict liability having nothing to do with reciprocity. When strict liability would reduce risk relative to negligence liability, strict liability better compensates potential victims both in terms of risk reduction and injury compensation.
Such contexts undoubtedly exist. Injurers can avoid negligence liability if they do not get caught. Even identifiable (and solvent) injurers can avoid negligence liability due to difficulties of proof. The amount of care actually required by the negligence standard depends upon the evidence available to plaintiffs and courts concerning the benefits and burdens of feasible risk-reducing measures. If good evidence concerning desirable safety precautions is unavailable, a negligent injurer who fails to take such precautions will escape liability.106 Similarly, a negligent injurer can avoid liability by relying on evidentiary problems concerning causation, problems that may be unique to negligence.107 Self-interested actors, aware of these avenues for escaping negligence liability, predictably disobey the tort duty to avoid the cost of exercising reasonable care.
In these contexts, strict liability can more effectively enforce the tort duty. A self-interested individual is concerned only about minimizing her costs: the sum of safety expenditures and liability costs. Under strict liability, the desire to minimize costs leads such actors to take safety precautions to reduce the incidence of injuries and payment of tort damages. If these actors would not take such precautions under an imperfectly enforced negligence rule, then strict liability will increase precautions and reduce risk below the level attainable by a negligence regime.
The rule of strict liability therefore can address the possibility that enforcement problems will induce self-interested actors to act negligently, a particular concern for contexts in which actors are motivated by profit maximization.108 Such an enforcement rationale for strict liability has been recognized by generations of tort scholars and is an express rationale for strict products liability.109 Moreover, the enforcement rationale provides the best interpretation of the rule of strict liability for abnormally dangerous activities promulgated in the Restatement (Second) of Torts.110
Consider the factor of common usage. If an activity is “customarily carried on by the great mass of mankind or by many people in the community,” it is not abnormally dangerous and not subject to strict liability.111 The factor of common usage would seem to express the concept of reciprocity, but as argued earlier reciprocity must be conceptualized in terms of risky interactions for compensatory purposes. That an activity like driving is common in the community at large does not mean that drivers and pedestrians impose reciprocal risks upon each other. Instead, such common usage is relevant to the issue of whether strict liability would reduce risk.112
An activity carried on by a large percentage of the population, like driving, presumably has significant private and social value. Consequently, strict liability is unlikely to affect significantly the amount of driving, even though the individual decision of whether to drive on a given occasion is an “activity” not fully regulated by negligence liability.113 The care people exercise while driving, though, can be effectively subjected to the more exacting requirements of a negligence standard, resulting in driver precautions that would not be taken in a regime of strict liability. All things considered, negligence appears to be the better rule for purposes of risk reduction, explaining why strict liability is not appropriate for common activities.
An activity that is not common and poses significant risk despite the exercise of reasonable care, by contrast, may be deterred by strict liability. For these activities, the negligence standard has a limited effect on risk reduction, whereas the lack of widespread participation in the activity suggests that strict liability may reduce risk by further reducing participation in the activity-a desirable outcome for activities otherwise lacking in social value. The deterrence rationale for strict liability accordingly explains why such activities are “abnormally dangerous” and subject to strict liability.114
Once again, a compensatory rationale for tort liability can explain an important structural feature of tort law, this one involving the division between the rules of negligence and strict liability. Ordinarily, the negligence standard can reduce risk below the level attainable by strict liability, providing an important compensatory advantage for negligence liability. For contexts in which the rules of negligence and strict liability would each attain identical risk levels, negligence liability is adequately compensatory for reciprocal risks. Absent reciprocity, negligence liability loses its compensatory advantage, so the greater availability of damages under strict liability explains why these activities are “abnormally dangerous.” In the remaining contexts, strict liability would reduce risk below the level attainable by an imperfectly enforced negligence standard, in addition to providing a greater range of injury compensation. The compensatory norm again explains why such activities are “abnormally dangerous” and governed by strict liability.
The remaining important limitations of tort liability involve assumption of risk and contributory negligence, defenses available to a negligent defendant that reduce or eliminate any compensation owed to the injured plaintiff. Like the other important limitations of tort liability, these limitations can also be explained in compensatory terms.
