Law anü i'niiosopny (ZUiU) zy:i;>y-is/ DOI 10.1007/sl0982-009-9061-2
© Springer zuuy
ANTHONY R. REEVES
DO JUDGES HAVE AN OBLIGATION TO ENFORCE THE LAW?: MORAL RESPONSIBILITY AND JUDICIAL-REASONING
ABSTRACT. Judicial obligation to enforce the law is typically regarded as both unproblematic and important: unproblematic because there is little reason to doubt that judges have a general, if prima facie, obligation to enforce law, and important because the obligation gives judges significant reason to limit their concern in adjudication to applying the law. I challenge both of these assumptions and argue that norms of political legitimacy, which may be extra-legal, are irretrievably at the basis of responsible judicial reasoning.
What form of reasoning is appropriate to the judicial office? We might think, along with Chief Justice John Roberts, that judges are a bit like umpires,1 impartially interpreting and applying pre-established standards in the game of law, insofar as that is possible. Roberts does not do much to elaborate what he is getting at here, but his statement seems to strike a chord with a common understanding of how judges ought to approach adjudication. Judges are arbiters of what the law demands and, although arbitration requires judgment, judicial decisions ought to be fundamentally guided by the relevant legal standards. The
1 Now U.S. Supreme Court Chief Justice John Roberts compared judges to umpires during his senate confirmation hearings. "I will be open to the considered views of my colleagues on the bench, and I will decide every case based on the record, according to the rule of law, without fear or favor, to the best of my ability, and I will remember that it's my job to call balls and strikes, not to pitch or bat." U.S. Congress. Senate. Committee on the Judiciary. Confirmation Hearing on the Nomination of John G. Roberts, Jr. to be Chief Justice of the United States. 109th Cong., 1st sess., Sep. 12, 2005. Serial J-109-37.
160
ANTHONY R. REEVES
judge normally acts responsibly when she implements the law impartially despite any personal reservations (moral or otherwise) about the law in question. In other words, the mode of practical reason appropriate to the judicial office will be largely determined by the nature of legal norms in general and by the valid legal standards of a judge's jurisdiction. A judge is to exclusively rely upon the relevant legal standards (again, insofar as this is possible) to decide matters that come before the court.
One intuitive way to justify the above model of judicial reasoning is to claim that judges have a moral obligation to apply the law: what I will call an obligation of fidelity. A moral obligation of fidelity would, in effect, preempt the introduction into judicial reasoning considerations heterogeneous to the content of the law and its impartial application. If judges generally have a powerful moral reason to be in fidelity, then they can normally ignore moral concerns that are not legally recognized, since their obligation would usually outweigh or otherwise invalidate competing considerations. A judicial obligation of fidelity, moreover, seems initially straightforward and promising so much so, in fact, that it has tended to receive little attention even when the nature of responsible adjudication is the topic of discussion. Steven Burton, for example, in a book-length treatment of the ethics of judging, spends only five pages on the moral grounds of a judicial duty to uphold the law. He argues that consent, the judicial oath, and the fact that the judicial office is a public trust unambiguously de-mand fidelity on the part of the judge. In an extended discussion of defensible common law legal reasoning, Melvin Eisenberg briefly remarks that a clear-cut judicial obligation to the law "follows from the voluntary assumption and retention of judicial office" and that the office is "held in trust." The belief that judges are so obligated is, moreover, a common one outside academic discussions. One often hears politicians and citizens alike demand fidelity to the law from judges, and frequently on the grounds of their solemn oath to uphold the office.
Given the wide acceptance of a judicial obligation of fidelity and the significance of such an obligation for a theory of
2 Burton 1992, pp. 217-221.
3 Eisenberg 1988, p. 160.
JUDICIAL OBLIGATION
161
adjudicative reasoning, it is important that we become clear about the nature of this obligation. In this paper, I contend that the idea of a judicial obligation of fidelity amounts to much less than we might initially suppose. The central thesis is that no generally binding judicial obligation of fidelity exists, even where judges assume their office voluntarily and deliver a public oath. The conclusion I draw is not that judges never have reason to enforce standing law. In many circumstances, it would be inappropriate for judges to let their assessment of how a particular political question should ideally be resolved stand in for how it has in fact been resolved by existing legal institutions. Rather, my conclusion is that since judges are not presumptively justified in ignoring extra-legal moral concerns, fidelity to the law is not the proper starting point for judicial ethics. The 'umpire model' grossly oversimplifies the nature of responsible judicial reasoning and distorts the moral position of the judge, particularly in cases where a polity is at risk of committing a serious moral wrong. A successful account of judicial reasoning will have to account for the way in which a special class of moral norms, which may or may not be legally enshrined, ought to figure into adjudication.
The paper proceeds as follows. First, I define more precisely the character of the position against which I argue, i.e., what it would mean for a judge to have a special obligation of fidelity. Second, I examine the moral position of the judge, identifying some common characteristics of the role I take to be morally significant and clarifying the kind of theory of law my approach assumes. Third, I will consider and reject several possible rationales for the view that judges have a general, special moral obligation to implement the law. Finally, I draw some general conclusions from the argument and indicate what I take to be its importance for a theory of judicial reasoning.
I. THE SPECIAL OBLIGATION OF FIDELITY
Let us first examine more closely the notion of a special obligation of fidelity. By "special" it is meant that judges have a stronger obligation to the law than the average citizen. By "obligation of fidelity" it is meant that judges always have a
162
ANTHONY R. REEVES
practically significant moral reason to adjudicate as existing law requires. A "practically significant reason" is a reason for acting that, although capable of being defeated by other reasons, ought to figure into a person's deliberations about what to do. If I promise my brother to go hiking some weekend afternoon, I have a practically significant reason for doing so. It should figure into my decision-making because the reason created by the promise outweighs or countervails at least some of the array of reasons for doing otherwise that would justifiably figure into my deliberations were it not for the promise. If I promise to kill an innocent person, I do not (typically) create a practically significant reason for doing so I do not create a reason capable of outweighing reasons for refraining from killing that are significant without the promise. Promise or not, the reasons I have to not kill have the same force, and thus the promise should not figure into my decision-making.4
If a special obligation of fidelity could be vindicated, then responsible judges would, in general, simply need to be concerned with the impartial application of the valid legal standards of their jurisdiction without recourse to extra-legal moral principles. At least some significant scope of moral considerations would be rendered irrelevant to judicial practical reason (again, because the obligation would outweigh or otherwise invalidate these considerations). The view I argue against here is that judges always have a practically significant reason for enforcing the law. Judges do not have a general, special obligation of fidelity to the law that justifies the preemption of reflection on the moral quality of valid law in adjudication.
