LAW'S EXPLANATORY FIRST VIRTUE
John Rawls begins and ends his magnum opus with the
Priority of Justice for Social Institutions
Justice is the first virtue of social institutions
which he treats as a commonsense conviction or endoxon. Footnote 1 However plausible this view might be, it stands in tension with the obvious fact that only some social institutions have anything special to do with justice. Given that many have nothing special to do with justice, why think that justice is the first virtue of social institutions as a whole? Indeed, why think anything is? But even if you are skeptical about the Priority of Justice for Social Institutions, you might still be attracted by the weaker
Priority of Justice for the Law
Justice is the first virtue of the law.
After all, we call our judges justices, the government's legal department the Justice Department, and the administration of law the administration of justice. The law enjoys a near-universal association with justice that most other social institutions do not seem to enjoy. However plausible the Priority of Justice for Social Institutions is, restricting its scope to the law can only make it more plausible.
What does it mean to say that justice is the first virtue (or point or principle) of the law? The obvious possibility is to regard justice as a straightforward moral virtue or principle that applies to the law but is entirely independent of it. This is how Rawls glosses the Priority of Justice for Social Institutions. For him, justice is a deontological constraint on social institutions. If social institutions fail by the external moral standards of justice, they must be revised even if they excel by certain other external moral standards, such as efficiency. Similarly, we could assert the
Priority of Justice for the Law (Constraint)
Justice is a deontological constraint on the law.
There are, however, reasons to avoid this interpretation. First, it fails to make the Priority of Justice for the Law more plausible than the Priority of Justice for Social Institutions. If justice is a deontological moral constraint independent of the law, it must surely apply to other social institutions. Why is not restricting justice's priority to the law arbitrary?
Second, even if you believe deontological constraints exist, and that justice is a deontological constraint on the law, this hardly explains why it is law's first virtue. For there can be some other deontological constraint(s) on the law, such as humanity or fraternity. To cash out justice's priority, we would have to assert the
Priority of Justice for the Law (Unique Constraint)
Justice is a unique deontological constraint on the law.
If social institutions fail by the external moral standards of justice, they must be revised even if they excel by any other external moral standard. But as long as we take justice seriously and treat it as one distinct moral virtue among others, it is not clear why the law ought to give justice more importance than humanity or fraternity.Footnote 2 From a purely moral perspective, it is simply not clear why the law ought to be just above all.
There is, however, another and, in my judgment, more rewarding way to read the Priority of Justice for the Law. In this interpretation, what makes justice law's first virtue is not some general priority it enjoys relative to the other moral virtues. Rather, it is the social institution of the law that gives justice priority over the other moral virtues. There is something about that institution—its distinctive structure, features, or nature—that singles out justice for special treatment in a way other social institutions do not. So understood, the Priority of Justice for the Law is not simply a claim in applied ethics but a largely explanatory claim about a social institution. Consider the
Priority of Justice for the Law (Explanatory Virtue)
Justice is the first explanatory virtue of the law or legal system. Distinctive institutional facts about the law give more importance to justice than to any other moral virtue or principle. Ascribing this moral prioritization to the law explains the law in a way ascribing to it some other (or no) prioritization does not.
This requires charitable disambiguation. Obviously, it must not be taken to deny that any particular law or legal system can be unjust or prioritize some moral virtue other than justice. The thought is that there is some general legal perspective—the perspective of law as an institution—from which it is more important that the law excel by the standards of justice than by any other moral standards. When the law is unjust or prioritizes some virtue other than justice, it fails not simply by the standards of morality but also by its own standards.
Second, the view need not assume that law's nature is immutable; it concerns the institution of the law as we know it. We must allow for the possibility that the law changes its nature. The thought is that the modern legal systems with which we are familiar share enough features to warrant the generalization that justice is the first explanatory virtue of the law.
Finally, we probably should not treat the explanatory priority of justice as a brute institutional fact. As I read it, the Priority of Justice for the Law presupposes a metaethics that allows institutional facts to constitute moral facts. When general institutional facts about the law prioritize justice, this gives the demands that justice makes on the law a genuinely moral priority. So when particular laws or legal systems are unjust, they are open to at least one moral criticism to which other immoral laws or systems are immune. On this view, the priority justice enjoys in the law is both moral and legal. To use a familiar metaphor, it is a matter of law's internal morality.Footnote 3
(Is there such thing as an explanatory moral virtue? I do not know for sure. The notion is not unproblematic but it is not obviously incoherent, either. Nor is it unfamiliar. In particular, John Gardner, who has recently taken up the relation between law and justice, seems to understand the Priority of Justice for the Law in much the same way as I do.Footnote 4 Much of my discussion will respond to Gardner's arguments, which appeal to various institutional facts about the law. Since his arguments make very good sense to me, I have no difficulty assuming that the notion of an explanatory moral virtue makes [or can be made to make] sense. Here I will simply take this notion on board sympathetically and propose claims and arguments that employ it. To the extent these arguments engage, we will have some indirect support for my optimism. The proof of the pudding is in the eating).
AGENDA
I suggest we reject the Priority of Justice for the Law in favor of the weaker
Priority of Justice for the Courts
Justice is the first virtue of the law courts.
Here is a rough road map. I start by presenting a venerable conception in which justice concerns intersubjective comparisons. I reject the claim that comparative justice is law's first virtue on the grounds that some particularly important (“primary”) legal duties have nothing special to do with such comparisons. I then entertain the weaker claim that comparative justice is the first virtue of the law courts. I argue that it is vulnerable to a counterexample.
Partly to complete my case against the Priority of Justice for the Law and partly to rescue the Priority of Justice for the Courts, I introduce another venerable conception of justice in which it concerns the requital of wrongs with sanctions. I claim that since primary legal duties are no more requitative than they are comparative, the argument that defeats the Priority of Justice for the Law under the comparative conception of justice also defeats it under the requitative one.
A lot happier is the result of applying the requitative conception of justice to the Priority of Justice for the Courts. I argue that, unlike the comparative conception of justice, the requitative one is capable of sustaining that weaker thesis about the relation between law and justice.
The essay's final challenge is to resolve the tension between accepting the Priority of Justice for the Courts and rejecting the Priority of Justice for the Law. The law courts are not just any old legal institution. Given the pivotal role they play in the legal system, it is only fair to ask how the priority that justice enjoys in the courts can fail to transmit to the legal system as a whole. I do not believe we can meet this challenge without examining the relative importance of courts to the legal system. Specifically, I will recommend meeting the challenge by rejecting the
Priority of Courts for the Law
Courts are the first institution of the law: courts are more important to the legal system than any other legal institution.
