In this chapter I continue my treatment of the immanence of corrective justice in tort law by focusing on strict liability, often regarded as the great competitor of negligence liability. To view negligence liability as corrective justice, as I did in the preceding chapter, implies the rejection of strict liability. Here I set out the reasons for this rejection and deal with the pockets of strict liability found in the common law. Taken together, the discussions in Chapters 6 and 7 reverse the more standard picture, in which strict liability is thought to be a more plausible manifestation of corrective justice than is negligence law.
To be sure, aligning corrective justice with strict liability has a superficial attraction. Under strict liability causa, not culpa, is paramount: liability follows from the occurrence of the damage at the defendant’s hands, regardless of whether the defendant’s behavior was faulty. By focusing on the sheer impact of one person’s behavior on another, strict liability links the parties only as doer and sufferer, and cuts off inquiry into distributive considerations that interrupt the immediacy of their relationship. Strict liability thus appears to fit easily within corrective justice. And this fit seems to be corroborated by the presence in the common law of doctrines that hold faultless defendants liable.
My argument that strict liability is nonetheless incompatible with corrective justice divides into two parts. First, I present the theoretical inadequacy of strict liability. To this end, I criticize Richard Epstein’s sustained effort to vindicate strict liability as a requirement of justice between the parties.1 My argument will be that Epstein’s position is consistent neither with corrective justice’s equality nor with its idea of agency nor with its correlativity of right and duty.
This part of the chapter contains a Kantian reformulation of Oliver Wendell Holmes’s masterly survey, more than a century ago, of the competing liability regimes.2 In seeking to discover “the common ground at the bottom of all liability in tort,”3 Holmes set out to explain why liability for unintended harm employs the objective standard of negligence rather than either the more expansive idea of strict liability or the more restrictive standard that reflects the actor’s subjective moral shortcoming. Holmes rooted his views in utilitarian considerations of the community’s expedience, rather than in the immediate normative connection of doer and sufferer. As we shall see, however, his arguments against strict liability and the subjective standard fit readily within Kantian right.
In the second part of the chapter, I go on to consider particular common law doctrines that are often thought to embody strict liability: respondeat superior, liability for abnormally dangerous activities, liability for nuisance, and liability for the use of another’s property to preserve one’s own. The question here is whether corrective justice can be seen as immanent in these well-established doctrines. My argument is that if we have regard for their specific contours, we shall see that these doctrines are either extensions of fault liability or are ways in which the common law regulates the use of property in accordance with corrective justice. Their existence, therefore, poses no challenge to the ideas I have been developing so far.
Epstein’s case for strict liability takes two forms. The first, which can be termed the argument from the hypothetical of self-injury, contends that strict liability follows from the fact that the actor would bear the loss if he or she had also suffered the injury. The second, which can be termed the argument from the concept of property, contends that strict liability is the analytic concomitant of the plaintiff’s ownership of what was injured. By setting up a framework within which commonsense notions of causation lead to liability even in the absence of fault, each of these arguments treats strict liability as corrective justice. Yet, as we shall see, they both fail.