II. THE MORAL POSITION OF THE JUDGE
I approach the judicial office from a general standpoint, not simply from the perspective of Anglo-American legal systems. With that said, some morally important features that are commonly part of the judicial office are: de facto authority over
4 This kind of point has led many to claim that promises to do evil are not really promises. However, there is some dispute about this matter, so I will not assume that promises to do evil are incapable of producing any reason, however insignificant compared to other relevant reasons, at all.
JUDICIAL OBLIGATION
163
state action, finality of arbitration, legal expertise, and location within a scheme of governance that separates/shares powers and functions. I will briefly look at each in turn.
A. Judicial Authority
The deployment of state power to regulate people's affairs requires justification. When a state justifiably exercises its power, its action is what I will call "politically legitimate." My use of political legitimacy closely corresponds to Christopher Heath Wellman's. Wellman explains this idea as follows:
Political states coerce those within their territorial borders; if you are in country X, X threatens to punish you if you disobey its legal commands. An account of political legitimacy explains why this coercion is permissible. In doing so, it explains why the state has a right to coerce its citizens and correlatively, why its citizens have no right to be free from this coercion.5
A coercion right makes no claim about any corresponding moral obligation of citizens to abide by state directives.6 It merely asserts that the affected have no right to be free from the state's intervention. State power, however, may not be essen-
in
tially coercive, so I will understand political legitimacy to simply refer to an "enforcement right" rather than a coercion right. A state's deployment of power is politically legitimate when it justifiably enforces its directives. The idea of political legitimacy, then, picks out those moral norms relevant to assessing whether use of state power is justified.
Judicial decisions, insofar as they are recognized by state institutions, are occasions for the use of state power. Given that a court's decision will influence how other state institutions behave, judges have de facto control over how the state will deploy its power to regulate the affairs of those within its jurisdiction. State regulation of people's affairs requires
5 Wellman 1996, pp. 211-212.
6 Noticing this here should help us avoid ongoing disputes about whether legitimate authority implies a political obligation on the part of subjects to abide by that authority. For a discussion of this issue, see Edmundson 1998a. See also, Simmons 1999.
7 Edmundson 1998b, pp. 73-124.
164
ANTHONY R. REEVES
justification: the use of force by the state is something that must be morally accounted for, i.e., it must be legitimate. Judicial decisions, therefore, should meet conditions of political legitimacy. The conclusion to draw from this is that judges should be concerned that they are deploying that power legitimately. Very simply, since their decisions are typically occasions for the use of state power, judges are responsible for considering moral norms of political legitimacy in the course of adjudication.
B. Finality
Judicial responsibility for the political legitimacy of court decisions might be deferred if there is some other institution that can easily review, through normal and readily available channels, those decisions. This would, to some extent, shift the moral burden elsewhere. To what extent this is the case will vary from polity to polity, but usually a court's decision is final, even if the possibility of appeal exists. Lower court decisions in a large legal system rarely receive attention from appellate courts, and thus even a lower court must accept the likelihood that their decision is the final determination of how the state will treat the matter (at least for the near future). The moral situation is even starker in constitutional and high courts, whose decisions are not reviewable and can be extremely difficult to reverse through other state processes. Since a court's decision is very likely the final arbitration of some matter to be regulated by the state, a judge is responsible for the political legitimacy of its decision.
C. Legal Expertise
We should not assume that judges are morally enlightened or above average moral thinkers, but judges typically are legal experts. Mature legal systems are sophisticated phenomena, and understanding what is valuable about a legal system, or even some department of it, frequently requires detailed knowledge of how it operates. It is this legal knowledge that gives a judge an advantage in assessing the political legitimacy of law, not any advanced moral knowledge of political legitimacy.
JUDICIAL OBLIGATION
165
The judge has an above average knowledge of law and thus is better able than most citizens and officials to understand how a legal order may be justified.
A simplified example may help. Tort judgments enforcing rules of strict liability seem to violate strong intuitions most of us have about being responsible for rectifying an injury. Where a rule of strict liability is enforced, a plaintiff must simply show that the defendant (by act or omission) caused the harm, not that the defendant is culpable for that harm. This conflicts with the standard moral view that someone ought to be at fault for an injury to be liable for its remediation. However, once one recognizes the circumstances in which strict liability is actually applied and the overall effect of a system of strict liability, one is in a position to identify competing rationales that might justify strict liability rules in appropriate situations. Strict liability serves to discourage behavior inherently dangerous to others and, in terms of product liability, rationally and fairly shifts the cost of harms to manufacturers. The point, I hope, is straightforward. Judges are in a position to assess the political legitimacy of legal requirements.
D. Separation of Powers and Functions
Importantly, the judicial office is often situated in an institutional scheme that divides governmental decision-making and functions amongst various branches of government the judiciary normally being assigned the task of law application. Such a scheme would appear to relieve a judge of a great deal of responsibility for the political legitimacy of her decisions, since it is not the court's place in the system of governance to develop the standards and policies that are to guide a state's behavior (again, insofar as straightforward application of the law is possible). In fact, rule-focused judicial reasoning may help maintain a desirable separation of powers. Frederick Schauer8 and Tom Campbell,9 for example, argue that rule-based legal-reasoning prevents administrators of the law from intruding
* Schauer 1991.
9 Campbell 1996. See also Campbell 2004.
166
ANTHONY R. REEVES
upon the rightful authority of legislators to create government policy. An ethic in the legal profession of applying rules, irrespective of any personal reservations, protects the democratic character of a polity by removing the personal preferences of officials from the determination of how political power is to be exercised. As Campbell puts it,
[0]nly if democracy is centred on the choice of rules can it begin to approach the ideal of providing real political power to the people as a whole. If the rules thus created are not followed or are subverted by processes that enlarge judicial discretion at the expense of rule-governed decision-making then democracy is thereby diminished.10
Ideally, democratic legislatures make determinate and clear law that judges follow closely in implementing state directives.