I. COMPARATIVE JUSTICE AND THE LAW
No doubt “injustice” can be used to describe any morally wrongful act. To be just in this general sense is simply to act rightly or treat others morally. But if the morality of right- and wrongdoing exhausts morality, the claim that rightness is law's first virtue is false. For in this case, rightness is no particular moral virtue but rather the master virtue of acting morally, all things considered. Since exhibiting this virtue already takes account of each relevant moral virtue, it cannot conflict with any. In the absence of a second virtue, there can be no first virtue either. And if the morality of right- and wrongdoing does not exhaust morality, the claim that rightness is law's first virtue is a truism. For the morality of right- and wrongdoing is certainly morality's most important part: all other things being equal, it is better to act morally rightly (comply with one's obligations) than to act morally supererogatorily (comply with one's optional reasons). Of course compliance with obligation is law's first virtue.
A. Comparative Justice
Narrower conceptions of justice make better sense of the Priority of Justice for the Law. We begin with a venerable conception of justice harking back to Aristotle. Justice in this conception is an essentially comparative virtue, and the comparison in question is among different subjects. So conceived, justice is distinguished from the rest of morality by its concern with the relative positions of two or more subjects—with how subjects are treated relative to one another. Justice strikes the right balance between the competing interests of subjects, always taking its subjects together rather than one at a time.Footnote 5
Aristotle famously divided “special” justice into distributive and corrective.Footnote 6 He makes the comparative nature of these forms clear by defining them in terms of equality. Distributive justice is defined as “geometrical” equality of ratios. It requires that the share each subject receives be a function of what others in the comparative class receive (given what they deserve absolutely). Corrective justice is defined as “numerical” equality of differences. It requires the shifting of the wrongful loss (gain) some interaction between two subjects has created for one of them onto the other. Hence corrective justice always restores one of two parties to some wrongful interaction back to the position in which she stood before that interaction, and moreover always does so by subtracting something from one party and adding it to the other, thereby changing their relative positions.
Aristotle's labels can mislead. It is not that the one form of comparative justice is distributive whereas the other is corrective; both are distributive. Indeed both are distributive in a comparative way: they both concern the distribution of some fixed amount of good or ill between two or more subjects, where what each subject should receive is a function of what others do. The difference is that distributive justice is not corrective: only corrective justice presupposes some wrongful interaction between two subjects and cares to bring at least one of them back to where she stood before.
Both distributive and corrective justice strike a balance between the conflicting claims of subjects, where satisfying the claims of one can only be achieved by frustrating those of others. Hence there are no winners (losers) without losers (winners). This is obvious enough in the corrective case, where taking the loss from the innocent party (or giving her the gain) necessarily gives that loss to the wrongdoer (or takes that gain from him). But the same point is true of distributive justice. Since what is being distributed is some good or ill—something that is in the interest of each subject to have more or less of—and is finite, for every distribution there is at least one subject who would do better under an alternative distribution.
B. Primary and Secondary Duties
We now explain why the Priority of Justice for the Law cannot be sustained under the comparative conception of justice. Our starting point is the traditional view of law as a system of social norms. Given this view, it is plausible to expect the relation between legal norms and justice to reflect the relation between law and justice. Rather than discussing legal norms generally, I focus on mandatory legal norms or legal duties. After all, duties are particularly important norms. Requiring subjects to engage in certain types of conduct or refrain from others is the most direct way in which the law affects human behavior. The Priority of Justice for the Law, I believe, can only be borne out by an examination of legal norms if it can be borne out by an examination of legal duties.
We now observe that legal duties generally divide into two classes, which are sometimes labeled primary and secondary. Call a legal duty secondary just when it presupposes a violation of some other legal duty; otherwise call it primary. I believe these labels aptly reflect the relative importance of the two duty classes within the system of legal duties. For every legal duty Ds that presupposes a violation of some legal duty Dp, the functions of the law would be better fulfilled when Dp is performed than when Dp is violated and Ds is performed. This seems to me to establish a major functional priority of primary legal duties over secondary.
Two examples will suffice to illustrate this point. Consider first criminal-law duties. The duty to submit to criminal punishment presupposes a crime, namely, a violation of some criminal-law duty. Obviously enough, it is better that people comply with criminal-law prohibitions (by refraining, say, from murder or rape) than that they violate them and readily submit to punishment. Consider second private-law duties. The duty to pay compensation for a private-law wrong presupposes a breach of some private-law duty. While this is somewhat less obvious, I think it is still clearly better if people perform their private-law obligations (by refraining, say, from driving negligently) than if they breach them and pay full compensation.
It is not for nothing that primary legal duties are labeled primary; they really are (functionally) primary.
C. Primary Duties and Comparative Justice
Next we argue that primary legal duties are not essentially duties of comparative justice. To discharge this claim, all I need to show is that some paradigmatic primary legal duties have nothing special to do with justice. Consider a paradigm of primary criminal-law duties: offences against the person. These are essentially protective duties: they proscribe such harmful treatments of persons as murder, rape, and grievous bodily harm (GBH). The crucial point is that the harmful treatment is typically defined in terms of the treatment's impact on the agent being treated. No reference is typically made to the relative positions of the treating agent and the treated agent—nor any other pair of subjects, for that matter. Consider the equal-opportunity murderer, who picks his victims at random without discrimination on the basis of objectionable distinguishing factors. At least insofar as the very concept or form of comparative justice is concerned, an equal-opportunity murderer might be regarded as impeccably just. In any event, such a murderer is certainly not in the slightest unjust, comparatively speaking. No essentially comparative considerations could be said to be contravened.
Now of course, a murder changes the relative positions of the offender and the victim. Think about the aggregate lives and deaths between them. Beginning with two persons each having a life, we end up with one person having a life and the other having a death. But to think that this pattern of distribution of lives and deaths is of the essence of murder is a perversion. A murder-suicide is murder all the same, but it seems to retain the relative positions of the parties. After all, the difference between two deaths is still zero.
Laws defining offences against the person are essentially protective. They protect persons from various forms of harms, and the harms they protect against are largely noncomparative: they consist of various setbacks to the well-being or dignity of persons taken one at a time. I think this is true of primary criminal-law duties in general. When they do not protect persons, they protect other vital human and social interests: national security (terrorism), private property (theft), stock markets (insider trading), transportation (dangerous driving), public health (selling old milk), clean environment (pollution), public safety (assault weapons). Here, too, there seems to be nothing particularly comparative about the proscribed harms. I believe the same point can be extended to primary private-law duties.
