The Implications of Jurisprudential Theories for Judicial Election, Selection and Accountability
|
... To establish the implications, if any, of jurisprudential theories for the election, selection, and accountability of judges, we need to establish links between typical answers to questions of legal theory presented by various theories of law, and criteria for choosing judges and for holding them accountable. ... Our next step is to see whether theories of adjudication and theories of law may help us in dealing with the implications of this complex set of considerations for selection and accountability of judges. ...
... They also concede that there are cases in which the law is ambiguous or unclear or apparently silent, and in which the judicial decision is indeed an interpretation of the norm, but the language and the function permit more than one right answer. ... A moral realist might nonetheless accept that the law and adjudication must take cognizance of moral controversy in society, since ignoring it is likely to harm the legitimacy of the law and courts with those who do not share its moral stand. ...
[*1618] To establish the implications, if any, of jurisprudential theories for the election, selection, and accountability of judges, we need to establish links between typical answers to questions of legal theory presented by various theories of law, and criteria for choosing judges and for holding them accountable.
But when we read the literature on these two topics we find that legal theories never address explicitly the questions of choice and accountability of judges, and that discussions concerning these questions are not usually linked to any particular theory of law. n2 Thus, the temptation is great to say that there are no such implications.
This situation is not unique to the law. In many other instances, the theory of a profession does not yield pictures of who is to be regarded as a good practitioner and how such people should be selected. n3 Nonetheless, the situation -- in law as in other contexts -- is surprising in that it suggests the existence of a gap between attempts to analyze and to understand the nature of the activity on the theoretical level and important aspects of its practice. My analysis indeed reveals the limited scope of relevance of general theories of law to modes of selection and accountability of judges. Yet the practical gap is more radical than the pattern of relevance suggested by this analysis. The discussion of selection and accountability does not seem to be informed even by the limited insights which general theories of law can provide. I shall try to evaluate this [*1619] situation in my conclusions, after I map links and areas of relevance and their limits.
Both choosing judges and holding them accountable constitute decisions. We want the decision making procedures to be such that they yield the best results overall. The best results, however, may be a misleadingly simple description of a cluster of distinct desiderata. They may include the quality of particular judicial decisions, the quality of judicial proceedings, strengthening the judicial ethos and the legitimacy of the judiciary in society, and the furtherance of ideals such as the rule of law and judicial independence. Furthermore, each of these may in turn be categorized according to a variety of distinct factors which may be independent of each other and at times may be inconsistent.
Our criteria for choice and accountability will reflect our answers to some normative questions (e.g., what are "good" judicial decisions? How should judges conduct cases? In cases of conflicts between laws and morality, what should the judge prefer? Should judges be "activists"?) and our answers to a variety of empirical questions (e.g., how judges in fact reach their decisions, the likelihood that a past record will reflect future decisions, the extent to which the judge's own values and perceptions (rather than legal norms) affect their decisions). The relevance of jurisprudential theories concerning choice and accountability will depend on the extent to which they address and seek to answer these questions. Moreover, the theories will only be relevant if the answers they provide are specific enough to be helpful.
Judges are central functionaries in all legal systems, and all systems must provide some method for selecting them. Judges will be selected even if we have no theory to guide us in selecting them. Nonetheless, a deliberate decision on modes of judicial selection involves two distinct but interrelated tasks: first, an elaboration of what we seek, and second, the identification of procedures which may optimize the choice so that we are likely to get what we seek. Procedures may affect the substance in the sense that if available and acceptable procedures cannot yield a successful search for the ideal, we may have to settle for an effective search for less than the ideal. What we seek, in short, is the best possible judges and judicature.
Accountability is a more complex notion, and its relationships to selection may be varied and complex. I shall refer to accountability in a broad sense to include anything from specific responses of the system concerning individual judges, with the aim of disciplining them or removing them from office, through official responses directed generally [*1620] at the judicature and their working conditions, to critical responses by society or social groups which may affect, in a diffuse way, the status of the judicature, their decisions, and their conduct generally.
In one important sense, accountability is the response of the system after a selection is made and the selected person has violated a norm. Holding a person accountable may have at least two purposes: to "punish" the person involved, and to help in the creation or elaboration of the norm for the violation of which the judge is "punished." If the system is to be just and coherent, the norms involved must be those that are invoked to evaluate judicial decisions and judicial performance in general. n4 Thus, although accountability is past-regarding, some of the consequences of imposing liability and holding a judge accountable may be future-regarding. One future use of such norms may be to use them in the process of selection to choose judges so as to maximize the chance that they will be able to meet the requirements of the norms.