A democratic rationale for rule-based adjudication cannot, however, relieve judges of a responsibility to ensure that their decisions are politically legitimate. First of all, considerations of this sort are institution-specific. They apply only when the actual procedures and institutional arrangements of a polity can ground a claim that decisions of certain types are the rightful province of a non-judicial body. For democratic separation of powers considerations to be relevant, a government must be sufficiently democratic in actual practice. The official self-understanding of a government is, moreover, insufficient. Although the statutory law of apartheid-era South Africa was enacted by an elected legislative body, it would be implausible to call such law democratic given the exclusion of large sections of the population from the political process. The same point applies to any argument that purports to limit the concern of judges based on the characteristics of some other law-making body (e.g., that the body is procedurally just, democratic, or epistemically privileged). Existing law-making bodies must actually possess the relevant characteristics for the argument to be relevant. Thus, judges are at least in the position of considering whether actual political and legal practices possess characteristics that warrant deference.
Campbell 2004, p. 119.
JUDICIAL OBLIGATION
167
Second, even where the institutional arrangements of a polity do warrant deference from judges concerning some matters, a judge must still determine whether the regulating institution acted within the scope of its legitimate authority. My point is not essentially a constitutional one, though the existence of a constitution might be relevant. Rather, the point is that even properly conducted legislative or executive decisions can only go so far in morally justifying state use of power. Even the most democratic decisions can demand what they have no right to demand, which is simply to say that democratic procedures do not automatically imply an enforcement right. Where there is some question about the legitimacy of the government's regulation of some matter before the court, a judge is responsible for considering whether the government's demands are, in fact, politically legitimate. Insofar as a judge's decision will determine how the state actually treats the matter, and given the other aspects of the judge's position outlined above, the judge ought to be concerned that her decision is politically legitimate that she does not direct the state to act in ways it has no right to act.
Before directly examining the plausibility of a special obligation of fidelity, I should clarify some assumptions implicit in my discussion here about the nature of law and the state. First, I assume that valid law does not automatically meet conditions that would justify its enforcement. This assumption is certainly consistent with legal positivism (both inclusive and exclusive), but also with at least some versions of natural law.11 Showing that a norm is legally valid or derivable from dispositive legal standards is one matter, showing that it ought to be enforced is
1 O
another. Second, I am assuming that states can, have, and do occasionally demand what they have no right to. I take this to be true not simply in wicked regimes, but also in states we might give that ambiguous designation of 'minimally just.' Even minimally
11 See Murphy 2003.
12 For a general discussion on the gap between legal derivability and moral defensibility, see Lyons 1993b. Thus, irrespective of whether one has to occasionally rely upon moral reasoning to identify the law, an ongoing dispute even amongst positivists, I assume that it is possible for some moral standards of political legitimacy to be extra-legal.
168
ANTHONY R. REEVES
just states can violate moral rights, institutionalize unjustifiable inequalities, regulate realms of morally protected personal liberty, act in great detriment to the common good, or legislate in a manner that compromises a claim to democratic legitimacy. If these assumptions are sound, a judge cannot assume that the state's legal requirements are politically legitimate.
III. AGAINST JUDICIAL OBLIGATION
So far, I have tried to characterize in general terms the moral position of the judge. As it stands, the judge appears to be fundamentally responsible for the political legitimacy of his decisions and, because of this, judicial reasoning will have at its foundation norms of political legitimacy even if such norms do not form a part of the valid law of his jurisdiction. Now we can see quite clearly the practical significance of a special obligation of fidelity. If judges have an independently grounded and weighty obligation to enforce standing law, it could preempt judicial consideration of political legitimacy in the normal course of adjudication. I will now consider three possible justifications for a special obligation of fidelity and argue that they are unsuccessful.
A. The Judicial Oath of Office
One straightforward ground for a special obligation of fidelity is the judicial oath of office. Usually, judges must publicly swear to uphold the laws, defend the constitution, impartially administer justice under the law, discharge the duties of the office and the like. Such an oath is an initially plausible ground for an obligation strong enough to preempt extra-legal moral reasoning about issues before the court. When one makes a promise, one creates morally forceful reasons to carry out the terms of promise, reasons different from whatever other considerations might lead one to act in that way. Moreover, upon promising, certain other reasons (e.g., prudential ones) seem to be preempted, or at least outweighed, in favor of carrying out the promise. When I promise to take my brother hiking on a certain day, the fact that by forgoing I can make some progress
JUDICIAL OBLIGATION
169
at work does not, under normal circumstances, seem to matter. I should take my brother hiking. Therefore, since most judges are required to take an oath to uphold the law upon assuming their position, we seem to have a strong basis for the kind of special obligation that would support an umpire-like approach to adjudication.
Of course, certain limiting conditions apply to promises. For example, we generally recognize that a promise is not binding, or at least is made very weak, if made under coercion or when one is not in the right state of mind. Thus, if any coercive elements are part of the judicial oath, it would undermine its obligatory force.
Nonetheless, let us leave these two conditions aside, as we might be inclined to think that for most judges, the oath of office is undertaken freely and in the right state of mind. Another oft-suggested condition that concerns the content of the promise, rather than the circumstances under which it is made, requires more thought. Sometimes it is claimed that promises to do what is morally objectionable do not give rise to obligations. As A. John Simmons puts it, "A promise to aid
1 ^
[a man] in his villainy, of course, would not bind us." If promises are limited by the moral worth of their content, then it seriously damages the judicial oath's ability to ground a special obligation of fidelity. Such an oath would only create an obligation of fidelity where the law that the judge has promised to enforce is independently morally justified, i.e., where the judge is not made party to an illegitimate use of state power. Crucially, then, an oath would be unable to preempt (extralegal) ethical reasoning in responsible adjudication since its obligatory force is contingent upon its coherence with other moral requirements. A judge would have to assess the political legitimacy of the law relevant to the case at hand in order to take seriously obligations generated by her oath of office.
