I Professor Jiménez: Institutions and interests

Very perceptively, Professor Jiménez situates my view against the two most sharply defined contemporary positions in tort theory. One view is the Kantian formalism of Ernest Weinrib and Arthur Ripstein. The other is instrumentalism in its economic form, a view which has dominated American tort theory for the past fifty years. Jiménez characterizes my view as ‘instrumentalist,’ not ‘formalist.’^[1] The significance 72of the characterization is that normative, non-economic tort theory generally is not instrumentalist. In characterizing my view as instrumentalist, Jiménez is careful to distinguish between instrumentalism about law and instrumentalism about morality. The instrumentalism that he sees in my work pertains to law. Legal institutions are instruments for realizing values, not sites of intrinsic value. Tort and related institutions are tools through which we spell out what we owe to one another with respect to coercively enforceable obligations not to interfere with, or impair, each other’s urgent interests as we go about our lives in civil society. With respect to morality, Jiménez thinks I am not an instrumentalist. On a morally instrumentalist view, persons do not have intrinsic value; something else does, and it is the connection of that something to persons that confers value on persons. On a standard welfarist view, for instance, welfare or well-being has intrinsic value. Persons matter only insofar as they are locations where welfare alights. By contrast, I accept the Kantian view that persons and their interests have intrinsic value. They command our respect, set limits on our wills, and obligate us to treat each other in some ways and not in other ways. We may not use them simply to increase the overall welfare of some population of sentient beings. Jiménez then goes on to argue that my view entails a standing tension between the intrinsic values – or, from a different angle, the interests – that legal institutions serve, and the rights and responsibilities that those values and interest justify.

Jiménez is fundamentally correct. Moreover, his comment picks out important features of my view. Although he does not put it this way, one feature of my view is that I am pushing for a kind of ‘third way’ between the formalism of Ripstein and Weinrib and the instrumentalism of law and economics. A second feature is that the contours of our legal rights and responsibilities are vulnerable to being endemically ‘second-best.’ They may fail to get the values and interests they purport to realize right. Perhaps more significantly, they may fail to realize those values because they must respond to special institutional pressures to which law is subject. For one thing, legal rules must be administrable. For another, legal rules are meant to guide the actions of an indefinitely large population of actors, and they may therefore need to oversimplify – and may consequently be at best approximate embodiments of the values they serve. For a third, interests are squishy and require spelling out. Legal doctrine may need to deploy distinctively legal ‘principles’ to give content to abstract moral ‘principles.’^[2] For yet a fourth, legal rules may have to pursue their 73objectives indirectly, a phenomenon familiar from discussions of utilitarianism and a commonplace of contemporary property scholarship. The simplicity, paucity, and rigidity of forms of ownership in property law, for example, may be efficient even though the simple, rigid forms themselves are never efficiently tailored to any given transaction. Their simplicity and rigidity may facilitate interaction among vast numbers of strangers. Terms tailored precisely to the tastes of particular parties would thwart, rather than enable, such interaction by requiring intimate knowledge of the parties, transactions, and legal structures tailored to the details of their desires.^[3]

These points seem to me to be both correct and well-taken. I wish to push back only insofar as the enduring dichotomy between ‘instrumentalism’ and ‘formalism’ may obscure the fact that I think of my view as trying to navigate a middle way between the Scylla of the one and the Charybdis of the other. Ardent contemporary proponents of formalism in tort theory often speak as if instrumentalism conceived of value as wholly independent of law whereas formalism thinks of value as wholly intrinsic to law. For instrumentalism, value is complete without law and law is a mere tool for value’s implementation. For formalism, law’s embodiment of value makes no reference to any other source of value. In Ernest Weinrib’s version of the view, it seems that we can derive the content of the law simply from the bilateral form of the normal tort lawsuit. In Arthur Ripstein’s, it seems that we can work out the details of the abstract idea that ‘everyone is in charge of themselves and no one is in charge of anyone else’ without reference to anyone’s interests.^[4] This dichotomy misleads us both with respect to what instrumentalism must assume and with respect to what formal concepts can do. Normally, instrumental conceptions are incomplete without institutional articulation and, normally, institutional articulation of formal concepts cannot be done in justifiable ways without reference to values or interests independent of the concepts.