Now, the protective nature of duties does not prevent them from being comparative: the harm protected against may be essentially comparative. In fact, some primary legal duties are protective duties of comparative justice—the primary duties of antidiscrimination law, for example. Wrongful discrimination necessarily involves treating members of a certain group (women, men) worse than members of another group (men, women) are or would be treated. It is of the very essence of discrimination that the discriminator treats one subject worse than he treats or would have treated some other, similarly situated person. Duties to refrain from direct (disparate treatment) or indirect (disparate impact) discrimination are duties of distributive comparative justice: they are duties to divide up some fixed amount of goods, such as jobs and university admissions, or ills, such as stop-and-searches and conscription summonses, in certain proportions and not others.Footnote 7 But this is a case where the exception proves the general rule. For this concern with intersubjective comparison at the level of primary duties is a highly contingent feature of the law. This is borne out by the relatively recent and parochial history of antidiscrimination law, which first developed in the United States only after World War II and has since spread to other Western jurisdictions but not much beyond.Footnote 8 But even in jurisdictions where antidiscrimination law is extremely important, the great majority of primary legal duties have nothing particularly to do with comparative justice.
We are only one step away from our first and negative conclusion—the rejection of the Priority of Comparative Justice for the Law. Remember our simplifying assumption that the legal system is a system of duties. If the discussion so far is accurate enough, then the functionally primary duties of that system—primary legal duties—have nothing special to do with comparative justice. But then it seems reasonable to conclude that comparative justice is not the first virtue of the law. On the face of it, if justice were the first virtue of the law as a whole, the law would deploy its functionally primary duties in the service of justice. Presumably, these duties would require us to act justly or do justice. The fact that they do not places anyone who is tempted by the Priority of Comparative Justice for the Law under an obligation to provide an explanation, in the absence of which we are entitled to reject that thesis.
II. COMPARATIVE JUSTICE AND THE LAW COURTS
At this point I would like to examine the possibility of retreating to a weaker and more plausible thesis about the relation between law and justice. Gardner argues that comparative justice is the first virtue of the law courts,Footnote 9 and his argument is both original and plausible. The basic thought capitalizes on the structure of adjudication. Litigation always pits two parties against each other. Every case is a competition in which there can be no winner without a loser. To accept the plaintiffs' claim is to reject the defendant's, and vice versa; to accept the case for the prosecution is to reject the case for the defense, and vice versa. Adjudication always involves the judge in striking a balance between the competing claims of the parties. When the courts strike this balance correctly, they cannot fail to do justice. Obviously, the courts often fail to strike the balance correctly: it can hardly be said that they are always comparatively just. But when they strike the balance incorrectly, they cannot but do comparative injustice. When the courts fail to be just, they fail by their own internal moral standard. Hence comparative justice is their first virtue. Gardner says:
this special association between courts and justice comes of the fact that, in court, there can normally be no winners without losers. Whatever the court decides, what is at stake is the relative positions of the parties. Thus the reasons which the court has for its decisions, whatever else they may include, must include reasons for or against granting a judgment to one person at another's expense, i.e., reasons for or against altering the parties' relative positions. This way of putting it may especially bring to mind the activities of courts dealing with private law disputes. But the same point applies, when you think about it, in criminal law and public law adjudication, where what is at stake is the state's right to punish someone or deprive them of some benefit, in which the victory of the state is also, of necessity, that person's loss. In fact, with the possible exception of purely declaratory judgment, there can be no escaping the issue of justice in the courtroom.Footnote 10
Let us accept for argument's sake Gardner's claim that comparative justice is the first virtue of the private law courts. Even so, I think we should reject his thesis, because comparative justice cannot plausibly be regarded as the first virtue of the criminal-law courts, which are at least as important to the court system as the private-law courts.
The justice we bring criminals to when we bring them to justice is essentially retributive. The central question is always what the defendant deserves for his crime. Asked in the context of a criminal-law case, this question is not essentially comparative in the intersubjective sense. If it calls for any comparison, that comparison is not between crimes or punishments of different persons but rather between the crime and punishment of the very same person: the defendant.
Notice that this lends no support to the familiar claim that comparative facts are strictly irrelevant to the issue of deserved punishment.Footnote 11 It is reasonable to believe that what punishment the defendant deserves may well depend on how the legal system punishes others for similar crimes. This much, however, is compatible with thinking that the question of deserved punishment is essentially noncomparative, because the relevance of the said comparative facts may well be derivative rather than intrinsic. The power of acts of punishment to give a defendant what he deserves for his crime depends on the intensity of its condemnatory message, which in turn depends on how the system punishes others for similar crimes. When the system is overly harsh (lenient), the condemnatory intensity of a unit of punishment is diluted and more (less) punishment is needed to give the defendant what she deserves. But this appears to presuppose that one always deserves a certain punishment absolutely, in virtue of one's crime and regardless of how others are punished. The ultimate question of deserved punishment may well remain noncomparative.Footnote 12
The adversarial structure of adjudication in common-law jurisdictions naturally creates the impression that every court proceeding is a symmetrical competition in which there can be no winner (loser) without a loser (winner). In the private-law case, this impression happens to be accurate, because the procedural symmetry is accompanied by a substantive one. The procedural winners and losers are necessarily substantive winners and losers. But this tidy correspondence between procedure and substance does not carry over to the criminal-law courts. To be sure, here, too, there are no procedural winners without losers: by accepting the case for the prosecution, the court cannot but reject the case for the defense, and vice versa. On reflection, however, these corresponding victories and defeats do not tell the true story. When a private-law defendant is rightly ordered to pay the plaintiff compensation, the defendant loses as much money as the plaintiff wins. Moreover, the losing and winning are in terms of the genuine interests of the parties. However, when the defendant in a criminal case is rightly acquitted (say, of a crime he did not commit), who could possibly be said to suffer a comparable loss in terms of her genuine interests?
I see exactly two possibilities, neither of which is appealing. It is implausible to say that the prosecution or the state suffers a genuine loss in such a case. The mission of the prosecution is to help the court reach the right verdict, not maximize punishment. The state's securing the punishment of the innocent is not even a Pyrrhic victory; it is not a victory at all. So when a defendant is rightly acquitted, the state is not defeated. If we must use this terminology, we should say that both litigating parties emerge victorious.
Recall Gardner's claim that in criminal-law adjudication “the victory of the state is also, of necessity, that person's loss.” Notice first that this appears to contradict my discussion so far. If I am right, there is at least one case (where the defendant is rightly acquitted) in which the state's victory (in securing justice) is also the victory of the defendant. To my mind, this is a win-win situation. Second, it is clear that while Gardner's claim is entirely general, he has in mind the case in which the defendant is rightly convicted. You might think that at least in this case, the state's victory corresponds to the defendant's loss. But this is not so. It is true, of course, that the defendant in this case suffers a genuine loss. Conviction and punishment clearly set the defendant's interests back. Now, perhaps it also makes sense to say that the state gains something of genuine value in this case. In particular, that justice has been done might itself be taken to represent a genuine gain for the state. But the mere fact that the state gains one thing and the defendant loses another in a single case cannot bring it under comparative justice. The gain and the loss must also be comparable. And yet the state's gain and the defendant's loss in the case at bar are clearly incomparable. The state wins none of the well-being lost by the defendant; the defendant loses none of the justice gained by the state.