Consequently, the important question appears to be: do promises to do what is unethical on other grounds create reasons for fulfilling the terms of the promise? The answer is not indisputably negative. Margaret Gilbert argues that promises
13 Simmons 1979, p. 78.
170
ANTHONY R. REEVES
to do what is immoral still have obligatory force since they involve commitment to a joint enterprise that is not structurally different than promises to do what is morally permissible. In both cases, an expression of readiness is made, in common knowledge between the parties, to be jointly committed to a project or course of action. Our intuition that promises to do wrong are not binding is better accounted for by the fact that the obligations engendered by such promises are generally outweighed by other moral considerations.14
The dispute over whether promises to act immorally fail to generate obligations or simply have their corollary obligations outweighed need not detain us. Our question is the practical one of how a person should regard a promise to do what would otherwise be unethical. (Again, if a promise is morally forceful even where its terms require what would otherwise be a moral wrong, then the judicial oath might justifiably preempt considerations of political legitimacy from judicial practical reason.) The way of getting to the heart of this issue is, I think, by considering the following question: Is there any action that, being otherwise immoral, is made morally permissible by the existence of a promise to so act? In other words, can we conceive of any morally indefensible action that is made morally worthwhile for some person by the person's promise to do it? If we cannot, then either the obligations generated by promises are too weak to outweigh other moral concerns or promises to do what is immoral do not generate obligations. In either case, a promise to act unethically should be disregarded from a practical standpoint. In answer to the question, I have to simply confess that I cannot think of any action that is immoral unless it is the object of a promise.
Nonetheless, let us consider an initially plausible candidate.15 Imagine that my friend Jared and I go to the casino. Jared has a gambling problem. Jared takes SI00 out of the ATM, gives me his debit card, and I promise Jared that I will not give the card back to him tonight, even if he asks for it. Given this promise, should not I refuse Jared the card when he
14 Gilbert 1999, pp. 241-248.
15 My thanks to Simon Keller for pointing out this example.
JUDICIAL OBLIGATION
171
asks for it later that night? Would not this act preventing Jared from getting access to his own money be morally impermissible, were it not the object of a promise?
Despite initial appearances, the answer is: no, the act is not made permissible by the promise. We should notice two background features of the example that do the work of grounding the moral permissibility of withholding the debit card. First, Jared presumably consents to me withholding his card. Consider if he had not consented to my keeping the card but, rather, handed me the debit card to hold while he pulled out his wallet. After handing me the card, I promise not to give it back to him. This promise would hardly justify withholding his debit card when he demands it back moments later. Consider a second scenario: he consents to my keeping the card, but I do not promise to. He says, "Please hold on to the card until tomorrow." I respond, "Maybe, I will hold onto it for now, but I do not want to make you my responsibility." He shrugs his shoulders and says, "Do what you think is best." Later, when he asks for his card, it would seem permissible for me to withhold it even though I had not promised to do so. If I had promised, I would be obligated to withhold, rather than being simply permitted. The point, though, is that the presumed consent in the example makes the promissory obligation capable of producing practically significant reasons.
Also notice that substantive goods may make it morally permissible to withhold the debit card. As Jared's friend, I recognize his gambling problem and that he tends to make poor financial decisions (decisions he later regrets) after he begins gambling. After he loses his initial SI00 gambling, I may be justified in withholding the card whether I had promised to do so or not, especially if the threat to his financial welfare is great. Much like withholding borrowed weapons from an insane friend,16 I can keep the card if I genuinely expect Jared to seriously hurt himself or others (e.g., his family). If this is the case, then it suggests that independent moral considerations are establishing the moral permissibility of withholding the debit card. Of course, as I just mentioned, the promise may create
16 Plato 2004, p. 5.
172
ANTHONY R. REEVES
additional (practically significant) reasons so long as the terms of the promise are permissible on other grounds.
If we are unable to construct a situation in which a normally objectionable action is vindicated by a promise outside of legal practice, then I see no grounds for asserting that such vindication is possible when it concerns the judicial oath. Even a promise to humiliate someone in private does not seem to justify doing so. Given the fact that judicial decisions usually have outcomes that are far more significant, it is hard to see how a judge's promise to abide by the law, when the relevant portion of the law requires what is immoral, can generate obligations that are important from a practical standpoint. Law must be politically legitimate before it can be the subject of practically significant promissory obligations. The judicial oath of office, therefore, is incapable of supporting a special obligation of fidelity to the law.
B. Role Identification and Judicial Obligation
On one common view of the moral significance of socially defined roles, a role requirement does not amount to a moral requirement. The mere fact that some course of action is required by a role does not, by itself, give a purported participant in the role a (moral) reason to so act. Rather, an independent moral justification of the course of action or of the role itself is necessary to show that one should act as the role requires. Simmons offers a clear statement of this view:
The existence of a positional duty (i.e., someone's filling a position tied to certain duties) is a morally neutral fact. If a positional duty is binding on us, it is because there are grounds for a moral requirement to perform that positional duty which are independent of the position and the scheme which defines it. The existence of a positional duty, then, never establishes (by itself) a moral requirement.17
Whatever sanctions or disapproval that departure from a defined role may engender, it is no less the case that social roles are a proper object of ethical evaluation.
Simmons 1979, p. 21. His emphasis.
JUDICIAL OBLIGATION
173
Simmons' view of the significance of role requirements is well-motivated. First, what turns out to be a socially recognized role is a rather arbitrary matter from a moral standpoint. Morally objectionable but socially established roles have not been uncommon historically, as seen in cases like the slave trader, Spanish Inquisitor, Gestapo official, concentration camp guard, and, to use Arthur Applbaum's well-illustrated example, Executioner of Paris.18 It is hard to see how the social establishment of these roles adds any measure of defensibility to their requirements.19 Second, where a role requires what is worthwhile, or where failure to fulfill the role is blameworthy, it is easy to explain the importance of fulfilling the role in terms of independent moral requirements. Taking Simmons' example, an army medic who shirks his duties to spend a day at the bar during wartime is certainly morally culpable. However, his duty to attend to the wounded can be explained in terms of a moral requirement to help those in dire need, rather than the mere existence of the socially defined role, since the medic is one of the few present capable of treating major battle injuries. Third, when failure to fill relevant role requirements does not
18 Applbaum 1999, pp. 15-42.
19 Ibid, at pp. 3-75.
20 Simmons 1979, pp. 18-20. Of course, social roles may divide up duties in some social scheme so as to give particular individuals responsibility for performing certain tasks. So long as this division of labor and the scheme itself are justifiable, the content of the assigned roles may determine what is morally required for particular individuals within a certain sphere of action. A soldier assigned to delivering supplies to an army hospital is responsible for doing so because she is the one assigned - not because she has specialized skills. Everyone expects her to do it, no one else will do it if she does not, and it is important that it get done. Nonetheless, the scheme and the content of the role still require independent justification. See also Luban 1988, pp. 104-147. On Luban's analysis, a role occupant is in a position of assessing the value of the institution of which they are a part, the significance of the role in achieving that value, the essentialness of the particular obligation to the role, and whether an act is really required by an obligation on particular occasions of acting. The agent has to determine whether the strength of the justifications for carrying out the role, in any given situation, outweigh any competing moral considerations. If a role requirement lacks any justification in the circumstances, it makes no claim on the agent. Thus, I consider a position like Luban's "liberal" in the sense described below.