The defense of assumption of risk absolves a negligent defendant of liability for a risk that was knowingly and voluntarily consented to by the plaintiff.115 The existence of knowing and voluntary consent to the risk implies that this limitation of liability conforms to the compensation rationale. The compensatory norm requires that a potential victim be made no worse off by the risky interaction. In its ideal form, the norm is embodied in a consensual transfer between the two parties of the WTA risk proceeds. However, in important circumstances, a potential victim can benefit from the risky interaction even if she does not receive any transfer from the potential injurer. For example, individuals routinely consent to the risks created by contact sports because the benefit they derive from participating in the risky activity outweighs any risk of injury. That benefit explains why someone would knowingly and voluntarily consent to a risk, even if the consent would absolve a potential injurer of the duty to compensate. The risky activity confers a benefit on the potential victim that makes her no worse off than she would be without participating in the activity-an outcome that can be inferred from the fact of knowledgeable consent-so compensation from the potential injurer is not required.
The other liability-limiting defense is the doctrine of contributory negligence. In its orthodox form, contributory negligence barred a plaintiff from recovery, whereas today the doctrine reduces damages under principles of comparative fault.116 The rationale for the doctrine has eluded explanation.117 The compensatory rationale persuasively explains the doctrine of contributory negligence, while being agnostic on the question of who should bear liability when an injury is jointly caused by a negligent defendant and a contributorily negligent plaintiff. Contributory negligence in its orthodox and modern forms is consistent with the compensatory norm.
The analysis so far has assumed that the risk of physical injury can be reduced by precautions unilaterally taken by the potential injurer. Most risks, however, can be reduced if the potential victim also takes precautions. Any precautionary burden incurred by the potential victim must be compensated, as an uncompensated precaution would make the potential victim worse off than she would be without the risk exposure.118
To determine what precautions should be taken by the potential victim, we must reconsider the idealized compensatory exchange between the potential injurer and victim. By definition, the potential victim would agree to face a specified risk in exchange for the WTA risk proceeds. The parties therefore would agree that the potential injurer must take precautions costing B for which B
Like negligent conduct by the potential injurer, contributorily negligent conduct by the potential victim breaches the compensatory agreement. Hence it is not difficult to identify a compensatory rationale for contributory negligence. Rather, the difficult issue would seem to be one of allocation. Each party’s breach was a cause of the injury. Should the negligent injurer pay for the injury? The contributorily negligent victim? Or should the damages be apportioned between them?
Perhaps surprisingly, the compensatory norm does not answer these questions. As discussed earlier, the norm gives priority to the security interest over the liberty interest only for interpersonal interactions.120 Thus the pedestrian’s interest in physical security has legal priority over the driver’s liberty interest. The compensatory norm does not require an intrapersonal priority of interests giving the pedestrian’s security interest priority over her own liberty interest. However, the nature of the potential victim’s liberty interests change once they involve precautions required by the compensatory exchange with the potential injurer. The potential victim would be compensated for any precautions required by that exchange, so her failure to take such precautions necessarily involves interpersonal liberty interests governed by the compensatory norm.
Cases of contributory negligence accordingly involve two injurers for compensatory purposes: the driver and the contributorily negligent pedestrian. The compensatory norm merely requires that the injurer compensate the victim. In cases of multiple injurers, which injurer should provide compensation is an issue without any necessary analytic resolution. Payment by either injurer would satisfy the compensatory requirement.
Prior to the adoption of comparative fault, contributory negligence completely barred a plaintiff from recovery. In effect, the rule treated the contributorily negligent plaintiff as the single injurer for compensatory purposes. This compensatory interpretation explains why courts justified the bar to recovery in causal terms, “saying that the plaintiff’s negligence is an intervening, or insulating, cause between the defendant’s negligence and the result.”121 Of the two causes (or injurers), the courts singled out the victim, unless the negligent defendant had the last clear chance of avoiding the injury.122 Again, this allocation of liability was frequently explained in causal terms: “[T]he later negligence of the defendant is a superceding cause which relieves plaintiff of liability for it.”123 By identifying a single “cause” of the accident, the courts in effect identified a single injurer for compensatory purposes, eliminating the possibility of apportionment.
Rather than identifying a single “cause” of the accident, the modern rule of comparative fault recognizes that each party caused the accident, resulting in multiple injurers for compensatory purposes. Apportioning damages among the multiple injurers has evident fairness that explains the widespread adoption of comparative fault. That outcome, though, was not mandated by the legal priority of security over liberty and the associated norm of compensation.