The way in which instrumentalism and formalism work together is illustrated by an institution as mundane as the traffic code. We do not need the traffic code to know that we all owe to one another obligations to cycle, drive, scooter, skateboard, and walk safely. Safety is the primary value that the traffic code serves, and it can be stated independent of the code. Reasonable care is the abstract obligation that the code makes concrete and it, too, can be stated independently of the code. Tort law reflects these truths. Statutory negligence takes the rules of the road to be specifications of duties independently recognized by the law of torts. But, in this case 74and many others, we cannot discharge our obligations without constructing an institution. We need the traffic code to articulate the precise content of our duties in a defensible way. To drive safely, we must coordinate our conduct, and to coordinate our conduct, we need an articulated system of reciprocal rights and responsibilities. Independent, uncoordinated action, no matter how conscientiously careful, will not enable us to discharge our duties of care. On the one hand, instrumentalism needs law to make abstract obligations concrete and to complete incomplete, if independently identifiable, moral concepts. On the other hand, the law is working in a way that is discernibly formal. It is specifying an antecedent, abstract obligation. It is not simply pursuing a wholly independent end.

When we reflect on the relation of our duties of care to the traffic code, some of the lessons that we learn resonate deeply with the teachings of formalists like Ripstein and Weinrib. Our highly general, abstract duty of care imposes a further duty to create an institution that enables us to discharge that duty. That institution, in turn, makes our abstract obligation concrete. Other thoughts resonate with instrumentalism: we can state the value of safety independent of the traffic code and, in spelling out the code’s provisions, we do need to attend to the diverse interests that bear on its design. We care about safety, but we also care about other desiderata: administrability, convenience, comprehensiveness, relative simplicity and more. Among these interests, safety has a special urgency because of the foundational role that it plays in our lives. It has that special urgency for the contingent, if deep-rooted, reason that Zorzetto emphasizes when she argues that deep metaphysics underlies my normative framework.^[5] We are physically embodied agents, and our bodies are fragile. For every one of us, the physical integrity of our person is a precondition for leading a life where we are, in fact, in charge of ourselves and not dependent on the benevolence of others.

Further thoughts bounce us dialectically back to formalism. We cannot realize the important value (namely, safety) that we can identify independent of the institution without constructing an institution. Constructing that institution changes our moral position. The institution imposes obligations to which we were not previously subject, and which might have been articulated differently. First principles dictate that – to discharge our duties of safe driving – we must coordinate our rights and responsibilities, but first principles do not dictate that we must drive on the right and not on the left. Once we adopt a traffic code, we will be subject to just such an obligation. So, too, once we have a traffic code, we will also have reasons we did not previously have to support and reform the code that we have created. When no 75institution exists – and we need an institution in order to discharge our obligations to one another – we have a duty to create the institution. When an institution exists, we have a duty to comply with its terms if it is reasonably just and are responsible for reforming it to the extent that it is not.

The thought that the traffic code must be something like ‘reasonably adequate’ or ‘reasonably just’ bounces us back towards ‘instrumentalist’ considerations. Not just any institution will do. We need an institution whose terms are as responsive as they can be to the interests and values that the institution serves. The conditions that enable us to be ‘in charge of ourselves’ are partly determined by our interests as physically embodied, vulnerable beings. Our interests must therefore have a say in determining the rules laid down by our institutions. We need both form and substance. On the one hand, our specific duties of safe driving must be plausible incarnations of our abstract obligation of reasonable care. On the other hand, our specific duties must be sensitive to interests we have as embodied agents engaging in activities that play diverse roles in our lives.

One other example, not mentioned in my book, might help to show the basic interplay of form and substance in articulating tort law. Two famous mid-19th century cases – Rylands v Fletcher^[6] and Losee v Buchanan^[7] – sparred over the role that reciprocity should play in dividing tort law’s labor between the domains of negligence and strict liability. Reciprocity of risk is a formal idea. Risks are reciprocal when the plaintiff and defendant impose equal risks on one another. Addressing instances of non-reciprocal risk imposition, Rylands and Losee embraced opposite ways of restoring reciprocity. Losee settled on subjecting non-reciprocal risk impositions to negligence liability. It justified doing so in part by pointing out that negligence liability compensates the victims of non-reciprocal risk impositions by permitting those victims to impose equivalent risks on others (including those who imposed non-reciprocal risks on the victims) without incurring liability for non-negligent harms issuing from that imposition. For Losee, the right to impose increased risks on the original risk imposer is fair compensation for bearing the increased risk exposure created by the original risk imposer’s activities. Rylands, by contrast, concluded that the proper way to restore mutuality between the parties was to subject non-reciprocal risks to strict liability. Losee’s solution invites an increase in the level of mutually imposed risk because it invites people to ‘level up.’ Rylands’ solution tends to damp down the level of risk imposition by subjecting the imposition of non-reciprocal risks to liability.