174
ANTHONY R. REEVES
result in outcomes that are otherwise morally significant, it is difficult to say what one has done wrong by that failure. If no one's moral rights have been violated, if no one has been endangered, and if no one's interests have been affected for the worse, it is hard to specify the wrongness of violating the role requirements.21
I will call the view that socially accepted roles have no inherent moral worth the "liberal conception of roles." The liberal conception has recently been challenged by a number of moral and political philosophers, including Alistair Macln-tyre, John Horton, Michael Hardimon, Justin Oakley, and Dean Cocking. I will focus here on Hardimon's account of role obligations as it does not depend upon the acceptance of a particular theory of normative ethics and because it summarizes strands of thought present in the other works. Hardimon contends that social roles can be a genuine source of obligations. Hardimon defines a role obligation as "a moral requirement, which attaches to an institutional role, whose content is fixed by the function of the role, and whose nor-mative force flows from the role." Importantly, then, the content of the role is not merely conventional, but is interpretive with regard to function, "which is to say that people can reasonably argue about the proper interpretation or under-standing of role terms or concepts." Citing Ronald Dworkin, Hardimon appears to model the interpretation of role requirements on constructive interpretation, which requires construing some item or practice as best serving the values that the item or practice is taken to serve. As Hardimon explains, "[I]n order to settle what the requirements of a given role are, it is often
21 Simmons 1996a, pp. 269-270.
22 Maclntyre 1981.
23 Horton 1992.
24 Hardimon 1994.
25 Oakley and Cocking 2001.
26 Hardimon 1994, p. 334.
27 Ibid, at p. 336.
28 Dworkin 1986, pp. 49-55.
JUDICIAL OBLIGATION
175
necessary to determine how the institution of which it is a part is   or ought to be - structured."
Before moving on to the rest of Hardimon's account, I should note that making the content of role requirements dependent on constructive interpretation undermines the thesis that the normative force of role obligations flows from the role itself. If what a role really requires, rather than what it merely appears to require, depends on how it can best serve some independent value, then the normative force of the role would seem to derive from that value. Apparently, according to Hardimon, conventional understandings of the role that cannot be construed as serving the relevant value are not genuine elements of the role and do not give rise to role obligations. The independent value, then, is doing all the work in establishing obligatory force, not the social recognition of the role.
That independent values account for the obligatory power of the roles Hardimon has in mind is further confirmed by a condition he places on role obligations. In order to ground such obligations, a role must be "reflectively acceptable." Hardimon explains:
To say that a social role is reflectively acceptable is to say that one would accept it upon reflection. Determining whether a given social role is reflectively acceptable involves stepping back from that role in thought and asking whether it is a role people ought to occupy and play. Determining
Hardimon 1994, p. 337. My emphasis. A similar constraint on the content of professional roles is suggested by Oakley and Cocking. "Broadly speaking, what counts as acting well in the context of a professional role is in our view importantly determined by how well that role functions in serving the goals of the profession, and by how those goals are connected with characteristic human activities. That is, good professional roles must be part of a good profession, and a good profession, on our virtue ethics approach, is one which involves a commitment to a key human good, a good which plays a crucial role in enabling us to live a humanly flourishing life...Thus, in order to generate a defensible professional ethic, the norms of the profession in question cannot simply be taken as given; rather they must be shown to reflect a commitment to an important substantive human good that contributes to our living a flourishing human life." Oakley and Cocking 2001, pp. 74-75.
176
ANTHONY R. REEVES
that a given social role is reflectively acceptable involves judging that it is (in some sense) meaningful, rational, or good.30
The motivation for placing this constraint on role obligations is clear: otherwise one is committed to saying that morally perverse roles are a source of obligations. As Simmons points out, however, insisting on this condition is tantamount to
n i
admitting that social roles require external justification. In considering reflective acceptability, we are considering whether a role is independently valuable.
One might drop the above constraints Hardimon places on the content of role obligations, so we should consider the arguments Hardimon offers for the existence of obligations 'whose normative force flows from the role.' In his discussion of contractual role obligations, which arise in the consensual adoption of a role, Hardimon identifies two sources of obligation: voluntary acceptance of the role requirements and individual identification with the role. Since the judicial oath was our last topic, I will now focus on role identification. Hardimon lays out three conditions for identifying with a role. First, one must occupy the role, and this is presumably a matter of social fact. Second, one must recognize (correctly) that one occupies the role. Third, one must "conceive of oneself as someone for whom the norms of the role function as rea-sons." When one identifies with a role in this way, there is a basis for role obligation independent of one's consent to honor the role.
The key component of the idea of identifying with a role is that of conceiving of oneself as someone for whom the norms, that is, the evaluative standards associated with a role, its rights, duties, virtues, ideals, and supererogations, have reason-giving force. Here is the idea. If you identify with a role, its norms will function for you as reasons. If you are a judge who identifies with the role of judge, the fact that this is something judges do (in the normative sense) will give you a reason for doing it.33
30 Hardimon 1994, p. 348. His emphasis.
31 Simmons 1996b, p. 31.
32 Hardimon 1994, p. 358.
33 Ibid. His emphasis.
JUDICIAL OBLIGATION
177
Role obligations flow from a role itself when one correctly identifies with the role.
As it stands, the above line of reasoning is fallacious. The third condition for role identification is a psychological one, i.e., it simply concerns facts about one's beliefs. It does not follow from the fact that one believes oneself to have a reason for acting in a certain way that one really has a reason for so acting. As Simmons colorfully illustrates, "A person who believed himself to be Napoleon could not intelligibly deny his obligation to, say, lead the French army, but this would not show that this person in fact had a moral obligation to lead the French army."34 Of course, such delusion would fail to meet the first condition of role identification, actually filling a social role, but the central point is no less pertinent. Typically, belief in an obligation is insufficient for substantiating an obligation. One can be wrong about what one has reason or is morally required to do.