VI. COMPENSATION AND THE DEVELOPMENT OF TORT LAW
The important substantive doctrines of tort law are consistent with the compensatory norm. However, consistency does not constitute a sufficient description of tort law. To persuasively explain the practice of tort law, the compensation rationale must also adequately describe the self-understandings or motives of participants in the tort system.124
No single theory will fully describe the self-understandings of those who participated in the evolution of the tort system. The most that can be expected of a theory in this regard has been well articulated by Oliver Wendell Holmes:
The law did not begin with a theory. It has never worked one out. The point from which it started and that at which I shall try to show that it has arrived, are on different planes. In the progress from one to the other, it is to be expected that its course should not be straight and its direction not always visible. All that can be done is to point out a tendency, and to justify it.125
If the inquiry is understood in these terms, then it is fair to say that over time participants in the tort system have understood tort law in compensatory terms.
Originally, tort damages were awarded as an incident of criminal prosecution. The linkage of criminal and tort liability meant that the early common law courts “approach[ed] the field of tort through the field of crime.”126 Tort actions continued to be quasi-criminal until the late seventeenth century.127 The initial linkage of tort actions to criminally caused harms implies that tort law originated as a means of compensating wrongfully (criminally) caused injuries. Once freed from the limits of criminal law, the tort system faced a choice. It could continue to require wrongful conduct as a predicate for liability, or it could instead base liability on the fact of harm, whether caused wrongfully or not.
As a matter of precedent, the tort system could adopt a compensatory conception given the priority of the security interest over the liberty interest. That priority of interests was recognized by the early common law. According to Sir William Blackstone, the first absolute right held by individuals involves the “right of personal security” whereas “[n]ext to personal security, the law of England regards, asserts, and preserves the personal liberty of individuals.”128
In Blackstone’s time, however, there was no separate body of law called “torts.” No treatises were published on the law of torts, in either England or the United States, before 1850.129 The modern tort system emerged in the nineteenth century as a response to the increasing number of accidental injuries caused by industrialization.130 The field of torts was created by the legal scientists, most notably Holmes, following the demise of the writ system in the latter half of the nineteenth century.131
Given the priority of security interests inherent in the intentional torts, precedent would seem to favor the rule of strict liability for accidental harms. This priority explains why some of the “greatest common law authorities” had maintained that “under the common law a man acts at his peril.”132 The claim that precedent supported such an extreme form of strict liability was famously rejected by Holmes.133 But Holmes did not reject the priority of security interests over liberty interests. Instead, he argued that tort liability is limited by the requirement of foreseeability,134 a limitation consistent with the compensatory norm. Holmes first rejected strict liability defined exclusively in terms of physical causation and then argued that precedent supported negligence liability subject only to the foreseeability limitation inherent in the standard that “the defendant was bound to use such care as a prudent man would do under the circumstances.”135
The negligence principle unified the field.136 It was widely recognized, though, that negligent conduct does not equate with moral culpability.137 Conduct is negligent if it breaches the standard of care. That standard, in turn, requires a comparison of liberty and security interests. Tort law in the nineteenth century did not reject the priority of security over liberty, and that priority had compensatory implications that would be worked out over the next hundred years by the evolving tort system.
At the time when the modern U.S. tort system emerged, the states were abolishing the writ system, which used pleading requirements in a manner that often masked difficult issues of substantive law.138 Although this procedural reform was not supposed to affect substantive law, the elimination of the writ system led to some substantive issues of great importance, such as the choice between the rules of negligence and strict liability, being presented for the first time in their most general form to the courts.139
Around the same time, legal decisionmakers were rejecting natural-law justifications in favor of more pragmatic, instrumentalist justifications. Whether this shift occurred before or after the Civil War is a point of some contention, but it is clear that judicial decisionmaking became increasingly pragmatic and policy-oriented as the nineteenth century progressed.140
These developments occurred at a time when increased industrialization and use of technology, particularly the railroad, substantially increased the number of accidental injuries.141 In light of the enormous toll of the Civil War and its impact on society, the problem of accidental injury and death became particularly salient.142 “[O]bservers of the industrial scene in the 1880s were struck by the fact that the peacetime economy produced more death and injuries than the war that preceded it.”143
The confluence of these factors profoundly affected the newly emergent tort system. The abolition of the writ system forced the courts to address directly the problem of accidental harms. At this time, judges were increasingly pragmatic, focusing on the pronounced problem of severe physical injuries and death. In these circumstances, the limitations of strict liability were perhaps more apparent than they are today. Damages in wrongful death actions do not provide a remedy for the life prematurely taken from the decedent, and a damages award for physical injury is less desirable than prevention. For good and obvious reasons, pragmatic judges would have been attracted to the deterrence advantages of negligence liability.