When judicial tenure is limited and a judge may be reelected, the processes of selection and accountability are united: reelection is a form of accountability. On the other hand, when tenure is unlimited, the only form of explicit accountability for one's conduct is through rarely used disciplinary measures, culminating in removal from office.
Two senses of accountability may be distinguished: accountability to, identifying those to whom the judge is answerable, and accountability for, identifying the norms governing the judge's conduct. Both senses of accountability have implications for selection. Thus, an argument may be made, although it is not conclusive, for placing the power of selection in the hands of those to whom the judge is deemed to be accountable. This argument, if successful, will decide the identity of the chooser. The criteria for choice will tend to be, as mentioned above, related to criteria for accountability. n5
[*1621] In many contexts, and in particular where the legal system confers on officials powers of decision, we want to enable the person -- appointed, chosen, or elected -- to be free to exercise those powers without constant supervision and review. Consequently, many of the guarantees against abuse of powers must lie with the procedure for choosing the judge. In such circumstances, initial selection and the articulation of criteria for selection are extremely important, since only major transgressions can be dealt with by holding the person accountable.
This is particularly true for positions that contain an inherent ethos of independence.
Judges pose a particular problem, because they are the authoritative determinants of legal (as distinct from political) accountability for all other power holders. Furthermore, because they must decide disputes between authorities and individuals, they must be independent of these authorities. Since accountability must be imposed by some organ, this raises the difficulty of finding a procedure for accountability which will neither negate judicial independence, nor seem to negate it. For accountability, too, the absence of a suitable procedure for the desirable scope of accountability may lead us to modify our goal.
Ideally, criteria for selection should reflect the norms for which a judge is accountable. Nonetheless, criteria for selection may include factors which are not part of personal accountability, while we may pass judgments on judges (accountability in the broad sense) without necessarily suggesting that something was wrong with criteria for selection or with the original selection of the criticized judge. Furthermore, our desire to promote independence may suggest that only some of the factors taken into account in selection should be allowed to feature in the [*1622] process of accountability. n7
Thus, both selection and accountability presuppose that we know what judging entails, and what constitutes performing well as a judge. But to decide on modes of selection and accountability, we need to know more, on both substance and procedures.
Theories of adjudication and theories of law may be expected to help us in elaborating the ideal of the good judge and the conditions required for the successful and stable performance of judicial functions. They may also have something to say on questions such as the legitimacy of judging and the identity of those to whom a judge should be accountable. There seems to be nothing unique which theories of law may contribute to our choice of modes of selection and ways to optimize social choice. Theories about the law cannot tell us much about other factors central to selection and accountability: how to predict a person's balance between open-mindedness and persuadability? What are the effects of education, social class, personal traits, on a judge's performance as a judge? Theories of adjudication try to answer some of these questions, but the proper domain of the quest is psychology, individual, and social, rather than legal theory.
Thus, what we need is not merely a mapping of where theory is relevant to the practical decision of selection and accountability of judges, but what is the possible contribution of theories about the law and about adjudication.
This analysis dictates my strategy: I shall first draw a pre-theoretical skeleton of the good judge and the good judicature. This pre-theoretical skeleton is complex and controversial, and on many important points it is incomplete. I shall then proceed to investigate whether theories of adjudication and theories of law may clarify these desiderata and their implications. Some of these theories may simply enhance our understanding of the phenomenon of judging. Others may provide us with alternative pictures of adjudicating which may have different implications to accountability and selection. I shall then explore the relevance of this "amended" picture to the issues of selection and accountability.
We seek the best judges to form the best judicature. "Best" here invokes criteria of evaluation. I believe these criteria are not system-dependent. Thus, [*1623] we can generalize this discussion to legal systems with similar conceptions of adjudication. However, I do not want to make the claim that this is a universally useful "ideal type," true for all legal systems at all times with all kinds of background cultures. The picture for modern Western systems is complex enough.
II. THE GOAL: A PRE-THEORETICAL SKETCH
Neither theories of law nor theories of adjudication seek to comprehensively describe adjudication or the character of a good judge. Nonetheless, what we need is a precise articulation of the criteria for the selection of judges. The picture I shall sketch will be mainly the result of a pre-theoretical observation of adjudication and the literature describing it. I am aware, of course, of the fact that this description of the good judge may be controversial. Yet I believe it does make a reasonable claim to being pre-theoretical in the sense that most observers of adjudication will agree on most of these requirements for a good judge.