76The formal concept of reciprocity favors neither solution. Reciprocity obtains when equal risks are mutually imposed: this condition can be satisfied (or not) at different levels of risk imposition. As long as we have, and enforce, a speed limit, the non-negligent risks of the road are equally reciprocal no matter where we set the speed limit. What varies is the level of legitimate risk imposition and the compensation given for the bearing of increased risk when increased risk ripens into harm. If we set the speed limit at 55 miles per hour, we declare the risks created by those who drive at 100 miles an hour illegitimate and compensate with damages awards those who suffer injury at the hands of those who drive at 100. If we set the speed limit at 100 miles per hour, we compensate those exposed to the risks of 100 miles per hour driving by entitling them to impose those risks on others. Both form and substance matter to setting a speed limit. Form matters because equal persons should presumably be entitled to impose, and only to impose, equal risks. Reciprocity of risk is an attractive idea because it expresses a conception of mutuality. Reciprocal risks are, in an important way, interpersonally fair risks. Substance matters because, as physically vulnerable beings, we have an especially urgent interest in safety. We care not just that risks be equal but that they be worth running – that the game be worth the candle in the sense that the threat to our lives posed by taking to the road be less than the good we stand to gain from doing so. We cannot do this without attending to the diverse ends for which we drive and their relative urgency. Emergency vehicles can impose non-reciprocally great risks on the rest of us because their travels serve especially important ends.

The general position that I try to develop in Reasonableness and Risk thus tries to steer a course between the Charybdis of formalism and the Scylla of instrumentalism. In so doing, it falls, I hope, into the shared ground occupied by John Rawls and HLA Hart.

II Professor Papayannis: Conditional wrongs and interpersonal justice

Papayannis subjects my view of strict liability to intense examination and searching criticism. Drawing on his own, independently important and interesting, theory of strict liability, his rich commentary covers a great deal of ground. This reply can respond to only a few of his criticisms. First, Papayannis objects to both the form (conditional wrong) and the substance (duty to compensate) of my view of strict liability because he regards strict liabilities as imposing a very stringent form of responsibility. The paradoxical thought that we are sometimes ‘bound by impossi77ble duties and this is routinely the case in tort law’^[8] is central to his argument. We have duties not to harm that are independent of our duties of care and these justify strict liability. ‘Strict liability rules are the legal response to the violation of a right not to be harmed by the risky (but lawful) activity of the defendant; fault liability is the legal response to a right not to be harmed as a consequence of the defendant’s lack of reasonable care.’^[9] The flip side of this coin is an interpretation of the account that I offer of strict liability wrongs. He understands my account to assert that strict liability wrongs are wrongs whose commission is acceptable so long as one pays for the harm done to others by one’s risky conduct. He objects, that is, that I assimilate strict liabilities to prices. Consequently, I do not have a defensible account of the wrongness to which strict liability responds. Papayannis also thinks that my account of enterprise liability hollows out the institution of tort law to the point where it is drained of interpersonal justice. Enterprise liability responds to accidents as social phenomena and in so doing kicks interpersonal justice to the curb. These are important criticisms. Before I take them up, however, I shall detour briefly to discuss a criticism Papayannis makes of my treatment of sovereignty-based wrongs. I believe that this criticism may be based on a misunderstanding.