Epstein raises the hypothetical of self-injury in his analysis of Vincent v. Lake Erie Transportation Co.4 The issue in that case was whether the plaintiff could recover for damage to his dock after the defendant deliberately kept his boat moored there to protect it from a storm. Epstein’s justification for imposing liability is as follows:
Had the Lake Erie Transportation Company owned both the dock and the ship, there could have been no lawsuit as a result of the incident. The Transportation Company, now the sole party involved, would, when faced with the storm, apply some form of cost-benefit analysis in order to decide whether to sacrifice its ship or its dock to the elements. Regardless of the choice made, it would bear the consequences and would have no recourse against anyone else. There is no reason why the company as a defendant in a lawsuit should be able to shift the loss in question because the dock belonged to someone else. The action in tort in effect enables the injured party to require the defendant to treat the loss he has inflicted on another as though it were his own. If the Transportation Company must bear all the costs in those cases in which it damages its own property, then it should bear those costs when it damages the property of another.5
Epstein accordingly proposes to determine the liability regime through a thought experiment that unites the interests of the plaintiff and the defendant and holds the defendant liable for whatever costs would be sustained by the amalgamated individual. The argument moves from the actual separateness of plaintiff and defendant, to the identification of the two in a superindividual who bears the costs of self-inflicted injuries, and then back again to the actuality of separate existences. In Epstein’s view, this procedure yields a justification for holding the defendant liable for damage done to the plaintiff whether negligently or not. The argument for strict liability proceeds “on the assumption that the defendant must bear the costs of those injuries that he inflicts upon others as though they were injuries that he suffered himself.”6
However, Epstein’s argument does not in fact point unambiguously to strict liability. Two other liability rules can be elicited from the amalgamation of the litigants:
A. The defendant should be liable even for harm that results from behavior that does not manifest the defendant’s volition. Epstein’s premise that “the defendant must bear the costs of those injuries he inflicts upon others as though they were injuries that he suffered himself” allows the argument that just as I would myself bear the costs of whatever injuries I sustained while sleepwalking or in the course of an epileptic seizure, so I should be held liable for any such harms that I inflicted on you. Since Epstein’s theory of strict liability accepts the uncontroversial common law doctrine that the defendant cannot be held liable for behavior that does not manifest volition,7 the hypothetical of self-injury can lead to a liability rule broader even than the one he supports.
B. The losses should lie where they fall. Once the parties are amalgamated, the person suffering the loss has no cause of action, since one cannot sue oneself. On this argument, the consequence of combining the litigants is to preclude tort liability for any losses.
Thus Epstein’s hypothetical of self-injury presents a large menu of possible liability regimes, from strict liability (Epstein’s preference) to liability even for nonvolitional conduct (possibility A) to the absence of liability (possibility B). It does not, however, make strict liability more plausible than any alternative. Epstein’s argument from the hypothetical of self-injury merely announces a conclusion that must be supported on other grounds.
Moreover, possibility B underlines the futility of reducing the two parties to one. Tort law presents the problem of whether the plaintiff or the defendant should bear the loss. Epstein’s amalgamation of the two parties merely transforms this problem into one of selecting the feature of the resulting hypothetical situation that is to be regarded as decisive. Is the decisive feature, as Epstein assumes, that the super-person suffers a loss that should remain the actor’s loss in the two-party situation? Or is it that the superperson suffers an irrecoverable loss that should remain irrecoverable when transposed into the actuality of litigation? Epstein views the harm as something that the defendant would have to bear in the hypothetical situation. But one can equally view it as something for which the plaintiff would not be able to recover. Thus the distinct normative claim of each of the parties survives their fusion into the notional superperson. The thought experiment that unites the parties cannot determine which of them is to bear the loss in the real world.
The root problem with the hypothetical of self-injury is that its notional amalgamation of the litigants is at odds with the irreducibly bipolar nature of corrective justice. As the ordering of the relationship of doer and sufferer, corrective justice necessarily connects two parties, no more and no less. Epstein’s transformation of the two-party problem of justice into a one-party loss is a misleading diversion. And as the need to choose between strict liability (Epstein’s preference) and no liability (possibility B) indicates, the amalgamation of the litigants does not in the end obviate the necessity of dealing with the irreducible bipolarity of the parties’ relationship.8
Epstein’s second argument is that “the idea of ownership necessarily entails a strict liability standard in all tort cases between strangers.”9 Ownership and property are, for Epstein, omnibus terms that refer to the entire range of one’s entitlements to external possessions and personal integrity.10 Viewing corrective justice noninstrumentally as a regime of rights, Epstein conceives of property as “an external manifestation of the principle of personal autonomy.”11 He contends that strict liability is conceptually implied by the very notion of private property because the absence of liability for nonnegligent injury amounts to the taking of a limited property interest.12 Epstein regards the boundary of what I own as circumscribing the area of my moral space, the domain within which I am entitled to be free of the intrusions of others. Your damaging my Ming vase even without fault, for example, is a penetration of this space that ought to trigger your liability. The idea of property makes the location of the action’s effects, not its innocence, decisive for liability. Allowing you to harm what I own is inconsistent with its being my property.