A. THE GOOD JUDGE
The good judge is one who makes good judicial decisions. This apparently simple notion is in fact a catchword for a complex cluster of evaluation criteria, with tensions between some of the criteria which determine desirability. Some of the requirements are applicable to all judges qua judges, irrespective of the kind of regime within which they work or of the position they hold on the bench. Others are more essential for some judicial positions than for others. In a democracy we add another requirement to the evaluation of judicial decisions. Good decisions should correspond to the rules for division of powers and competencies which are typical of democracies.
Most of the cases, and thus most of the judicial work, involve trials; either criminal prosecutions or civil disputes. There are a few important skills which are required of the judges dealing with such trials. First and often overlooked, they must be good triers of fact. They must know how to conduct the trial efficiently but with respect and courtesy, so that the parties will have a sense that justice is being done. They must have the legal competence to identify the laws relevant to the facts and the basic ways of legal reasoning and interpretation. They must have a basic sense of justice and a measure of common sense which will check the application of these norms. They must act, and seem to be acting, impartially and without bias or prejudice to either party. They must also act, and seem to be acting, in obedience to the laws of the land. They must have [*1624] the ability to make decisions with caution and good judgment but without delay, so that legal justice will be as speedy as it can be, on the one hand, but that the parties will not feel that the judge came to the trial with a closed mind about the case, on the other hand. Finally, judges must know how to justify their decisions in a way that will be persuasive both to the parties and to the legal community.
This complex set of desiderata is the relatively unproblematic core of the judicial role, and it is clear that the skills required by it are many and distinct, and that while some of them are connected, others are independent of each other. Even this simple core-description poses formidable difficulties for a selection process which might seek to optimize all variables. Furthermore, this description is oversimplified, and hides further difficulties behind some vague terms. What is the good balance between "caution" and "speedy trial"? What is involved in exercising "legal competence"? What is a "persuasive" justification?
Many believe that the key to a good judicial appointment is "legal competence," and that all the other elements of the description are less important. But even if we accept this (and we should not), legal competence is not enough to guarantee a good resolution of legal cases.
First, observers stress that much of the work in the lower courts is not adjudication; rather, the work consists of a combination of mediation on the one hand and management on the other. n8 Moreover, problems resulting from overload and expense make it imperative to increase the efficiency of case load management. In addition, most judicial decisions in courts of first instance in the
Furthermore, disputes exist for which a legal approach may be disastrous. An obvious example is complex disputes between neighbors or in families. These are cases in which it is a mistake to encourage the [*1625] Kolhausian tendency of some to seek justice at all cost, or to solve the particular legal problem that is in court at the moment. What is important is not to mete justice, but to help the parties design an arrangement which will be least destructive and vindictive, especially when children are involved. The prime commitment of the decision maker should not be justice or the application of (legal) norms, but a viable resolution of the dispute. For this commitment, legal education is neither necessary nor clearly desirable.
Similarly, for cases in which intervention is needed to supervise a complex administrative agency structure, such as a company or a prison system, judges must have the skills to help them identify the right and the feasible arrangement, and design the supervision mechanism which might be most effective and least demanding of constant review. I know of no law school which seeks to provide its students with such skills, nor of a selection criterion which reflects them.
Another complication of the description of the judicial function is that the law may be silent or unclear. n10 In cases like this, judges may disagree, and the description of their activity as simply that of applying the norms to the facts is only partial. It is difficult in such situations to predict the judicial decision, since the ordinary language of the law does not dictate a unique legal answer. In all modern systems judges must decide such cases. Moreover, all modern legal system include norms and principles whose key terms are extremely vague, such as "reasonable" or "public interest," and whose application requires considerable skills of application which cannot be reduced to legal or logical competence. The question is what exactly judges are doing when they decide cases like this: Can we still say that they are simply applying preexisting law, or do they create new law? If the latter is true, how do they do it, by what authority and what norms govern and should govern this activity? Does it conform to the requirements of the principle of legality and the rule of law? Does it conform to the democratic principle of separation of powers? If courts perform a dual function of deciding disputes and articulating new laws, and there is a tension between the two, which function should be given primacy? The answers to these questions depend on some theoretical presentation of the nature of adjudication in such cases, and on the evaluation of such activity. The answers to these questions [*1626] also may reveal additional desiderata for the good judge. These questions are especially important when the decisions involve answers to value laden issues which are hotly debated in society. Still, these questions apply with equal force to more modest disputes in which the law is silent or unclear.