Then perhaps it is the victim who suffers the comparable loss? But first and most obviously, criminal adjudication sometimes proceeds in the absence of a victim. A large number of crimes, for example, pollution, driving while intoxicated, or tax evasion, are virtually victimless. They are harmful to valuable social institutions and common goods, but not to anyone in particular. Second, many if not most crime victims want to see the defendant convicted and punished if—but only if—he deserves to be. What loss do they suffer, in terms of their genuine interests, when the defendant is rightly acquitted or spared undeserved punishment?
III. REQUITATIVE JUSTICE AND THE LAW
We have now seen that the comparative conception of justice fails to sustain the general Priority of Justice for the Law (Section I) or the weaker, more plausible Priority of Justice for the Courts (Section II). But it may be objected that the comparative conception is rather narrow. For all I have said, either one of these theses might be true under some other, more interesting conception of justice. I cannot refute the objection, because I cannot examine every—or even every plausible—conception of justice. But perhaps I can mitigate its force by examining what strikes me as the most plausible alternative conception of justice in the neighborhood.
A. Requitative Justice
At least as broad and pertinent as the comparative conception of justice is the age-old notion that justice is the giving or getting of what is deserved. Desert-based justice is not necessarily comparative. Treating an agent according to her deserts is a matter of maximizing fit between her treatment and her deserts. This might require us to strike a right balance between the agent and someone else. But as our discussion of retributive justice shows, this is not necessarily so. That discussion also shows that the desert-based conception of justice is pertinent to the study of law: it explains criminal-law adjudication better than the comparative conception.
If the worry that prompted me to bring the desert-based conception into the discussion was a perceived narrowness of the comparative one, the worry now is that the desert-based conception might be all-encompassing. The standard analysis says that desert is a three-place relation between a deserving subject, a deserved object, and a desert basis, where the subject always deserves the object in virtue of the basis. This analysis does little to prevent us from describing every morally wrongful (rightful) act as undeserved (deserved). (And not just acts: we also talk about deserved states of affair that are not the product of human agency). However, there may well be useful ways of delimiting the concept of desert.Footnote 13 And we are not in the business of offering a general theory of desert. All we need is a minimally circumscribed concept that will help us explain the law in particular.
Call it requitative justice. One distinguishing factor of requitative justice, let us say, is that it always involves some agent who does the requiting by giving another agent the treatment he deserves to receive (or not to receive). The primary way of being requitatively just (unjust), then, is to treat agents according (contrary) to what they deserve. This rules out talk of (un)deserved weather, for example. Second, let us stipulate that the desert basis of requitative justice is some human conduct on the part of the deserving agent. There is always some act or omission in virtue of which the agent deserves the treatment. There is no requital for overall virtuousness, for example. Finally, assume that requital is (doubly) negative: it consists of some treatment that is bad from the deserving agent's perspective and it is deserved in virtue of some conduct that is bad from the treating agent's perspective. Call these sanction and wrong, respectively.
This requitative model of desert is predicated on the law. I make no apology for this. We need some way of delimiting the potentially all-encompassing concept of desert, and since the point of the exercise is to explain the law, the legal model of desert is as good a place to start as any. Later I consider relaxing this conception. For now I am asking the reader to acquiesce in it.
B. Primary Duties and Requitative Justice
Recall the argument against the general Priority of Justice for the Law under the comparative conception of justice. We now extend that argument to show that the Priority of Justice for the Law cannot be sustained under the requitative conception. All we need to show is that primary legal duties have nothing special to do with requitative justice; everything else remains the same as before. I claim that: requitative justice necessarily presupposes some wrong to requite; but primary legal duties do not necessarily presuppose some wrong to requite; hence they are not duties of requitative justice.
Consider again the paradigmatic primary criminal-law prohibitions associated with offenses against the person. Saying that the prohibitions on (wrongs of) murder, rape, or GBH are essentially requitative implies that they necessarily presuppose some requitable negative conduct—some wrong—on the part of the criminally treated agent, the would-be murder, rape, or GBH victim. But this is absurd. It means that to be protected by these criminal prohibitions, the would-be victim must first be guilty of some wrong—that the wrongs of murder, rape, and GBH can only be committed against a wrongdoer and never against a right-doer.
I have previously claimed that primary criminal-law duties are essentially protective but not essentially comparative. We have just seen that they are not essentially requitative, either. Given the functional priority of primary duties and the reasonable assumption that they would bear some special relation to the first virtue of the law (if there were one), it follows that requitative justice is not the first virtue of the law.
C. One Objection
To allay the suspicion that this is too quick, let us consider two objections. The first emphasizes that my argument crucially depends on the assumption that requital presupposes some wrong or negative conduct. This assumption in turn depends on the particular conception of requital in which you have patiently acquiesced. Having seen the consequence, you may now wish to reconsider. Even if legal requital necessarily presupposes some wrong to requite, this need not be true of requital generally. What about (doubly) positive nonconduct requital, the requital of treatment that is good from the deserving agent's perspective and is deserved in virtue of some fact about that agent other than her conduct, one, moreover, that is good from the treating agent's perspective? Perhaps what is wrong with murder, rape, and GBH is that they fail to requite good with good (and requite good with ill). My argument simply rules this possibility out by fiat.
Under the present proposal, to say that the prohibitions against murder and rape are essentially requitative is to imply that they necessarily presuppose some requitable positive fact about or good-making property of the criminally treated agent (would-be victim). But what could this fact possibly be? What good-making property of the would-be victim do primary criminal-law duties require the would-be murderer or rapist to requite with abstention from murder or rape? One possibility is the would-be victim's innocence, the fact that she does not attack the would-be assailant. But all that this can possibly explain is why the assailant does not have self-defense. It cannot possibly explain why he needs such defense in the first place, why murder would be wrong in the absence of self-defense.
Another possibility is that the would-be victim is innocent in a more robust sense. He is not guilty of an offense, not only against the would-be assailant but also against others. But of course, the protection of the prohibitions on murder or rape is not restricted to those who have not offended, not even to those who have not themselves committed murder or rape. A murderer's murder is murder; a rapist's rape is rape. So perhaps it is the basic needs of the would-be victim that the prohibitions require us to requite with abstention from crime: the basic needs not to be murdered or raped. But how does the sheer fact that the agent has a basic need make her good? Indeed, it is one of the most entrenched sentiments in the theory of desert that the satisfaction of someone's sheer needs can be justified without being personally deserved.Footnote 14
Finally, what about the notion that the would-be victim deserves abstention from crime in virtue of her humanity? Now, since we can claim no more personal credit for our humanity than for our needs, I do not think humanity's claims are desert claims in a minimally robust sense. But even if humanity is a good-making property of agents, the claim that the would-be victim deserves not to be murdered or raped in virtue of her humanity is nonexplanatory. It adds no content to the claim that murder and rape (of humans—what else?) are morally wrong; it simply reformulates it. The language of desert can be cheap.