Maybe, though, there is something morally important about retaining one's role identities, important in a way that makes one's beliefs concerning role obligations different from other moral beliefs, e.g., whether human cloning is morally acceptable. The fact that the former are important elements of my self-conception, whereas the latter are not, may be significant. Given the context of Hardimon's argument, this seems to be the right way to understand his claim. In an earlier discussion of the non-voluntary roles of citizen and family member, Hardi-mon emphasizes that such roles are an essential element of our identity and to give them up "would require a radical revision of our self-conception." Such revision is costly to the individual; it leaves a person worse off in morally significant ways. "For those of us who are committed to the forms of ethical life lived within the family and the state, giving up our conception of ourselves as family members and citizens would...carry a tremendous normative price."    Since role requirements are
34 Simmons 1996, pp. 262-263.
35 Hardimon 1994, p. 346.
36 Ibid, at p. 347.
178
ANTHONY R. REEVES
constitutive of the role, one cannot abandon them without paying this price. Given one's identification with these roles, then, one has reason to act as they stipulate, independent of any external value the roles serve. The social recognition of, and individual identification with, a role is a genuine source of moral requirements.
The point is similar with voluntarily assumed roles. When one conceives of oneself as a judge, for example, abandoning this conception by violating the role requirements of that position carries a cost for the individual albeit not a cost so great as abandoning more basic roles of citizen and family member. In this sense, therefore, one has reason to act as the role requires simply because one conceives of oneself "as someone for whom the norms of the role function as reasons." No non sequitur here. Moreover, such reasoning is not unprecedented. Joseph Raz has also argued that individual commitments and choices can create reasons for a person. Raz contends that committing to a way of life can help determine what counts as a good for oneself. In pursuing a career in philosophy, I create new standards of success and failure that may not have been relevant otherwise, e.g., publishing, becoming an effective teacher, treating students fairly, etc. These would not have been important, at least not in the same way, had I chosen to become a cemetery sexton. "In embracing goals and commitments, in coming to care about one thing or another, one progressively gives shape to one's life, determines what would count as a successful life and what would be a failure."40
Even if the above explication does somewhat rehabilitate Hardimon's line of thought, it does little to establish role identification as a substantial source of moral obligation. The problem is familiar. Social roles can require action that is morally objectionable. In such cases, the fact that the role is socially recognized, and that the person who occupies the role identifies with it, seems to add little to the role requirements' defensibility. A military general charged with forcing the
37 Ibid, at pp. 345-351.
38 Ibid, at p. 359.
39 Ibid, at p. 358.
40 Raz 1986, p. 387.
JUDICIAL OBLIGATION
179
migration of an ethnic minority may identify with his role and even believe that his action is justified. This, however, tells us little about how we should regard the general's actions from a moral standpoint. Of course, we may insist with Hardimon that the role itself must be justified or valuable in some way, i.e., reflectively acceptable,' in order to give rise to role obligations. However, apart from the fact that this seems to implicate external moral principles as the source of obligatory force, it leaves a role requirement in the weak position of being both unnecessary and insufficient for the existence of an obligation. It is clearly not necessary since we may have a duty even if there is no applicable role. Moreover, it is insufficient since the role must satisfy the demands of external moral principles to give rise to obligations. On Hardimon's account, genuine role obligations grounded in one's self-conception will always be coextensive with what is deemed morally acceptable by other ethical considerations. This makes role obligations a much less interesting source of moral requirements.41 It certainly leaves them incapable of grounding a special obligation of fidelity that preempts moral reasoning in adjudication since external moral considerations must validate a judicial role requirement before it becomes relevant.
One might try a different tactic by dropping reflective acceptability as a condition of role obligations. In cases where a role requires what is otherwise morally indefensible, it is not that role requirements fail to give rise to obligations (based in personal identification with the role); rather, it is that external ethical principles frequently outweigh the role obligation. The aforementioned general has a genuine obligation to carry out
If role requirements do (independently) give rise to obligations when those obligations are morally defensible on other grounds, then they may be powerful enough to give one reason to select one morally acceptable alternative over another. In fact, this seems to be the kind of reason-making of personal commitment that Raz has in mind (see The Morality of Freedom, pp. 378-381). However, this does not help substantiate a general judicial obligation that would normally preempt non-legal moral reflection in adjudication since a decision must first be shown to be defensible on independent moral grounds before role obligations enter the picture.
180
ANTHONY R. REEVES
the commands of his superiors, but his role obligation is overridden by the grave moral wrong of forced migration.
Although role identification is now sufficient for establishing obligations, there is still a question of their practical significance. If the role requires a grave moral crime, it is difficult to discern the difference between a role obligation having no weight versus it having some, but limited weight. Let us take a different case. The general is ordered to extract an 'ethnic tax' from the minority of five hundred dollars. Here the moral wrong is much less grave, but still somewhat substantial. Do the role requirements of the military general mitigate in any way the moral wrong of extracting an ethnic tax? Would the general be in any way justified in carrying out his orders if he can get away with not doing so? It appears not. It seems, rather, that we would expect the general to do what he can to reduce the injustice of his orders. If this is right, though, it speaks heavily against the strength of role obligations since they are not significant enough to mitigate even this more limited moral wrong.
We can make the moral crime even less severe by having the general ordered to extract a ten dollar ethnic tax. Would the general be justified in not extracting the tax if he is able to get away with not doing so? If we answer in the affirmative, then this strongly suggests that role obligations have no weight or extremely limited weight since they do not counter even nearly inconsequential injustice. I am inclined to agree, then, with the liberal conception of roles. Even if the liberal conception is wrong in the fine details, i.e., even if role identification is a source of some moral obligation, it is still the case that role obligations are of such limited weight that, practically speaking, they are nearly insignificant. Of course, the judge's role will be different than that of a general, but the central point here is that the identification with a role, by itself, is not a source of powerful moral obligations. It certainly does not carry enough force
42 There may be practical issues for the general. Maybe the general cannot avoid doing as ordered without detection, and this puts him and his family at risk (in ways other than possible loss of a sense of self). Let us assume, though, that the general can get away with not extracting the tax or extracting less than demanded.
JUDICIAL OBLIGATION
181
to support a judicial obligation of fidelity to law (of the kind that preempts moral reasoning) since even small injustices counter the normative force of whatever role obligations are applicable. Judges must consider the legitimacy of their role requirements, including the administration of the law, on any particular occasion. In considering the topic of judicial responsibility, therefore, the liberal conception of roles ought to be our operative conception. Whatever way it distorts our moral experience is unimportant from a practical standpoint of how judges should act in their official capacity.