III 87Professor Steele: Enterprise liability in the law of negligence

Steele’s exploration of how to run with the ideas of activity liability and the heterogeneity of tort, and to bring these ideas to bear on other aspects of contemporary tort law, owe more to her own expertise and ingenuity than they do to my thinking, but I am very happy to have stimulated her thought. Parallel developments can be found in American negligence law. It, too, ‘is as imbued with issues relating to the world of activities – rather than discrete acts – as is strict liability.’^[25] Indeed, the influence of enterprise liability conceptions on negligence law is almost as old as the emergence of enterprise liability itself. Shortly after the enactment and rapid spread of the Workmen’s Compensation Acts, a Harvard Law Professor by the name of Jeremiah Smith argued that – by both withdrawing a major site of accidental harm from governance by the law of torts and repudiating the fault liability around which late 19th century tort law had been constructed – these Acts made tort law a house divided against itself. The Workmen’s Compensation Acts were therefore bound, in Smith’s view, to spark a revolution in the common law of torts. The revolution that Smith foresaw involved not only the spread of strict liability, but also the reconstruction of negligence law. Quite rightly, Smith saw that negligence liability itself could be recast as a much stricter form of liability by relaxing the requirements of the prima face case, construing res ipsa loquitur liberally, and inverting the burden of proof.^[26] Many of these developments have long since come to pass.^[27]

In the United States, as in England, over the course of the 20th century, negligence law was extensively renovated. Enterprise liability ideas guided many of the renovations. In an important article now nearly thirty years old, Robert Rabin cataloged diverse ways in which enterprise liability conceptions can also permeate fault-based doctrines.^[28] Anticipating some of my response to Papayannis in this Reply, Rabin argued that the idea of enterprise liability was much more than a conception of strict liability and, indeed, was better understood as the expression of a shift ‘from a corrective justice perspective on responsibility in tort law to a collective justice approach.’ Although I would describe this shift a bit differently – to 88emphasize the move from corrective to commutative justice and the move from the ‘world of acts’ to the ‘world of activities’ – I agree with the basic point. As both Rabin and Steele say, the legal transformation effected by enterprise liability is broader than, and significantly distinct from, any shift toward strict liability. In England, there has been no common law shift toward strict liability and yet enterprise liability has made its presence felt. As Steele says, enterprise liability marches on even when strict liability does not.^[29]

Examples of how modern tort law absorbs much of enterprise liability’s logic even when it adopts fault rules appear across a wide range of tort law’s domains. For example, when courts single out psychiatrists for special duties to warn potential victims of their dangerous patients, they are relying on the character of the psychiatric enterprise to justify moving from a regime of ‘no duty’ to a regime of fault.^[30] The Tarasoff duty to warn rests on the premise that psychiatrists are engaged in a special kind of expert activity which makes them uniquely well situated to identify those persons who are especially likely to inflict violent injury on others. The character of their activity plays a central role in justifying their special responsibility. The duty attaches to a role that is institutionally defined. Conversely, when a court relies on the amateur status of a man who constructed a home for his own use to justify not imposing a duty of careful construction to a subsequent purchaser – to justify moving from a fault regime to a regime of ‘no-duty’ – the court is also acting on an enterprise liability logic. Individual, amateur masons who are neither trained nor engaged in the enterprise of masonry can neither reduce the risks of defective construction nor disperse those risks across the enterprise of homebuilding in the way that professional masons and real estate developers can. Individual, amateur, masons still live in the ‘world of acts.’ They are not incarnations of organized activities.

Other examples of enterprise liability conceptions informing developments in negligence law include the slow but steady trend in medical malpractice law to move ‘to a more robust principle of fault in medical malpractice cases’^[31]by abandoning the same locality rule, using res ipsa loquitur more expansively, adopting less restrictive standards for qualifying experts, and recognizing robust informed consent claims. When courts make these changes, they are in part effecting enterprise liability ends by focusing on the risk-reducing and risk-spreading capacities of 89the organizations that provide health care.^[32]Changing domains, enterprise liability ends are also being effected when courts abolish charitable immunities on the ground that they are no longer needed to induce the provision of important services because the charitable institutions involved can insure against any liability.^[33] Two recent California Supreme Court decisions imposing affirmative duties to act on institutional actors also show the influence of enterprise liability logic.^[34] In both cases, the Court recognized duties to protect parties with whom the institutions had special relationships from harm at the hands of third parties. The institutional capacities of the respective defendants UCLA and USA Taekwondo played indispensable, if background, roles in the justification of those duties of care. The institutional (or enterprise) actor was not the principal wrongdoer, but the enterprise stood in an enabling relation to the intentional wrongs of third persons. Both cases illustrate a larger point of Steele’s: the cutting edge of duty doctrine now addresses ‘[t]he role played by parties who are not necessarily risk creators’ in a primary sense.^[35]