D. Another Objection
The second objection is a counterreply. In the previous paragraph I rejected the proposal that murder requites the would-be victim's innocence of some past wrongdoing. Good. But consider the case in which a court of law wrongly convicts and sentences an innocent defendant to death. Surely this is a case of undeserved treatment—of requitative injustice—if any is? But how does this case differ from an ordinary murder of an innocent victim? My account does not answer this question, because in neither case is there any wrong or negative conduct to requite. Indeed the only thing that seems to be requited in the sentencing case is the victim's innocence. But if innocence can be a desert basis in the sentencing case, why can it not be a desert basis in the ordinary murder case, too? If a person deserves not to be sentenced to death in virtue of her innocence, why does she not deserve not to be murdered in virtue of her innocence?
Since punishing the innocent is obviously a paradigm of undeserved treatment—sentencing an innocent defendant to death is requitative injustice, if anything is—I must explain why I think the relevant concept of desert or requital does not apply to murder generally. It seems to me that the only difference concerns the intention the negative treatment expresses. Consider the sentencing case first. It is often said, plausibly, that a minimally robust notion of punishment requires the punishing treatment to be of an offender for an offense. But we should read this to refer literally only to instances of rightful punishment. What is necessarily true of punishment more generally is quite different, namely, that it expresses an intention to condemn an offender for an offence.Footnote 15 And condemnation of an offender for an offense is precisely what criminal conviction and punishment cannot but express.
The criminal trial necessarily presents the judges or jurors with a defendant and a charge specifying a punishable offense he allegedly committed. In the overwhelming majority of cases, these officials will deliver their verdict with every intention of giving the defendant the punishment he deserves for committing the crimes charged (including no punishment for committing none). But the expressive content of the verdict is no hostage to the intentions of the presiding judges or jurors. It is largely a matter of convention. Even the judges or jurors who intend simply to inflict pain on the defendant for nothing other than his skin color, knowing full well he is innocent, cannot fail to condemn him for the crime charged. The public meaning of conviction and punishment is an integral part of the legal practice of adjudication.
Compare the ordinary murder case. There can be no suggestion that murder generally expresses condemnation of an offender for an offence. The murderer can murder the victim to get him out of her way, all the time considering him a moral saint deserving of nothing but infinite bliss. And unlike the case of state punishment, there is nothing in the conventional, public meaning of murder to contradict this. This is enough to show that murder is not undeserved or requitatively unjust punishment.
Why can the murder not be undeserved non punishing negative treatment? Because the previous point can be generalized. Not every negative treatment can be deserved, only those that express an intention to disapprove, blame, or otherwise negatively evaluate some negative conduct or bad-making property of the agent.Footnote 16 Similarly not every positive treatment can be deserved, only one that expresses an intention to approve, praise, or otherwise positively evaluate some positive conduct or good-making property of the agent. Murder and rape typically express one or more negative attitude towards the victim, such as hatred, anger, and at the very least disregard. But these are not essentially evaluative of the victim.
This last point also provides an alternative reply to the first objection. For even if such positive nonconduct facts about the would-be victim as her innocence, basic needs, or humanity could in principle serve as requitable desert bases, the claim that crimes necessarily express an intention to evaluate the would-be victim positively is even less credible than the claim that they express an intention to evaluate her negatively.
IV. REQUITATIVE JUSTICE AND THE LAW COURTS
Having argued against the Priority of Justice for the Law under the comparative and requitative conceptions of justice and against the Priority of Justice for the Courts under the comparative one, I wish to argue for the Priority of Justice for the Courts under the requitative conception. Like its comparative counterpart, the requitative conception can explain why justice is the first virtue of the private-law courts. But unlike the comparative conception, the requitative one can also explain why justice is the first virtue of the criminal-law courts. The job of the comparative conception is not to unify the private-law and the criminal-law courts but to distinguish them.
A. Secondary Duties and Requitative Justice
The theoretical apparatus that has served our case against the Priority of Justice for the Law will also serve our case for the Priority of Justice for the Courts. Previously we focused on the fact that primary legal duties and wrongs do not necessarily presuppose any wrong or negative conduct to requite. We now focus on the fact that secondary legal duties and wrongs cannot possibly fail to presuppose some requitable legal wrong—namely, the violation of some legal duty. And yet to bring a case before a court of law is just to claim that some secondary duty applies to the defendant. So by the time a case properly reaches the court, a sanctionable legal wrong has been committed.
This explains why the courts virtually always have opportunity to requite. We also need to show that they virtually always take this opportunity. Now, although this was not built into the definition of secondary legal duties, it is fairly clear that they are always “bad” from the perspective of the legal wrongdoer. In point of fact, duties that answer to our definition of secondary legal duties virtually always require legal wrongdoers to absorb some genuine cost—some remedial burden or punishment that sets their interests back in one way or another. In any event, when the courts apply secondary duties to defendants in response to their legal wrongs, they give treatment that is negative from the defendant's perspective for conduct that is negative from the legal perspective. They impose sanctions for wrongs. This all but establishes that the courts are essentially a system of negative conduct requital.
To clinch this conclusion, we do not have to show that secondary legal duties as such are essentially requitative. It will suffice to show that their application by the courts is. The only missing link in the story so far is of course the expressive content of judicial decisions. But it seems clear that when judges in black robes apply a secondary legal duty to the defendant by imposing a legal sanction on her on account of her legal wrong, they cannot but express official disapproval of the defendant's conduct. Now, no doubt the negative evaluation expressed by conviction and punishment goes well beyond the one expressed by an order to pay compensation. Only the former officially stigmatizes the offender as antisocial. Moreover, a compensation order expresses more than disapproval of the defendant's wrong; it also publicly vindicates the plaintiff's right and validates her complaint. My claim is that judicial imposition of legal sanction at least expresses official disapproval of the defendant's conduct. The application of remedial secondary duties, like that of punitive ones, is never indifferent to the legal wrongs that trigger their application. Whatever else they do, they at least disapprove of these wrongs. This makes the court system one of negative conduct requital—and requitative justice the first virtue of the law courts.