C. Formal Justice and the Rule of Law
Another possible route to a special obligation of fidelity is the idea of fair and impartial administration.43 The idea is this: whatever the substantive merits or injustices of legal standards, there are issues of justice relating to administration itself such that whenever officials depart from a valid legal rule relevant to the case at hand, they commit an injustice by failing to implement it. Consider our expectation that judges apply the law without regard to their own interests, prejudice towards any of the involved parties, or caprice. Judges are to treat like cases alike and all those that come before the court equally to do otherwise is to treat at least one party unjustly. The idea of administrative, or "formal," justice seems to explain our expectation here. This is not to deny that other substantive moral concerns could outweigh the requirements of formal justice. The claim is simply that some injustice necessarily results from official departure from valid legal standards.44 If formal justice is to serve as a basis for a special obligation of fidelity, then it must produce a practically significant reason for
For a review of the adherents of formal justice, see Lyons 1993a, pp. 14-21.
44 Formal justice is different, then, from procedural justice. Procedural justice concerns moral standards for the assessment of procedures that regulate some domain of affairs. If established institutional procedures (e.g. those of a criminal trial) do not meet the relevant standards, then departure from those procedures would not give rise to claims of procedural injustice.
182
ANTHONY R. REEVES
impartial judicial application of the law generally. We can easily see that it does not.
First, as David Lyons points out, there are other ways of explaining the injustice of official partiality or prejudice than through formal justice. If a person is punished abnormally strongly and illegally upon conviction because of their skin color, the injustice might consist in substantive factors concerning the moral relevance of skin color, and of how individuals should be treated generally, rather than official departure from legal requirements. In fact, it seems likely that principles of non-formal justice or morality generally are at work in criticisms of unjust partiality in administration given that such concerns do not arise when no one's moral rights are violated and no one is harmed by official departure. As Lyons points out, "it is difficult to imagine a case in which injustice seems intuitively to be done by official deviation which has no adverse effect upon anyone at all."45
Second, it is difficult to see that our conviction that 'like cases be treated alike' attaches itself to institutional norms simpliciter rather than institutional norms that are morally important or, at least, morally acceptable. Lyons' discussion of the topic is again worth following. Consider what the maxim 'treat like cases alike and different ones differently' amounts to. As Hart notes, the maxim tells us very little by itself. Individuals are similar and different in a variety of ways, and until we determine what similarities and differences are relevant, the maxim offers no guidance. Since identifying what counts as a relevant similarity or difference can be a difficult and contentious matter generally when questions of justice arise,46 we might be tempted by a legalistic solution. In the case of administering the law, the problem of determining relevancy is avoided since the law itself specifies what similarities are salient for equal treatment. Justice requires treating like cases alike, and the law stipulates relevant similarities and differences for equal treatment. We may be tempted to conclude, then, that where there is law that applies to a case, some manner of real
Lyons 1993a, p. 23. Hart 1994, p. 159.
JUDICIAL OBLIGATION
183
justice is only achieved through impartial application of it. Hart makes some suggestion to this effect, and we get a stronger endorsement of this kind of claim from Chaim Perelman.48
If we are tempted to draw this conclusion, we should resist the temptation. Even if part of what justice requires is equal treatment, this does not imply that any stipulation whatsoever of relevant similarities and differences will realize the requirements of justice. As Lyons puts it, "[T]his argument begs the question at issue, which is whether the pattern of treatment prescribed by law is identical (or even compatible) with the pattern required by justice."49 The stipulations of the law must realize substantially the requirements of justice in order for claims of injustice to arise out of unequal treatment under the law.50 If the claim is that one part of justice is equal treatment and, since fidelity to the law always meets this requirement formally, justice is partially realized when the law is applied impartially, one is simply confusing necessary and sufficient conditions.51 Frederick Schauer puts the same point in terms of rule-based decision making. "Insofar as factors screened from consideration by a rule might in a particular case turn out to be
Ibid, at p. 160. "In certain cases, indeed, the resemblances and differences between human beings which are relevant for the criticism of legal arrangements as just or unjust are quite obvious. This is pre-eminently the case when we are concerned not with the justice or injustice of the law but of its application in particular cases. For here the relevant resemblances and differences between individuals, to which the person who administers the law must attend, are determined by the law itself. To say that the law against murder is justly applied is to say that it is impartially applied to all those and only those who are alike in having done what the law forbids; no prejudice or interest has deflected the administrator from treating them 'equally'." It is not clear that Hart would endorse the final step in the line of reasoning I outline above since he recognizes that justice is not identical with conformity to the law.
48 See Perelman 1963, pp. 1-60. Importantly, the context of Perelman's view is a belief in the arbitrariness of moral judgment.
49 Lyons 1993a, p. 29.
50 Of course, the law does not have to directly stipulate the requirements of justice. Sometimes a law will contribute indirectly to the justice of a society while in its application appearing somewhat unjust. Such is often the case when a standard of strict liability is applied in torts.
51 Lyons 1984, pp. 82-83.
184
ANTHONY R. REEVES
those necessary to reach a just result, rules stand in the way of justice in those cases and impede optimal justice in the long term." Justice does not simply require equal treatment, it requires equal treatment with regards to morally relevant similarities. Whether the law is substantially just or is instrumental in achieving justice is a contingent matter.
Lyons' discussion of formal justice brings into relief just how misleading the characterization of judges as umpires is. Umpires operate in a context where the rules and standards of the game are, presumably, fair and voluntary (joining the game is optional). It is these features of the umpire's position that successfully ground the claim that umpires ought to apply the pre-existing rules of the game in the course of making rulings. If playing the game were mandatory, the rules unfair, and the outcomes determinative of how people's rights and interests would be respected, it becomes much less clear why the umpire ought to be primarily concerned with impartial rule application. Concerning the judge, she cannot assume that existing law warrants enforcement, that people consented to be governed by it, or that people's rights and interests will not be affected in a way that requires moral justification. Given this, and the fact treating like cases alike is only a necessary condition for justice, the maxim cannot substantiate a general, special obligation of fidelity.
IV. CONCLUSION
I conclude, then, that we have little reason for thinking that judges have a general, special obligation to enforce standing law and abide by existing legal practice. Judges do not always have a practically significant reason to adjudicate in fidelity to the law. The importance of this conclusion is fourfold. First, in some situations, a judge will simply have no obligation of any kind that is actually relevant to decision-making to uphold the law. If the judge can successfully mitigate the injustice of some law through unlawful adjudication, and if the nature of the existing legal institutions do not produce some procedural
Schauer 1991, p. 137.