There is every reason to expect the influence of enterprise liability conceptions, broadly understood, to continue and to increase. The most obvious reason why is probably also the most important. We live in a world of institutions and organized activities. These create perils and possibilities unknown in the ‘world of acts.’ Tort law must respond to the perils, and the best responses involve making use of the possibilities for superior precaution that arise hand-in-hand with those perils because the institutions that now dominate our landscape have historically unprecedented and extensive capacities and competencies. Nonetheless, something under-appreciated is unfolding. Steele is right to call the subterranean role of enterprise liability within negligence law to our attention. The phenomenon tends to fly under the radar, and it does so because it confounds our expectations. Enterprise liability 90logic presses toward strict liability, and strict liability is an alternative to, and competitor of, fault liability.

Why, then, has enterprise liability ceased moving forward in the guise of strict liability, but continued to do so in the guise of fault liability? Part of the answer, I suspect, is that the long march of American tort law in an enterprise liability direction was brought to an abrupt end by the insurance crisis of the mid-1980s. Thwarted by that crisis, some of the momentum behind enterprise liability may have been deflected into negligence law. But there is also a fundamental limitation affecting the practical implementation of strict liabilities that pushes enterprise liability conceptions towards negligence incarnations. All accidents arise at the intersection of two or more activities. In some circumstances, it is impossible for the common law of torts to attribute responsibility for an accident to one of the parties to it without employing some criterion of fault.^[36] Highway accidents are the canonical case.^[37] In the absence of norms – usually statutes– specifying duties of precaution, rights of way, and so on, it is often impossible to attribute responsibility for accidental injury. In the absence of crosswalks, we may not be able to say if a pedestrian or a driver was responsible for an accident between the two. In the absence of rules ordering priorities among vehicles at four-way intersections, we may not be able to say whose activity is responsible for an accident between two cars at such an intersection.

Highway accidents are (to my mind) somewhat atypical in that acute problems of attribution arise even in cases where accidents arise at the intersection of only two activities – when only two vehicles are involved, or only one vehicle and a pedestrian. Problems of causal indeterminacy tend to be more acute when accidents arise at the intersection of multiple activities. When a bus bearing schoolchildren is struck by a train at a railroad crossing, for example, the accident arises at the inter91section of numerous activities. The railroad contributes to the accident by its design and operation of both the crossing and the train. Surface grade crossings can be replaced by underpasses and overpasses; warning bells and horns can be more and less piercing; gates can bar access to the tracks more and less effectively; the schedule and maintenance of the train can affect the incidence of accidents; and so too can the selection and training of the engineers who operate the trains. The traffic department of the city can affect the incidence of accidents by its designing of roads, routing of traffic, and timing of lights. The school department can affect the incidence of accidents by its choice of routes and its selection and training of drivers. The manufacturer of the bus can, by its choice of design, affect the ease with which accidents can be avoided and the severity of the injuries that they cause. Buses can be more or less maneuverable, more or less soundproof, and more or less crashworthy.

It can be extremely difficult and even impossible to devise non-fault attribution rules – such as the ‘scope of the employment’ test for vicarious liability or the ‘scope of the risk’ test for abnormally dangerous activities – to apportion responsibility for accidents that arise at the intersection of multiple activities.^[38] Negligence norms must often be deployed by default. Fairness may require that each of these enterprises – railroading, designing automobiles, transporting schoolchildren, designing traffic systems – bear its characteristic accident costs, but complexity tends to defeat the identification of characteristic risks. Fault liability therefore tends to expand beyond the ‘world of acts’ to which the logic of enterprise liability might wish to keep it confined.

Whatever its sources, though, the subterranean role of enterprise liability in contemporary negligence law is a role that we should welcome. It is a fitting response to the world of activities and institutions in which we live and, as cutting-edge duty decisions show, negligence doctrine is a flexible instrument. It can be used to make incremental and intelligent adaptations of the law.