This argument mistakenly supposes that ownership immediately entails the immunity of what is owned from change through someone else’s action. That no such immediate entailment obtains is apparent as soon as you accidentally drop my Ming vase. I notice at once that the physical condition of my vase has changed drastically. What was previously a thing of beauty and value is now a worthless scattering of shards. Nothing, however, has affected my ownership as such. What had been my vase has become my shards, and the idea of property is embodied in the shards as surely as it was in the vase.
Epstein’s argument from property to liability is missing an intermediate step. With respect to my Ming vase, all we have so far is the conjunction of my owning it and your damaging it. These are merely two separate facts about the vase which in themselves no more entail liability than do any other facts about the vase (its color, its shape, its hardness, and so on). The normative connection between my property and your action is yet to be established. We must still determine what it is about property that morally limits the action of others. The question is not whether you have intruded into my moral space but whether the intrusion is compatible with the idea of moral space that this particular chunk of it, the Ming vase, instantiates.
For corrective justice, Kantian right supplies the applicable idea of moral space. Grounded as it is in the normative dimension of free and purposive agency, Kantian right construes the right of property as the embodiment of the agent’s freedom in the external world. To this extent Kantian right coincides with Epstein’s notion of property as the external manifestation of the principle of personal autonomy. In governing the interaction of free and purposive agents, Kant’s principle of right requires that the action of the defendant be capable of coexisting with the freedom of the plaintiff. As I have noted, this principle reflects the equality of the parties under corrective justice. Liability therefore arises when the defendant’s act is inconsistent with the very idea of free agency underlying both the defendant’s act and the plaintiff’s proprietary and other rights.
We thus return to the workings of corrective justice and Kantian right. In the following section I suggest that the ideas of equality under corrective justice and the concept of agency under Kantian right exclude strict liability. In other words, once we fill in the missing step in Epstein’s argument, we see that the case for strict liability is not only unsupported but wrong.
At common law, negligence is a failure to act in accordance with an objective standard of reasonableness. As Holmes observed, in making negligence the criterion of liability the common law rejects two other possibilities. One is strict liability. The other is that actors be liable on a subjective standard for failing to act as safely as their personal capacities allow. Each of these alternatives embodies apparently plausible normative claims. Strict liability reflects the idea that one should not be allowed to encroach on another’s moral space with impunity. The subjective standard reflects the idea that individuals should not be held to a standard they are incapable of meeting.
I propose to argue that neither strict liability nor the subjective standard conforms to corrective justice. My reason for considering them together is that, from the standpoint of corrective justice, the two have parallel defects. Whereas corrective justice treats the litigants as equals, strict liability and the subjective standard center themselves on only one of the parties—the former on the plaintiff, the latter on the defendant.
The inequality in strict liability emerges from the principle that the defendant is to be liable for any penetration of the plaintiff’s space. What is decisive for the parties’ relationship is the demarcation of the domain within which the law grants the plaintiff immunity from the effects of the actions of others; the activity of the defendant is then restricted to whatever falls outside this sphere. Thus the interests of the plaintiff unilaterally determine the contours of what is supposed to be a bilateral relationship of equals.
That the subjective standard is the mirror image of this inequality is apparent from the defendant’s argument in the leading common law case on the point. In Vaughan v. Menlove13 the defendant had placed his rick of hay close to his neighbor’s barn, ignoring warnings that spontaneous combustion in the rick might set the barn afire. When the barn subsequently burned down after fire spread to it from the rick, the defendant’s lawyer argued that his client should be absolved because he meant no harm: he was a stupid man, and “he ought not to be responsible for the misfortune of not possessing the highest order of intelligence.”14 The court, however, ruled that his lack of subjective blameworthiness was legally irrelevant: his failure to act as a person of reasonable and ordinary prudence was sufficient for his liability.
