5.1. Introduction

With the elucidation of Kantian right, the main theoretical components for understanding private law are in place. Starting from the premise that private law is an exercise in justificatory coherence that can be understood from its own immanent perspective, I have set out three mutually supporting ideas. The first, formalism, constitutes the methodological framework for integrating the unity, the kind, and the character of a juridical relationship. The second, corrective justice, is the form of the private law relationship. The third, Kantian right, allows one to trace corrective justice back to its normative roots in self-determining agency and forward to the values and institutions of a coherent legal order.

I now want to illuminate the inner workings of the private law relationship by following the implications of Kantian right into the interior of corrective justice. In Aristotle’s presentation of corrective justice, the correlativity of gain and loss is the organizing feature of liability. But what exactly is the meaning of this correlativity? How do we identify the gains and losses and see them as expressions of Kantian right? And how do the main grounds of liability—tort, contract, and unjust enrichment—reflect the correlativity of gain and loss that Aristotle mentions?

In dealing with these questions, I first distinguish between two aspects of gain and loss, the factual and the normative. In their factual aspect, gains and losses refer to changes in the condition of the litigants’ holdings; in their normative aspect, gains and losses refer to discrepancies between what the parties have and what they should have according to the norm governing the parties’ interaction. My argument is that the gains and losses must be understood as normative, not factual, and that we can therefore identify the correlativity of this normative gain and loss as a correlativity of Kantian rights and duties. I then consider the objection that in postulating a normative rather than a factual correlativity, Kantian right ignores the particularity and concreteness of human welfare. My answer is that Kantian right does not ignore human welfare but sees it from the moral perspective appropriate to the bipolar interaction of free and purposive beings. Finally, I turn to an examination of how the structure of liability in tort, contract, and unjust enrichment reflects the correlativity of right and duty.

5.2. The Two Aspects of Gain and Loss

To clarify the meaning of gain and loss under corrective justice, I propose a distinction between what we might call the “factual” and the “normative” aspects of gains and losses. Suppose, for example, that you negligently injure me. A comparison of my present and my previous condition reveals that I am materially worse off than I was before. This is the factual aspect of the loss. In addition, however, I am also worse off than I should be, given the norm against negligent injuring. The loss considered from the standpoint of the relevant norm is the normative aspect of the loss.

Analytically, the normative and the factual aspects are distinguishable, although both may be present in a particular case. Thus, in the example of negligent injury, the injury that makes me materially worse off than I was before is also the injury that I should not have suffered, given the norm against negligence. If, however, you innocently injure me, the loss I suffer has a factual but no normative aspect.

The factual aspect of gain and loss refers to the effect of the interaction on the amount or condition of one’s holdings (broadly construed to include both one’s body and the external objects at one’s disposal). In its factual aspect, a gain is a change for the better from the standpoint of the person whose holdings are increased; a loss is a change for the worse.

The normative aspect of gain and loss derives from the justificatory function of corrective justice. As a justificatory structure, corrective justice embodies norms that set the terms of fair interaction. In their normative aspect, gains and losses involve a comparison between what one has and what one should have through the operation of those norms. A normative gain occurs when one’s holdings are greater than they ought to be under those norms; a normative loss occurs when one’s holdings are smaller than they ought to be. Taking our cue from Aristotle’s treatment of gains and losses as the just grounds for a court’s taking from one party and giving to another, we may say that a person enjoys a normative gain when there is justification for the law’s diminishing his or her holdings, and that a person endures a normative loss when there is justification for the law’s augmenting his or her holdings.1

The two aspects reflect different conceptions of the baseline for the characterization of gain and loss. The baseline for factual gains and losses is the preexisting condition of one’s holdings: a transacting party whose holdings have increased or improved as a result of the transaction has realized a factual gain; a party whose holdings have decreased or deteriorated has suffered a factual loss. The baseline for normative gains and losses is one’s due under the justifications that obtain within corrective justice: a gain is an excess over, and a loss a shortfall from, one’s due.