IV 92Professor Zorzetto: From metaphysics to political morality

Zorzetto’s thoughtful commentary raises deep and difficult questions.^[39] The concept of reasonableness is at the center of her commentary, as it is of my book. Early in her commentary, Zorzetto suggests that: ‘Keating’s concept of reasonableness possesses significant explanatory power, even for those who do not subscribe to Rawlsian thought or who are not aligned with the political liberalism developed from Mill and Rawls’ theories.’^[40] Relatedly, she suggests that my approach is rooted in metaphysically basic features of the human condition, namely, that we are vulnerable and that we interact. In both cases, I agree. Reasonableness is a shared concept, not a technical construct. And tort law, as Ernest Weinrib has insisted, responds to a very basic aspect of the human condition, namely, the unity of doing and suffering.^[41] In the crowded and bustling world in which we live, we inevitably run afoul of one another. Zorzetto writes, ‘Given that interferences are inevitable and cannot be wholly or abstractly eliminated, the central issue is determining which interventions are justified and which are not.’^[42]

I suspect that most, but not all, theorists of tort agree that the subject matter of the field, if elusive in some way, connects to a fundamental feature of human existence. But views of the feature do differ. For economic analyses, tort is a domain in which we must decide how to resolve conflicts that arise between competing uses in a way that puts scarce resources to their highest uses.^[43] And theorists extract very different theories from their views of tort law’s metaphysics. The physical vulnerability of persons does not play a central role in either Ripstein’s or Weinrib’s theories, for instance. In light of pervasive disagreement as to how tort law engages basic features of human existence, we need to build from metaphysics into moral and political theory. We cannot lose touch with the metaphysical circumstances that give rise to tort, but we also cannot merely work from the fact that human interaction is such that 93our actions and activities inevitably interfere. Even to spell that general idea out a bit more precisely by saying that tort law addresses the fact that our actions and activities pervasively threaten to impair each other’s interests and persons is to move beyond common ground. On the view that the book advances, the role of those torts that form the historic core of the subject is to secure basic preconditions of effective agency by reconciling liberty and security on terms that enable people to lead independent and decent lives. This is one way of sharpening our understanding of the basic situation that tort confronts. As with other ways of sharpening our understanding of the basic situation that requires a law of torts, this way of doing so is contestable.

Zorzetto is also right to emphasize that reasonableness as a concept is not the particular property of Rawlsian or contractualist theories. On one eminently plausible view, reasonableness is a ‘buck-passing’ concept in negligence law. It simply directs our attention to the question of whether the defendant acted justifiably. That, in turn, is a moral question, full stop.^[44] Basic moral questions, and basic moral concepts, are not the property of any one philosophical tradition. My theory does not mean to deny this. The belief that underlies my arguments is that we can develop the idea of reasonableness in tort law – and understand tort law’s preoccupation with harm, especially physical harm – in a perspicuous way by drawing on the resources of liberal political philosophy in its Millian and Rawlsian incarnations. Mill is particularly useful because harm is fundamental to tort, and Mill made harm a topic in and for political philosophy. Mill’s work founds a tradition that conceives of harm as the impairment of powers of agency. This idea resonates powerfully with the modern American law of torts. Rawls is particularly useful because he theorizes reasonableness in an important and illuminating way. Acting reasonably is about acting in a way that you can justify to others. It is not about showing that you are acting justifiably by showing that you are pursuing the general good.

Getting a grip on the intrinsic, negative significance of harm for the physically embodied, vulnerable agents that we are, and building up from the shared moral idea that reasonableness implicates a particular way of justifying conduct, helps us to understand not just the artificial reason of the law, but also the ordinary moral judgments made by jurors and other lay people about tort lawsuits. Those judgments are often at odds with the judgments of economic and technocratic experts. In the United States, plaintiffs’ lawyers argue, as one plaintiff’s lawyer did in a case involving a defective door latch in a Chrysler minivan, that the tortious wrong in much corporate conduct has its roots in a perverse conception of rationality:

94Chrysler officials at the highest level cold-bloodedly calculated that acknowledging the problem and fixing it would be more expensive . . . than concealing the defect and litigating the wrongful death suits that inevitably would result.^[45]

Not infrequently, this kind of charge hits home with juries, prompting explosions of outrage and awards of punitive damages so large that the national media takes note.

When the moment passes and the outrage subsides, some ostensibly level-headed commentator suggests that such indignation is misguided and, in fact, foolish. Everyone – people as well as corporations – makes tradeoffs involving safety. Kip Viscusi, an eminent economist, crisply states the standard argument:

We take chances all the time. We ride in motor vehicles, fly on planes, eat potentially risky foods, and live in an environment that is not risk free. Some tradeoffs of this kind are inevitable as we seek to strike an appropriate balance between the harm inflicted by risks and the benefits such activities offer for our lives.