Another complexity of the judicial task stems from the possibility that some legal norms may appear, to the judge or to others, to be immoral. When an unclear norm may be interpreted in a moral way and in an immoral way, the judge should probably give it the moral interpretation. Even this modest claim requires that the judge has, in addition to legal competence, the capacity to identify immorality and make moral judgments. Moreover, it is only simple if there is a broad consensus regarding the morality or immorality of the interpretations. Once there is a controversy on those, the legitimacy of choosing by morals becomes questionable. However, there may be cases in which the legal norm is clear, but is seen by the judge and others as clearly immoral. What should the judge do? Some theories may define such situations away by saying that blatantly immoral "laws" are no laws. n11 But the theory needs to give the judge the tools to move from his own judgment concerning the morality of the "law" to the kind of judgment which permits him, under the theory, to disregard it as law. Besides, some other theories do not offer this easy way out.
In any event, when there is a controversy concerning the moral norms, a situation of conflicted judicial duty is likely to develop. On the one hand, judges must apply the law as it is, not as it should be according to the judge. On the other hand, after World War II, we prosecuted and convicted judges in Nazi Germany for doing just that. This fact implies that we do not always regard the judicial application of norms which are legal by reasonable interpretation an adequate (moral) discharge of judicial duty. n12
Under all theories, what the judge needs to meet this demanding test, which will happily be encountered only in rare cases, is both an acute sense of justice to see that the norm is blatantly unjust, and the moral integrity and courage not to apply it. Whether this is done by [*1627] distorting the law and applying a just norm or by accepting the norm as legal and refusing to apply it is a matter of tactics. n13
Another systemic requirement of the "good" judicial decision is that it will contribute to, or at least not detract from, the respect and legitimacy which the population feels towards the judicature. Courts and judges may lose respect and even legitimacy when they work within the core of adjudication: a judge who is systematically drunk or ineffective, who makes decisions arbitrarily or not at all, or who conducts judicial proceedings with arrogance and superiority, cannot expect to enjoy the respect of those who come before him. But it is easy to see why the question of respect and legitimacy arises most acutely around controversial decisions for which the law does not provide a clear answer, the kind of answer on which all competent lawyers are likely to agree.
Loss of legitimacy, even if it is unjustified, may weaken the court in the long run. Justified weakening of respect for the judiciary is a very serious blow to the institution. On the other hand, judges should not decide according to popular wishes. This is particularly true since in the most relevant cases, whatever the decision, some people will be angry. By taking these possible consequences into account, judges should seek to avoid justified weakening of respect to the court. They should seek to avoid justified weakening of their legitimacy by weighing the circumstances carefully, opting for self-restraint when this is not objectionable on moral grounds, or when no violation of a basic human right is involved, and writing the reasoning carefully and sensitively, in order to minimize the risks involved.
Again, what is needed here is more than mere legal competence. What is needed is political wisdom and sensitivity, a correct reading of social trends, and the capacity to make the non-legal value judgments involved.
To complicate matters even more, courts sometimes face two kinds of issues which do not raise problems of justice between the parties at all. These are problems of the political order itself, and the degree of protection it grants the basic human rights. These issues are likely to be emotionally loaded. They are likely to be controversial. Also, these issues are essentially political in nature. Yet the courts have a special role to [*1628] play in them, whether or not their country has an entrenched constitution with judicial review. Of course, this role is clearer and stronger when the judges work within such a constitution. The special role of the judges in such cases stems from the fact that majority vote, the regular decision making procedure for democracies, may have a systemic failure in some political areas. Similarly, the other power holders of the regime may be even less concerned about democracy and human rights than the legislature, so they cannot be relied upon to exercise control over the legislature.
The importance of this function is not primarily in its connections (if any) to dispute resolution. n14 What is at stake is the affirmation of the basic value commitments of the society, and the granting (or denial) of legitimacy to some decisions or actions. Naturally, decisions like this will be of high visibility, they will make the court both powerful and vulnerable, and they will be the obvious basis for risks to the legitimacy and respect given to the court.
What judges need for this task is moral and political wisdom and courage, the legal competence to identify the guidelines given by the legal material itself, and a sensitive reading of the social reality within which the judicial decision is sought. A judicial virtue which is crucial for this function is independence. Judicial independence is probably necessary for all judicial tasks, since its absence is likely to hurt impartiality. It is indispensable in this type of task since the pressures are likely to be strong and persistent.