B. The Job of Comparative Justice
Requitative justice, then, can be said to unify the criminal-law and the private-law courts in a way comparative justice cannot. Both the criminal-law and the private-law courts are systematically called upon to requite the defendant's legal wrong with some legal sanction—some treatment that imposes a genuine cost on the defendant or sets her interests back. Nevertheless, the requitative conception cannot explain the difference between criminal-law and private-law courts. That difference lies in the kind of requital sought in the two cases. And it is here that the comparative conception of justice becomes handy. In fact, it is just one of the two Aristotelian forms of comparative justice that will do the trick.
The requital of the criminal-law courts is essentially retributive. As we have seen, retribution is essentially noncomparative. The cost that punishment imposes on the defendant is mirrored by no one's comparable gain; just punishment involves a loser but no winner. We can say, then, that retributive justice is exclusively requitative, namely, requitative noncomparative. By contrast, the requital of the private-law courts is essentially corrective. It involves the taking of something from one party to a wrongful interaction and giving it to the other with a view to bringing at least one of them back to the position in which she was before that interaction. This has an essentially comparative element. Thus, for example, the cost a compensation order imposes on the defendant is mirrored by the comparable gain it confers on the plaintiff. So unlike retributive justice, corrective justice is not exclusively requitative; it stands at the intersection of requitative and comparative justice. It is requitative because it gives the defendant-wrongdoer the legal sanction she deserves in virtue of having committed a legal wrong. But it is also comparative because the deserved sanction necessarily changes the relative positions of the defendant-wrongdoer and the plaintiff-victim.
Aristotle famously describes the judge as justice personified.Footnote 17 He probably had in mind private-law judges and corrective justice. To extend the metaphor to judges generally, we can abstract from the comparative aspect of corrective justice and concentrate on its requitative aspect. Judges personify corrective justice sometimes, retributive justice at other times, but requitative justice always. This is another way of saying that requitative justice is the first virtue of the law courts.
C. Completing the Explanation
The Priority of Justice for the Courts can explain the near-universal association between law and justice with which we began. All we need to add is the relatively uncontroversial
Essentiality of Courts to the Law
Courts are essential to the modern municipal legal system.
At least insofar as modern municipal legal systems are concerned, the law must include institutions that enforce primary legal duties by applying secondary ones to those who violate them and that resolve disputes over such violations. So the law cannot do without institutions the first virtue of which is justice. Hence the near-universal association between law and justice.
That association is mostly confined to the court system. After all, it is judges whom we call justices, the government's litigation department that we call the Justice Department, and the court—not legal—system that we call the administration of justice. It is claims that the courts will hear that we call justiciable, bringing criminals to trial that we consider bringing them to justice, and our court—not legal—system that our scales of justice symbolize.
What about the paradoxical-sounding maxim, “Unjust law is not law”? Surely this statement is not confined to the court system. Does not the fact that natural-law scholars have seriously entertained this claim but not similar paradoxical-sounding claims about unprotective, inhumane, or inefficient law suggest that our association between law and justice goes beyond the court system?Footnote 18
But I think the injustice in the maxim has nothing special to do with requitative justice, comparative justice, or indeed any conception in which justice is one distinct moral virtue among others. Historically, lex injusta has been used to designate morally wrongful, bad, or reprehensible laws or legal systems.Footnote 19 The injustice in question concerns primary as much as secondary legal duties. Laws or legal systems that fail to criminalize what ought morally to be criminalized (rape, genocide, lynching) or that criminalize what ought morally not to be criminalized (sodomy, smoking, protesting) are clear-cut examples of unjust laws or legal systems in the scholastic sense. So are laws or legal systems that require or permit mass murder of dissidents. And yet mass murder, like murder generally, is not necessarily unjust, requitatively or comparatively. The injustice of unjust law is simply the master wrong of treating people wrongfully, morally speaking.
D. Courts as Law-Making Institutions
When I say that (requitative) justice is the first virtue of the law courts, I mean to say that it is the first virtue of the courts as such, as what they are essentially. Courts are virtually always called upon to apply some secondary legal duty, thereby authoritatively resolving some particular dispute over the violation of some primary duty. Call this (essential) function of the courts their law-applying function. In addition, some appellate courts in common-law jurisdictions lay down general legal rules that are binding not simply on the parties at bar but also on the public at large. They do so by setting precedents. Call this (nonessential) function of the courts their law-making function. (Notice that in a sense, courts could be said to make law whenever they apply law; but the new law that courts necessarily make through the application of other laws applies to the parties at bar only. By “law-making,” I mean the making of general rules—the making of law other than by applying law.)
As a thesis about courts that are empowered to make general legal rules as well as apply them in particular cases, the Priority of Justice for the Courts is unsustainable. That thesis obviously owes its plausibility to the essential law-applying function of the courts. The courts do the requiting when they apply some secondary duty to the defendant in virtue of her wrong. When the courts make law in the form of a general rule by setting a precedent, they need not and typically do not requite anything at all. Specifically, appellate courts constantly impose new primary legal duties or reshape old ones, and as we established in Section II, primary legal duties are not essentially requitative. Judicial law-making, like law-making generally, has nothing special to do with requitative (or comparative) justice. And yet there is no special reason to think that the law-applying function of such hybrid law-making and -applying institutions is more important than their law-making one. In fact, the higher the court in the judicial hierarchy, the more important its law-making function relative to its law-applying one. Surely it is the law-making power of the U.S. Supreme Court that accounts for its immense importance in American history and justifies the heightened attention given to the president's judicial nominees. The ongoing controversy over that famous decision from 1973 is not about whether justice was done between Roe and Wade but whether the general rule in Roe v. Wade is good law.
Still, we are entitled to regard requitative justice as the first virtue of the courts. While all courts are law-applying institutions, only some are also law-making institutions. Courts in Continental jurisdictions and courts of first instance in common-law jurisdictions typically have no power to lay down general legal rules. Nevertheless, they are always empowered to apply secondary legal duties in response to violations of primary ones and to resolve disputes over such violations.
In sum, requitative justice is the first virtue of the courts as such (law-applying institutions) but not of courts as hybrids (law-making and -applying institutions).
V. THE PRIORITY OF JUSTICE AND THE PRIORITY OF COURTS
At this point it may be objected that the positive and negative theses of this essay are mutually undermining. The objection I have in mind accepts that primary legal duties have nothing special to do with justice but seeks to lever the Priority of Justice for the Courts into the Priority of Justice for the Law, with the aid of the
Priority of Courts for the Law
The courts are the first institution of the law: courts are more important to the legal system than any other legal institution.