The tort system could adopt negligence liability consistently with the priority of security over liberty. As argued earlier, that priority would lead courts to consider the risk-reducing features of negligence liability as being the decisive reason for choosing the rule of negligence over strict liability. This explains why most observers of the time thought that negligence liability served the purpose of punishing or deterring civilly blameworthy or “negligent” behavior.144 That understanding of tort liability was frequently expressed in judicial opinions, as Gary Schwartz found in an extensive study of the case law during this period:
Two [themes that recurred in the case law] were a judicial concern for the risks created by modern enterprise and a judicial willingness to deploy liability rules so as to control those risks. Another set of themes included a judicial solicitude for the victims of enterprise-occasioned accidents and a judicial willingness to resolve uncertainties in the law liberally in favor of those victims’ opportunities to secure recoveries.145
How could there be “judicial solicitude for the victims of enterprise-occasioned accidents” if courts rejected strict liability in favor of a negligence regime? Perhaps the judicial concern for potential victims was an insincere rationale for rejecting strict liability and the burden it may have imposed on business development.146 Alternatively, the courts may have believed sincerely that negligence liability offered the best overall protection for potential victims. Courts apparently conceived of the protection as taking the forms of deterrence and compensation, reflected in the “judicial concern for the risks created by modern enterprise,” the “judicial willingness to deploy liability rules so as to control those risks,” and the “judicial willingness to resolve uncertainties in the law liberally in favor of those victims’ opportunities to secure recoveries.”
Consequently, the compensatory advantages of negligence liability-deterrence plus monetary damages-were important considerations for the courts at the time they were adopting the negligence regime. These considerations may not have supplied the true motivation for the courts. But insofar as judicial opinions reflect sincerely held beliefs, the compensation rationale is consistent with the judicial understanding of negligence liability when the modern tort system emerged.
This reasoning explains why one prominent scholar can conclude that the “modern negligence principle” was “an intellectual response” to the increased number of accidents in the nineteenth century,147 whereas another can just as defensibly conclude that the new nineteenth century liability standards developed from precedent in an evolutionary way.148 Precedent gave priority to security over liberty, and although many great common law jurists thought the priority justified strict liability, the better reasoned response to the pronounced problem of accidental injury lay in the development of modern negligence law.
It was not until the twentieth century, though, that the compensatory nature of tort liability became apparent to participants in the tort system. During the period from 1910 to 1945, the conceptual jurisprudence of the legal scientists was largely displaced by Legal Realism.149 The legal scientists, like Holmes, had conceptualized negligence in terms of a universalized general standard of care owed by all to all.150 The generality or vagueness of a universal “fault” standard obscured the comparison of conflicting individual interests necessarily involved in a negligence inquiry. The Realists transformed negligence theory in the 1920s in a manner that would expose the underlying compensatory purpose of tort law:[T]he notion of an abstract universal duty was gradually abandoned and a “relational” concept of duty was substituted. Relational negligence theory introduced questions of “interest-balancing,” inviting judges to compare the magnitude of the risks to which a plaintiff was exposed, and the social worth of the class . . . a plaintiff represented, with the social utility of the defendant’s conduct.151
As the analyses in prior sections show, if one focuses on the relationship between the plaintiff and defendant, and expressly balances their interests consistently with the priority of security over liberty, then the compensatory structure of tort law becomes apparent. A focus on compensation would force one to look to strict liability as a compensatory complement to negligence liability. Thus, the relational theory of negligence and the interest-balancing advocated by the Realists revitalized the theory of strict liability:
Relational negligence, while retaining the concept of fault, had made the determination of where fault lay in a given case a process of interest-balancing. Strict liability theory proposed to conduct the interest-balancing altogether unburdened by notions of fault. In 1917 Jeremiah Smith had found those isolated instances of “strict” liability in tort law exceptional in their repudiation of the general fault standard. By 1941 Presser was prepared to ‘question whether “fault,” with its popular connotation of personal guilt and moral blame, and its more or less arbitrary legal meaning, which [would] vary with the requirements of social conduct imposed by the law, . . . [was] of any real assistance in dealing with . . . questions of [risk allocation], except perhaps as a descriptive term.’152
Whereas tort liability was necessarily predicated on criminal wrongdoing when tort liability was an incident of criminal liability, upon their separation tort law needed to develop some independent standard for defining tortious conduct. That standard necessarily involves the interest-balancing uncovered by the Realists. Once tortious conduct was defined in terms of interest-balancing, it became clear that tort liability could be decoupled from notions of moral culpability or “fault” in its popular sense. If the balance of interests dictated as much, one could be responsible for an injury even if the injury-causing conduct was not morally blameworthy.