JUDICIAL OBLIGATION
185
reason for fidelity (e.g., because they are genuinely democratic and are operating within the scope of their proper authority), then there is no moral dilemma for the judge in terms of an obligation of fidelity. The judge has no (practically significant) reason to enforce the law simply because it is law or because she occupies the judicial role.
Second, judicial fidelity to the law is not presumptively justified. Only when the law itself is justified (on procedural or substantive grounds) are judges bound to enforce it. Judges cannot simply assume that by enforcing law, they are acting responsibly. Third, given this, there is no general moral requirement capable of preempting non-legal moral reasoning in the normal course of adjudication. It is a contingent matter whether the rule of law serves democratic values, or whether adjudication in fidelity to the law serves some morally worthwhile purpose. Judges, in order to act ethically, must be responsive to these contingencies and how it affects the (de facto) role they play in the state. They must consider what values the rule of law in fact serves in the circumstances on particular occasions of adjudication. Again, there is no automatic moral safety in fidelity for judges, so they must take account of non-legal ethical requirements in the course of adjudication.
Fourth, judicial fidelity to the law is not the proper starting point for an inquiry into judicial reasoning. Given that judges are responsible for ensuring the political legitimacy of their official decisions, we need an approach to adjudicative reasoning that displays the significance of such norms for judicial decision-making independent of a theory of law. Undoubtedly an understanding of the nature of law will shed some light on how judges should approach their office. Yet, if we accept that law can and does demand what it has no right to, a theory of legal reasoning cannot exhaust a normative theory of judicial reasoning. A central problem any theory of adjudication faces is how judges ought incorporate non-legal considerations of political legitimacy in the normal course of adjudication. I do have ideas about how we might address this problem, but that is clearly outside the scope of this article. For now, I will simply
186
ANTHONY R. REEVES
say that judges are irretrievably in the position where they have the responsibility to consider the value(s) that standing law serves, if any, in the circumstances and whether those goods (procedural, substantive, or otherwise) sufficiently justify the enforcement of that law. Judges are not simply umpires, impartially enforcing the fair rules of a voluntary and morally benign sporting event. Judges have political power in a legal system that may only be partially just or legitimate, or that may occasionally demand what it ought not demand. This significantly complicates the moral position of the judge, and we need an approach to adjudication that is cognizant of this complexity.
ACKNOWLEDGEMENTS
I am grateful to David Lyons, Hugh Baxter, Simon Keller, Jamie Kelly, Timothy Brownlee and an anonymous reviewer for this journal for their careful review of earlier versions of this essay.
REFERENCES
Applbaum, Arthur Isak, Ethics for Adversaries: The Morality of Roles in Public and Professional Life (Princeton: Princeton University Press, 1999).
Burton, Steven J, Judging in Good Faith (Cambridge: Cambridge University Press, 1992).
Campbell, Tom, The Legal Theory of Ethical Positivism (Brookfield, VT: Dartmouth Publishing, 1996).
Campbell, Tom, Prescriptive Legal Positivism: Law, Rights and Democracy (Portland: Cavendish Publishing, 2004).
Dworkin, Ronald, Law's Empire (Cambridge, MA: Harvard University Press, 1986).
Edmundson, William A 'Legitimate Authority Without Political Obligation', Law and Philosophy 17(1) (1998): 43-60.
Edmundson, William A, Three Anarchical Fallacies: An Essay on Political Authority (Cambridge: Cambridge University Press, 1998).
Eisenberg, Melvin A, The Nature of the Common Law (Cambridge, MA: Harvard University Press, 1988).
Gilbert, Margaret 'Reconsidering the "Actual Contract" Theory of Political Obligation', Ethics 109 (1999): 236-260.
Hardimon, Michael 'Role Obligations', The Journal of Philosophy 41(7) (1994): 333-363. Hart, HLA, The Concept of Law (2nd ed.). (Oxford: Oxford University Press, 1994). Horton, John, Political Obligation (Atlantic Highlands, NJ: Humanities, 1992). Luban, David, Lawyers and Justice: An Ethical Study (Princeton: Princeton University Press, 1988).
Lyons, David, Ethics and the Rule of Law (Cambridge: Cambridge University Press, 1984). Lyons, David., 'On Formal Justice,' in Moral Aspects of Legal Theory: Essays on Law, Justice, and Political Responsibility (Cambridge: Cambridge University Press, 1993a), pp. 13^0.
JUDICIAL OBLIGATION
187
Lyons, David., 'Derivability, Defensibility, and Judicial Decisions,' in Moral Aspects of Legal Theory: Essays on Law, Justice, and Political Responsibility (Cambridge: Cambridge University Press, 1993b).
Maclntyre, Alasdair, After Virtue (Notre Dame: University of Notre Dame Press, 1981).
Murphy, Mark C 'Natural Law Jurisprudence', Legal Theory 9 (2003): 241-267.
Oakley, JustinCocking, Dean, Virtue Ethics and Professional Roles (Cambridge: Cambridge University Press, 2001).
Perelman, Chaim., The Idea of Justice and the Problem of Argument. Translated by John Petrie
(New York: Humanities Press, 1963). Raz, Joseph, The Morality of Freedom (Oxford: Oxford University Press, 1986). Plato. Republic. Translated by C. D. C. Reeve (Indianapolis: Hackett Publishing Company.
2004).
Schauer, Frederick F, Playing by the Rules (Oxford: Oxford University Press, 1991). Simmons, A John, Moral Principles and Political Obligations (Princeton, NJ: Princeton
University Press, 1979). Simmons, A John 'Associative Political Obligations', Ethics 106(2) (1996): 247-273. Simmons, A John 'External Justifications and Institutional Roles', The Journal of Philosophy
93(1) (1996): 28-36.
Simmons, A John 'Justification and Legitimacy', Ethics 109(4) (1999): 739-771. Wellman, Christopher Heath 'Liberalism, Samaritanism, and Political Legitimacy', Philosophy and Public Affairs 25(3) (1996): 211-237.
Xavier University,
Cincinnati, OH, USA
E-mail: anthonyroyreeves@gmail.com
Present Address Department of Philosophy, Binghamton University, Box 6000, Binghamton, NY 13902-6000, USA