The distinction between factual and normative gains and losses allows the possibility that a gain or loss of one type may be unaccompanied by a gain or loss of the other. All the possible combinations are recognized in sophisticated systems of private law. A party may realize: (1) a normative gain but no factual gain: if I negligently injure another, I have acted wrongly but no holding of mine has been improved by the wrong; (2) a factual gain but no normative gain: if another mistakenly paves my driveway without my knowledge, the condition of my holdings has been improved, but I owe the improver nothing; (3) a normative loss, but no factual loss: if someone trespasses on my property without impairing its condition, a common law court may award me nominal damages to mark the breach of a norm, despite the absence of actual damage; (4) a factual loss, but no normative loss: if someone injures me without fault, I generally cannot recover despite the impairment of my physical condition.

Given the two aspects of gain and loss, what precisely has to be correlative to what if there is to be liability under corrective justice? We can eliminate the possibility of a factual loss correlative to a normative gain or of a factual gain correlative to a normative loss. The logic of correlativity requires that what is predicated of one element in the pairing be also predicated of the other. Accordingly, the gains and losses must be of the same kind. We are then left with the question of whether corrective justice features a correlativity of factual gain to factual loss, or of normative gain to normative loss.

That the gains and losses of corrective justice are normative is evident from the equality that constitutes their baseline. This equality is not itself factual: as discussed in Chapter 3, it does not refer to an equality in the amount or condition of the parties’ holdings. Rather, equality is a formal representation of the norm that ought to obtain between doer and sufferer. Action that conforms to this norm, whatever it is, maintains the equality between the parties, so that no complaint is justified. Action that breaches this norm produces a gain to the injurer and a loss to the person injured. Then the court, “justice ensouled” in Aristotle’s graphic phrase, restores the parties to the equality that would have prevailed had the norm been observed. The normative nature of the equality indicates that the variations from that equality are also normative.

This conclusion accords with corrective justice’s being a justificatory structure. The gains and losses have the same character as the structure they define: they refer to the norm that figures in the process of justification. Accordingly, gain and loss are the excess over and the shortfall from one’s due.2

The modern common law confirms this conclusion by having liability reflect the correlativity of normative, rather than factual, losses and gains. Tort law allows recovery where factual loss is unaccompanied by factual gain; and the law of unjust enrichment allows recovery where factual gain is unaccompanied by factual loss.

In tort law, the plaintiff typically complains of the factual loss of injury, but only exceptionally has the defendant realized a corresponding factual gain from the wrong. If tort law insisted on factual correlativity, liability would be clear only for takings of property, where the taker’s holdings are increased and the victim’s diminished to the same extent.3 Negligence would be excluded from liability, because the plaintiff’s suffering of an unintended injury in no way improves the situation of the defendant. Normative correlativity, however, is present even in negligence liability, because the wrongful injury represents both a normative surplus for the defendant (who has too much in view of the wrong) and a normative shortfall for the plaintiff (who has too little).4

The converse case of factual gain without a corresponding factual loss arises under the law of unjust enrichment. Suppose the defendant profits through the unauthorized use of the plaintiff’s asset, and the plaintiff claims a part of the profit or, at least, the defendant’s saving in not having to rent such an asset. If the plaintiff did not intend to use the thing during the period of the defendant’s use and if the property was returned unimpaired at the end of that period, we have a situation where the defendant has gained but the plaintiff has realized no correlative factual loss. The modern treatment of unjust enrichment in the common law grants restitution in such a case.5 Because of the absence of factual loss, such results accord with corrective justice only if the correlativity of loss and gain is normative. Then the restitutionary requirement that the defendant’s gain be “at the plaintiff’s expense” can be understood normatively, as referring to what has been called the defendant’s “subtraction from the plaintiff’s right to the exclusive enjoyment of the property.”6

The conclusion that gains and losses of corrective justice are normative rather than factual ought to dispel a common error in contemporary thinking about corrective justice. The requirement of correlativity is sometimes thought to limit or undermine the Aristotelian conception of corrective justice, on the ground that if gain and loss are to be correlative, they must necessarily be equal, which is obviously not the case for the factual gains and losses resulting from negligence.7 However, that factual gain does not necessarily equal factual loss is true but irrelevant. What matters to the justificatory structure of corrective justice is the correlativity of normative, not factual, gain and loss.8