As before, “courts” here designates what courts are essentially, namely, institutions that resolve disputes over the application of general legal rules to particular cases or law-applying institutions. In particular, they contrast with institutions that make general legal rules, or law-making institutions. The Priority of Courts for the Law is popular. It is most naturally associated with legal realism but has also attracted many who reject the realist picture of law. Perhaps most notably, it informs the version of legal positivism championed by Hart, whose celebrated rule of recognition consists of judicial law-applying practices.Footnote 20 Several followers have explicitly endorsed that thesis. According to Gardner, for example, “adjudicative institutions are, in a sense, the linchpin of all legal systems. . . . Indeed, as Hart did not spell out but others have added, the presence of courts turns out to be more crucial to the existence of a legal system than the presence of any other legal institutions.”Footnote 21 These others include Joseph Raz, who claims that “the existence of norm-creating institutions. . . is not a necessary feature of all legal systems,” but “the existence of certain types of norm-applying institutions is.”Footnote 22 According to this approach, “a law is part of the system only if it is recognized by legal institutions. The emphasis is, however, on the law-applying rather than the law-creating institutions.”Footnote 23 In short, “the key to the systematic nature of law lies with the functioning of the law-applying institutions.”Footnote 24 Now consider the
Levering Argument (1) the Priority of Justice for the Courts: justice is the first virtue of the law courts (as such); (2) the Priority of Courts for the Law: the law courts (again, as such) are the first institution of the law; it follows that (3) justice is the first virtue of the first institution of the law; but (4) what more can be required of a virtue before it can be said to be law's first? Therefore, (5) the Priority of Justice for the Law: justice is the first virtue of the law (as a whole).
Now, one can question the inference from the first two premises to the conclusion. One might deny, for example, that if something is the first virtue of law's first institution, then it is law's first virtue. I wish to ignore this line of attack for two reasons. First, I am not sure it can succeed. Intuitively, if justice really is the first virtue of law's first institution, then this makes it okay to regard it as law's first virtue. Second, the argument is unsound anyway: the courts as such are not law's first institution. Doing justice to the Priority of Courts for the Law requires a separate paper.Footnote 25 Here I will present what I take to be the basic case against it.
A simple thought supports the claim that law-making institutions are just as important to the existence of the legal system as law-applying ones. On the face of it, law-applying institutions presuppose laws (general legal rules) to apply, which in turn presuppose law-making institutions. Laws, after all, are artifacts; they need to be made in order to exist. If so, law-making institutions—social institutions that produce general legal rules—seem to be as important to the legal system as law-applying ones. So courts as law-applying institutions do not seem to be more important to the legal system than law-making institutions as such. But law-applying institutions are what courts are essentially. So courts as such do not seem to be more important to the legal system than every other legal institution.
A. The Purely Customary System
Let us look at a possible rebuttal. Perhaps the best case for the Priority of Courts for the Law capitalizes on the possibility of a purely customary legal system. Gardner says:
One may have a legal system with no legislature and no police force and no legal professions—that is to say a purely customary legal system—but one has no legal system at all until one has courts, i.e., adjudicative institutions charged with administering a system of rules by which they themselves are bound.Footnote 26
This point seems to be a consequence of the claim, which Hart famously thought impugns the so-called command theory of law, that some legal systems have purely customary rules.Footnote 27 Notice that the legal system in question is purely customary only in the sense that custom is the only source of law. This implies that the system lacks courts with power to lay down general rules (by setting precedents) but not that the system lacks courts as such, namely, as law-applying institutions. Indeed, the quoted passage takes courts to be essential to the system (custom is a source of law for the system only because the courts apply customary rules). So the possibility of a customary legal system is the possibility of a legal system with (1) courts, (2) customary legal rules, and (3) no other legal rules (statutes or precedents).
This seems right as far as it goes, but it does not go very far at all. In particular, it does not rebut the simple thought that law-making institutions (as such) are as important to the legal system as courts (as such). At most, the case of the purely customary legal system shows that law-making institutions need not be law-enacting organizations (legislatures), namely, organizational institutions that make law by enactment (by performing acts with the intention to make law). What Hart shows is that customary, nonorganizational, nonenacting practices can be an ultimate source of general legal rules, one that does not derive from other law-making institutions. Such customary legal rules hardly look like commands. Still, this does not tend to support the Priority of Courts for the Law, because it leaves intact my claim that law-making institutions (as such) are just as important to the system as courts (as such). Customary law-making institutions are law-making institutions all the same.
B. Customary Law-Making Institutions?
It may immediately be objected that I am misrepresenting the Priority of Courts of the Law. Surely talk of law-making institutions is not meant to include custom? Presumably, those who accord courts or law-applying institutions priority over “law-making institutions” take this expression to mean legislatures. Nonorganizational, customary practices that make law by convention rather than by enactment are irrelevant. Their claim is the
Priority of Courts for the Law (Organizations)
The courts are the first organization of the law: courts are more important to the legal system than any other legal organization.
Before I explain why we should reject this reinterpretation, let me stress that there is no linguistic impropriety in talking about nonorganizational, nonenacting, law-making institutions. We talk about the (nonorganizational, customary) institutions of promising, marriage, the market, and, indeed, the law.Footnote 28 And we talk about conventional rules as (hu)manmade. The question is not linguistic but explanatory.
The reinterpretation on offer faces a dilemma. As a thesis about legal systems generally (including nonmodern or international ones), its indifference to customary law-making institutions is manifestly arbitrary. It is also too weak to do its intended levering duty. And as a thesis about the modern municipal system, the reinterpretation is just as objectionable as the interpretation it seeks to replace. It also renders the levering argument equivocal.
First Horn of the Dilemma
In the previous section, I relied on the relatively uncontroversial
Essentiality of Courts to the Law
Courts are essential to the modern municipal legal system.
It is indeed exceedingly hard to imagine a modern municipal system without courts. But as soon as we extend our inquiry to nonmodern (or international) legal systems, that assumption becomes dubious. What is there, for example, to prevent a small, close-knit tribe from having a regime of customary rules of conduct but no institutions that resolve disputes over their application? It is by no means clear how we could nonarbitrarily deny such a social regime the title of a legal system. The Priority of Courts for the Law (Organizations) simply ignores the possibility of courtless legal systems. But suppose courts are essential to the legal system and let us focus on the purely customary legal system—a system with courts and purely customary law-making practices. The Priority of Courts for the Law (Organizations) simply ignores the system's customary law-making practices. What is the rationale for the exclusion, the fact that they are not organizations? Given that the system is (1) obviously not quite modern or municipal, and (2) could not exist without these practices, how can an explanatory theory of law afford to ignore them? From a functional perspective, what matters is that some social institution performs a function that is important to the system.
Its arbitrary exclusion of customary law-making institutions also strips the Priority of Justice for the Courts (Organizations) of its power to lever the Priority of Justice for the Courts into the Priority of Justice for the Law. It is compatible with the claim that some institutions whose essential function is other than the essential function of the courts (law-applying) and that have nothing special to do with justice are just as important to the system. Specifically, law-making institutions as such—organizational or customary, enacting or conventional—have nothing special to do with justice but are just as important to the system as the courts. This does not seem to me to be compatible with the Priority of Justice for the Law, understood as a thesis that applies to legal systems generally.