Viscusi’s point is clear. ‘The plaintiff’s attorneys portrayed Chrysler as a corporate villain simply for undertaking a risk analysis.’^[46] Yet ‘risk analysis’ – balancing costs and benefits against each other to determine just how much safety is worth purchasing – is the only rational way to make safety tradeoffs. The value of safety is not infinite, and its pursuit consumes scarce resources that might be put to use elsewhere. No matter how highly we value safety, the benefits of achieving a particular level of safety must be traded off against the costs of doing so. And the rational way to make these tradeoffs is to balance costs and benefits in the way which nets us as much value as possible. It is foolish and sentimental to express outrage when we discover that corporations make just the kind of rational decisions that they should make.

Viscusi’s position is the standard economic position, and the dominant position even today in American policy discourse. One of my aims in my book is to show that this standard economic position is wrong. Once we understand the special negative significance of harm, and once we follow tort law’s lead in understanding interper95sonal justification as a matter of reasonableness, we understand the sense in both tort law and lay jury judgments that insist on levels of precaution that are irrationally stringent from an economic point of view. Zorzetto and I are very much on the same page here, I think. The intrinsic negative significance of harm arises because we are embodied, fragile, physically vulnerable beings.^[47] Physical harm is especially bad for us because it impairs essential conditions of human agency. Physical harms – death, disability, disease and the like – rob their victims of normal and foundational powers of action. Physical harm comes close to being unconditionally bad. Impairing the normal capacities of physical agency impedes the pursuit of a wide range of human ends and aspirations and denies normal human lives to those whose powers are impaired. Few benefits, by contrast, are unconditionally good. Benefits enhance lives, and their power to do so depends greatly on the details of the life in question. Whether a particular benefit is valuable for a particular person depends greatly on their aspirations and projects. Harms and benefits thus are not symmetrical in the way that economics takes costs and benefits to be. And the asymmetry of harm and benefit is not an irrational bit of sentimentality, but a justified ranking of phenomena that have very different impacts on the basic conditions of our agency. Showing that we have good reason to regard harms as especially bad for us, for its part, goes a long way to showing that stringent norms of precaution, such as the feasibility and safety norms, are justified norms of precaution. Balancing harm and benefit is inevitable when risk is at issue, but balancing which reflects the priority of avoiding harm is called for.

Giving a contractualist, Rawlsian inflection to reasonableness enables us to flesh out the contrast between reasonableness and rationality, and to understand justified precaution as something other than socially efficient precaution. My book takes tort law’s rhetoric of reasonableness as a signal that the subject takes our relations to one another to be a matter of morality not prudence. Relatedly, I think that there is more than sentimentality and squeamishness behind jury aversion to cost-benefit analysis of risks to life and limb. I take tort law’s relentless use of the word ‘reasonable’ to intimate an implicitly deontological outlook. Tort is about what we owe to each other in the way of coercively enforceable obligations not to impair or interfere with each other’s urgent interests as we go about our lives in civil society. These claims have their roots in the basic metaphysics of the problem that tort law addresses, but to have a conception that can both shine light on our law and guide our development of it, we must build up from basic metaphysics. Without that 96kind of building, we would have the basic metaphysical situation on the one hand, the details of the law on the other, and a large gap between the two.

In short, Zorzetto is right to think that my view has roots in something deeper than contemporary political theory. As she says, implicitly, my view reaches down to the metaphysics of human agency.^[48] But my view also supposes that we must build up from the metaphysics of human agency into history and political theory. We must build into history because our tort law is the contingent product of history. And we must move beyond the metaphysics of human agency into moral and political theory because the metaphysical circumstances that call tort into existence can be cited in support of very different views. The basic circumstance that calls tort law into existence – the inevitability of conflict and collision between our lives and our projects – can be cited in support of the view that tort law is an autonomous realm of corrective justice as well as in support of the view I develop in my book that tort is a part of the basic justice that we owe to one another. No doubt, it can be appropriated by other views, too. To choose among competing views, we need to develop, and then compare, the more complex moralities they embody.

Even so, Zorzetto is right to remind us that the predicament to which tort law responds goes very deep indeed. We lose our bearings when we lose sight of this fact.