Indeed, once one draws the distinction between the normative and the factual aspects of gain and loss, the conclusion that corrective justice involves a correlativity of the normative aspect can scarcely be in doubt. In the Aristotelian account, gain and loss are a way of representing the occurrence of an injustice that needs to be rectified through liability. Injustice and liability reflect the violation of certain norms, not the existence of certain facts. Even in the case of a taking, where the appropriator’s gain is factually equivalent to the victim’s loss, liability follows not from the fact of this equivalence, but from the breach of a property norm. Where gains and losses are solely factual (as in the case of a new enterprise that takes customers away from a competitor), there is no liability under corrective justice.

To sum up: The correlative gain and loss of corrective justice does not point to a factual loss and a corresponding factual gain. What matters is whether the transaction can be regarded as yielding a normative surplus for the defendant and a normative deficit for the plaintiff. Therefore, liability for a deterioration in the condition of the plaintiff’s holdings is predicated not on some parallel improvement in the condition of the defendant’s but on the defendant’s having unjustly inflicted that loss. Similarly, the plaintiff recovers the defendant’s gain not when the plaintiff has suffered merely a factual loss but when the defendant’s enrichment represents an injustice to the plaintiff.

5.3. The Correlativity of Normative Gain and Loss

Let me next consider what it means for normative gains and losses to be correlative. As the dynamic idea that links gain and loss, correlativity structures the normative content of corrective justice. To this point I have described normative gain and loss in general terms as excess and deficiency from the standpoint of a justificatory consideration, but I have not specified any particular justificatory consideration. Correlativity enables us to discern what justificatory considerations are possible because the only considerations that play a role in corrective justice are those that can be applied correlatively.

Here I move toward a particular set of justificatory considerations by following through on the structural implications of correlativity. My procedure is formalist: instead of positing substantive norms, I look to the conditions that any norm must fulfill if it is to conform to the dimension of correlativity. By progressively refining the considerations that might satisfy these conditions, I hope first to eliminate justificatory considerations that do not fit, and then to provide an increasingly definite account of the operation of Kantian right within corrective justice.

5.3.1. The Conditions of Correlativity

To satisfy the dimension of correlativity, the justificatory considerations at work in corrective justice must be unifying, bipolar, and expressive of transactional equality. They must be unifying in that for normative gain and normative loss to be relative to each other, the same norm must be the baseline for both. They must be bipolar in that because one party’s normative gain is the other’s normative loss, the justificatory considerations must link two, and only two, parties. And they must be expressive of transactional equality in that by being equally applicable to the party realizing the gain and to the party suffering the loss, they accord a preferential position to neither.

Consequently, a justificatory consideration that fits into the normative structure of corrective justice cannot have a justificatory force that reaches only one of the parties. Such a one-party consideration could account for the normative effect—be it gain or loss—on that party alone. But because the correlative normative effect on the other party lies beyond its scope, that consideration does not supply the single baseline for both gain and loss; nor does it establish a bipolar link between the gainer and the loser; nor does it express their equal normative standing in the transaction.

For instance, the idea that compensation is a goal of tort liability9 cannot fit within corrective justice. Because compensation reposes its justificatory force solely on the plaintiff’s exigence in the aftermath of injury, it applies to the plaintiff independently of the defendant. It therefore cannot serve the unifying function of supplying the baseline for both the defendant’s gain and the plaintiff’s loss. Moreover, the need for compensation does not forge a bilateral link between the injured party and any particular injurer. Instead, it relates the injured party to others (however many there are) who are similarly exigent and thus have similar claims as a matter of distributive justice. Finally, the plaintiff and the defendant do not have equal standing under this justificatory consideration—indeed, the defendant has no standing at all. In looking to the injured person’s exigence, the justification is intrinsically preferential to plaintiffs. Basing liability on it would violate, not vindicate, the parties’ equality.