Second Horn of the Dilemma
The truth of the matter, of course, is that the Essentiality of Courts to the Law owes its plausibility precisely to the exclusion of nonmodern or nonmunicipal legal systems. Alas, when we focus on modern municipal legal systems (as I have implicitly done throughout), the purely customary system ceases to be a viable possibility. The claim that law-applying organizations are more important to the (modern municipal) legal system than law-enacting ones (legislatures) is no more defensible than the claim that they are more important than law-making institutions generally. Trying to imagine a municipal modern legal system without a legislature is no easier than trying to imagine it without courts.Footnote 29
C. The Purely Judicial System
Imagine another rebuttal of the simple thought. What is special about the courts is not that they apply rather than make law but that they are capable of doing both. Unlike any other legal institution, courts can both make law and apply it. We've mentioned those appellate courts that have the power to lay down general rules that are binding on the public at large. We can therefore imagine a purely judicial legal system such that some of its courts make all of its general legal rules by setting precedents, and others apply these rules in particular cases. The entire body of general legal rules in such a system is judge-made. Remarkably, neither law-enacting organizations nor customary law-making institutions are needed; the system consists of its courts alone. The case that brings out the primacy of the courts then is not the purely customary legal system but the purely judicial one. Its possibility shows that courts might be self-sufficient in a way no other legal institution can be.
However, by treating courts as law-making and law-applying institutions, this rebuttal no longer treats them as what they are essentially, namely, law-applying institutions. After all, not all courts have the power to make general legal rules. A priority-of-courts thesis is interesting only insofar as it implies that there is some function all law courts perform that makes them more important to the legal system than other institutions are. The simple thought with which we started suggests that there is no such function. The possibility of a purely judicial legal system leaves this impression intact. Far from showing that courts (law-applying institutions) are more important to the legal system than law-making institutions, it merely shows that they might be combined in a single hybrid institution.
The second rebuttal renders the levering argument equivocal. As we saw at the end of the last section, when courts are understood as both law-making and -applying institutions, the first premise—the Priority of Justice for the Courts—becomes untenable. To repeat: justice is not the first virtue of the courts as law-making and -applying institutions; it is only their first virtue as law-applying institutions.
The rejection of the Priority of Courts for the Law should give us some confidence in the stability of our twofold result: justice is the first virtue of the law courts but not of the law.
VI. CONCLUSION
A. Final Thought
If justice is not law's explanatory first virtue, what is? My discussion of primary legal duties might be taken to suggest that the law is essentially protective; its first virtue is to protect subjects, institutions, and goods from various forms of harm. But the scope of our discussion is too limited to support such a general conclusion. We have focused exclusively on legal duties. Perhaps more to the point, we have drawn exclusively on criminal and private law duties. I believe these simplifications are justified in the context of arguing against the Priority of Justice for the Law. To reject this thesis it is sufficient to show that some paradigmatic primary law duties have nothing special to do with justice. They are also justified in the context of arguing for the Priority of (Requitative) Justice for the Courts. Indeed, the possibility that the first virtue of criminal-law and private-law courts as duty-applying institutions can fail to reflect the first virtue of law courts as law-applying institutions generally seems remote.
I do not, however, believe that these simplifications would be justified in the context of arguing that protection—or anything else for that matter—is the first virtue of the law as a whole. I do not see how one can convincingly argue for such a general view without going beyond the jurist-friendly arenas of private-law and criminal-law duties. In particular, I have in mind legislation and laws that either do not impose duties at all or do not impose duties that are applicable in the courts. Think about the national budget, health care reform, social security legislation, and the like. It is hard to fit these important instances of law-making and laws into the conceptual apparatus of primary and secondary duties that serve us well enough in the discussion of private and criminal law. I see no good reason to suppose that legislation in general is essentially protective.
But my doubts are actually more general. “If not justice, what?” presupposes that the law has some explanatory first virtue or other: if not justice, then it must be protection, humanity, veracity, fraternity, community, efficiency, or whatnot. I think the odds are against this suppressed assumption—and consequently against the various attempts to crown some single virtue as law's first. Law is an imperialistic institution. Unlike churches, golf clubs, or academic institutions, the law reserves to itself the right to regulate every single aspect of human existence as it deems fit—including churches, golf clubs, academic institutions, and every other social institution. This is supported by the extreme diversity of branches of law (compare family law and insurance law) and of legal institutions (compare courts and legislatures).
Even assuming that each branch or institution of the law has a unique, explanatory first virtue, since there is no obvious reason to think that any one of them is more important to our understanding of the law than every other, there is also no obvious reason to think that any of these virtues is the unique, explanatory first virtue of the law as a whole.
B. Summary of Claims
In this paper I have recommended, with varying degrees of support, the following claims.
(1) Comparative justice is not the first explanatory virtue of the law.
(2) Comparative justice is not the first explanatory virtue of the law courts.
(3) Requitative justice is not the first explanatory virtue of the law.
(4) Retributive (requitative noncomparative) justice is the first explanatory virtue of the criminal law courts.
(5) Corrective (requitative comparative) justice is the first explanatory virtue of the private law courts.
(6) Requitative justice is the first explanatory virtue of the law courts.
(7) The courts are not the first institution of the law.
(8) The law has no explanatory first virtue.
C. John Gardner
My discussion of the relation between law and justice is indebted to Gardner's pioneering contribution to the subject over the past decade, and it would be useful to summarize what I take to be the main similarities and differences between our views. Of the eight claims above, I take it that Gardner basically accepts (1), (3), (5), and (8) and rejects (2), (4), (6), and (7). Gardner and I agree that justice—requitative or comparative—is not the first explanatory virtue of the law. I also believe that we fundamentally agree on the most general reason for this, namely, the lack of essential relation between primary legal duties or laws and justice—requitative or comparative.
Our main differences are two. First and more clearly, Gardner thinks that comparative justice is the first explanatory virtue of the law courts. I reject this in favor of the view that the first explanatory virtue of the law courts is requitative justice. The second difference is subtler but just as significant. Gardner clearly shares the common view that the courts are uniquely important to the legal system. So he accepts the Priority of Courts for the Law (and the Priority of Justice for the Courts) but rejects the Priority of Justice for the Law. To me this position seems unstable. I am inclined to believe that, if the courts really were uniquely important to the legal system, this would tend to support the Priority of Justice for the Law. However, I reject the Priority of Courts for the Law. This allows me to block the levering of the Priority of Justice for the Courts into the Priority of Justice for the Law.