The court’s rejection of the subjective standard accords with corrective justice. The defendant’s argument, with its one-sided attentiveness to subjective moral capacity, is inconsistent with the transactional equality of the parties. The argument sets the boundary between the defendant’s right to act and the plaintiff’s freedom from the effects of that action at the limits of the defendant’s powers of evaluation. This means that a feature personal to the defendant sets the terms upon which the plaintiff must tolerate impingements. Whereas under strict liability something about the plaintiff is decisive in restricting the defendant to the space left over, under the subjective standard the positions are reversed.
In their preoccupation with one or the other of the poles of the relationship, neither strict liability nor the subjective standard treats the parties as equals. Strict liability one-sidedly orients the relationship to the standpoint of the plaintiff; the subjective standard one-sidedly orients the relationship to the standpoint of the defendant. Both liability regimes are inconsistent with corrective justice, and for the same reason.
Corresponding to such inequality is the absence, under both of these liability regimes, of right and correlative duty. Strict liability and the subjective standard are each marked by a converse one-sidedness: strict liability has right without duty, the subjective standard has duty without right.15
Because it is triggered solely by the causation of injury, strict liability has right without duty. Strict liability reflects extreme solicitude for the plaintiff’s rights. Under strict liability, the plaintiff’s person and property are a sacrosanct domain of autonomy, within which the plaintiff is entitled to freedom from interference by anyone else. But strict liability protects the plaintiff’s rights without allowing room for an intelligible conception of the defendant’s duty. A duty must be operative at the time of the act that the duty is supposed to govern. Under strict liability, however, the actor’s duty not to do the harm-causing act need not appear until the moment of injury. Only retrospectively through the fortuity of harm does it then turn out that the defendant’s act was a wrong. Thus under strict liability, the sufferer has a right to be free from the harm, but that right is not correlative to a duty, operative at the moment of action, to abstain from the act that causes the harm.
Under the subjective standard, by contrast, duty is present without right. Actors are under a duty to exercise care in accordance with their personal capacities, and when they fail to do so, those harmed can sue for compensation. The victim’s freedom from harm is thus derivative from the actor’s capacity to be aware of the harm’s likelihood. This freedom, however, is not the victim’s as of right. A right reflects the self-determining agency of the agent whose right it is. Persons have rights by virtue of being ends in themselves, not derivatively from the moral situation of others.16
The absence of equality under strict liability and the subjective standard reflects the incoherent conception of agency implicit in these liability rules. Because the equality of corrective justice is grounded in the equality of agents under Kantian right, an inequality under corrective justice also represents an incoherence regarding the outward exercise of one’s agency. This incoherence is evident in the judgment concerning the defendant’s action that the proponent of the subjective standard or of strict liability is implicitly inviting the court to make.
The incoherent judgment about action implicit in the defendant’s plea in Vaughan v. Menlove can be formulated as follows. An action is the actualization of the capacity for purposiveness in the external world. The initial stage of this actualization, in which a person’s purposive capacity is directed to a specific purpose, is on the Kantian view completely within the range of the actor’s self-determination: I can decide to put my rick of hay here or there. But the exercise of my freedom requires entry into a domain beyond my freedom. In extending my purposive capacity into the external world, I must step into a realm of nature, contingency, and the cross-purposes of others. In placing my rick here, I render it contiguous to my neighbor’s barn, and I expose both rick and barn to natural forces through which both may be destroyed.
The defendant’s argument that his purity of heart ought to constitute the standard of liability is an argument appropriate to the evaluation of action at the stage of potentiality, when the actor’s self-determination has not yet issued into the world beyond him and when an internal standpoint of judgment corresponds to the internal locus of the inchoate action. However, the judgment that he is inviting the court to make about his action is inconsistent with the stage to which the maturation of his purpose had progressed. In pleading that he is too stupid to have taken account of the external effects of his action, the defendant is claiming an entitlement to realize his projects in the world while retaining the exclusively internal standpoint applicable to projects as mere possibilities. He wishes to have the actuality of his projects treated from the standpoint of a now superseded potentiality.