Accordingly, the compensation rationale does not yield a correlative gain and loss. It is true that being injured might be thought of as a normative loss consisting in the shortfall from the plaintiff’s entitlement under the goal of compensation. But the defendant, whose position is unaffected by that justificatory consideration, could not be said to realize a gain with respect to it. From the standpoint of the compensation rationale, the plaintiff’s loss lacks bipolar significance. At most, the loss justifies improving the plaintiff’s situation; it does not state a ground for taking something from the defendant. Accordingly, the compensation rationale does not support the defendant’s liability to the plaintiff.

The same applies, mutatis mutandis, if the parties’ positions are reversed. Suppose a justificatory consideration applies to the defendant independently of the plaintiff (as when deterrence is taken to be a goal of tort law). Then the defendant’s holdings might be excessive in the light of that consideration, so that unless penalized by the law the defendant would realize a normative gain. However, because plaintiffs do not come within the reach of that justificatory consideration, they cannot be said to have less than they are entitled to under it. There is a normative gain with no correlative normative loss. Consequently, the justificatory consideration would provide no reason for the defendant’s liability to the plaintiff.

Nor can correlativity be satisfied by an accumulation of considerations that apply separately to the two parties but embrace them both only when taken together. (An example is the explanation of tort liability by reference to the need both to deter actors and to compensate sufferers.) To be sure, such an explanation produces a normative gain for the defendant and a normative loss for the plaintiff. But because the reason for thinking the defendant to have gained is not the same as the reason for thinking the plaintiff to have lost, the gain and the loss are not normatively correlative. The considerations may justify eliminating the gain and the loss through separate operations that decrease the holdings of the defendant and increase those of the plaintiff. But they do not justify the direct legal link of liability in corrective justice.

5.3.2. Right and Duty

In contrast to the inadequacy of such one-sided goals, Kantian right fleshes out corrective justice in a way that satisfies the requirement of correlativity. The fundamental principle of Kantian right, that one person’s action be united with the other’s freedom in accordance with practical reason, treats the relationship between the parties as unified, bipolar, and transactionally equal. Unity is present because the concept of right integrates into an idea of reason the juridical constituents of the parties’ relationship. Bipolarity obtains because of the focus on the relationship between the one person’s action and the other’s freedom. And the transactional equality is manifest in the equal moral status of the interacting parties as free and purposive beings.

The specific form of correlativity within Kantian right is that of right and duty. The requirement that one’s action be consistent with the other’s freedom means that every actor is obligated not to violate the rights of others. In Kantian theory, rights are the juridical manifestations of the freedom inherent in self-determining agency. An act is consistent with another’s freedom when it is compatible with that person’s rights. Thus having a right implies that other actors are under the moral necessity to refrain from infringing it. In Kant’s words, rights are “(moral) capacities to put others under obligations.”10

The Kantian approach, then, links the interacting parties through a right, on the one hand, and a corresponding duty, on the other. The right represents the moral position of the plaintiff; the duty represents the moral position of the defendant. Right and duty—and therefore plaintiff and defendant—are connected because the content of the right is the object of the duty. The Kantian principle that the action of one party be united with the freedom of another in accordance with practical reason can be explicated in terms of right and duty as follows: that principle is satisfied when the action of one party does not violate a duty, when the freedom of the other manifests itself in a right, and when the reason inherent in free and purposive agency grounds both the right and the duty and connects one to the other.

In making Kantian right more determinate, this conjunction of right and duty maintains the shape appropriate to the correlativity of corrective justice. The relationship of the parties is unified because both right and duty are expressions of the same notion of practical reason. Moreover, the direct connection between the obligee of the duty and the holder of the right makes the relationship a bipolar one: although a single right can generate a duty in many persons, each such duty is separately correlative to the right that generates it. Finally, transactional equality is assured because the holder of the right and the obligee of the duty have equal standing as free and purposive beings.

5.3.3. Correlativity as Articulated Unity

Given that the conjunction of right and duty satisfies the conditions of unity, bipolarity, and equality, I must now consider a further question: how do right and duty operate as correlatives? To put the question more accurately: how do these conditions manifest themselves in the internal workings of a juridical relationship marked by the correlativity of right and duty?

When right and duty operate as correlatives, they constitute an articulated