The influence of Realism waned after the second World War and tort scholars returned their attention to doctrine, now analyzing it with the interest-balancing and relational concepts developed by the Realists.153 A debate over the purpose of tort law ensued. It “was well established by the 1950s” that “tort law had become primarily a compensation system designed to distribute the costs of injuries throughout society efficiently and fairly . . . .”154 This compensatory rationale for tort liability was not based on a rigorous balancing of interests that prioritized security over liberty, although its methodological commitments were consistent with such a priority.155 A change in social attitudes towards injury undoubtedly played a major role.156 But as the foregoing analysis has sought to establish, the eventual emergence of the compensatory rationale for tort liability is the logical implication of the longstanding tort principle that security has priority over liberty.
Today, of course, debate continues over the appropriate purpose of tort liability. Despite the extensive debate concerning the comparative merits of allocative efficiency and corrective justice, a large number of scholars, and perhaps most participants in the tort system, believe that tort law serves the purposes of compensation and deterrence.157 Those objectives may be unconnected and inconsistent with each other.158 Alternatively, compensation and deterrence become unified if the security interest has priority over the liberty interest. The priority defines the categories of injurer and victim, making it possible to define the ideal compensatory exchange. That exchange creates precautionary obligations for both parties, explaining why deterrence is an integral component of compensation. Unless one believes that most participants in the tort system have an irrational understanding of the relation between compensation and deterrence, the tort norm of compensation plausibly describes their understanding of tort law.
Tort law has consistently given one’s interest in physical security priority over a conflicting liberty interest of another.159 That priority underlies the evolution of tort law, from the early intentional torts to the subsequent emergence of negligence liability and its modern development. As an analytic matter, this priority yields a compensation rationale for tort liability. The Restatement (Second) of Torts therefore is on firm doctrinal ground when it states that “the purposes for which actions of tort are maintainable . . . are: (a) to give compensation, indemnity or restitution for harms; (b) to determine rights; (c) to punish wrongdoers and deter wrongful conduct; and (d) to vindicate parties and deter retaliation or violent and unlawful self-help.”160 Rather than being unrelated reasons for tort liability, each of these “purposes” coherently implement a norm of compensation in a world of scarce resources.
Understood in nonideal terms, the compensatory rationale also finds expression in the draft Restatement (Third) of Torts: Basic Principles. As an apparent compromise to the ongoing debate whether the appropriate purpose of tort law is one of efficiency or fairness, the Restatement (Third) justifies negligence liability as a remedy for wrongful behavior and as a deterrent to such behavior.161 This rationale is consistent with the compensatory norm, even though the justification for compensation-the priority of security over liberty-is not based on the wrongfulness of the injury-producing conduct.
Consider someone who chooses to act negligently. That conduct does not become permissible merely because the tortfeasor is obligated to pay damages for any injuries. An important aspect of nonideal compensation involves the reduction of risk below the ideal level attained in a fully compensatory regime. By acting negligently, the tortfeasor has failed to provide the safety required to make the risky interaction adequately compensatory. When that failure injures or kills someone, the outcome is precisely the consequence the negligence regime sought to avoid. Monetary damages for the injury are not always adequately compensatory, so the mere payment of such damages does not sufficiently remedy the wrong.162 Negligent behavior can be wrongful, even if the general justification for tort liability is based on compensation.
Although the compensation rationale can explain the important features of tort law, it does not follow that compensation is a normatively defensible basis for tort liability. The priority of security over liberty, however, has normative appeal.163 And, even though the compensatory norm justifies allocatively inefficient tort rules, those rules satisfy the requirements of modern welfare economics.164 The compensatory norm has evident appeal, but the conclusion for present purposes is more limited. Tort law can be coherently understood in compensatory terms, despite the large number of accidental injuries that receive no damages compensation.
* Professor of Law, New York University School of Law. Special thanks to the symposium participants, particularly Heidi Li Feldman, for their comments. I am also grateful for the helpful comments I received from Daryl Levinson, Liam Murphy and Stephen Perry, and from Marshall Shapo and other participants in the faculty workshop at the Northwestern University School of Law. This research was supported by the Filomen D’Agostino and Max E. Greenberg Research Fund of the New York University School of Law.
Copyright Georgetown University Law Center Mar 2003
Provided by ProQuest Information and Learning Company. All rights Reserved