So the conclusions of this pre-theoretical sketch are that judges need to have legal competence (which is itself a complex characteristic, involving many skills), commitment to the laws of the land, capacity at mediation, compassion, a sense of justice, knowledge of the beliefs and preferences in their society and the willingness to grant them proper weight in their decisions, the skills of a fact finder and an administrator, and the capacity to conduct a trial with courtesy and efficiency. The judge needs powers of moral and political reasoning, respect for others, moral courage and integrity, and the ability to strike balances between activism and restraint, daring and humility, vision and caution. A Herculean creature indeed. In addition, the good judge seemingly needs to [*1629] be slightly schizophrenic as well: a person with the required wisdom and integrity is unlikely to be self-restrained and humble; a person with this deep sense of responsibility is unlikely to bend uncritically to what he may see as absurd or unjust laws. The balancing is thus not only difficult and unclear. It may require inconsistent psychological tendencies. This fact has obvious implications to modes of selecting judges.
B. THE GOOD JUDICATURE AND THE
The above characteristics of a good judge must be held by each particular judge. For non-professionals, it is only the judge with whom they meet who matters. Therefore, each judge should be a close approximation to the ideal judge. Yet, each judge is a unique individual. Courts are social institutions. Thus, the question rises: are there features of the judicial institution which require additions to the previously discussed qualifications and traits which individual judges should possess?
The main question is whether courts should be "representative" of the population. n15 Especially when this population is heterogeneous, it is sometimes argued that a non-representative judiciary weakens the extent to which people feel respect towards the court and the extent to which they feel that members of the court are legitimate decision makers. This is true in part because the judges are seen as people who do not share the views and sentiments of the population at large, and partly because the judicature often alienates certain segments of the population which are not represented. n16 If we accept some requirement of representation, this should be reflected in our modes of selecting judges.
Our next step is to see whether theories of adjudication and theories of law may help us in dealing with the implications of this complex set of considerations for selection and accountability of judges.
III. THEORIES OF ADJUDICATION
By "theories of adjudication" I mean theoretical attempts to answer the questions of what judges and courts do in fact and what they should do. I also include attempts at articulating the presuppositions, logical, linguistic, and others, of judging.
[*1630] Theories of adjudication often have not included implications for selection and accountability of judges. n17 Moreover, such theories have not sought to articulate a full and comprehensive image of how judges do in fact act, or how they should act and the kinds of skills which adjudication requires. n18 Theories of adjudication do address, however, at least five questions which may have relevance for these issues. All these questions relate to the role of the judge in applying norms to legal disputes.
While all these questions are part of a theoretical attempt to understand and evaluate adjudication, it is difficult to differentiate between distinct theories of adjudication which form coherent wholes and which can be distinguished from other equally coherent theories of adjudication.
Theoretical insights into adjudication may be of interest in two different ways. Some insights will not be theory dependent, and will simply expand on the common sense pre-theoretical analysis given above. Such a theory may deepen our understanding of some features of judging and may improve our ability to decide on selection and accountability. Other theoretical statements may put forward one picture of adjudication and challenge alternative ones. In these cases, the implications for selection and accountability may depend on the theory one decides to adopt.
A. IS THERE JUDICIAL DISCRETION AND JUDICIAL LEGISLATION?
The core description of the judge applying preexisting norms and declaring the law applicable to the facts does not raise questions regarding the judge's personal values. So long as the judge is fully constrained by preexisting norms, his personality does not play a role in deciding the case (this is true in the interpretation stage; there is a room for personality in the fact finding stage).
I have already illustrated how this core description becomes problematical when the law is unclear or silent. The difficulty becomes even greater when the unclarity in the law relates to controversial questions or to issues relating to human rights and the political order itself.
The source of the difficulty is that if judges are not fully constrained by the norms, they must, in a sense, go beyond the law. If they do, the [*1631] question is how in fact this is accomplished, what values are invoked, and what authorizes judges to make such a choice? Furthermore, judges in such cases do not only decide the case for the parties. They develop the general law of the land. n19 In functional terms, a judge legislates through these cases. This description raises various problems: separation of powers, legitimacy, and competence. In addition, judicial legislation is retrospective, and there is an inherent tension between it and the principle of legality. n20
Thus, it is not surprising that a persistent thesis about adjudication denies that judges ever make law by going beyond the clear mandate of the language of the law, or the rules decided by precedent (which, logically speaking, are a proof of judicial legislation). This thesis has been given a variety of names: the declaratory theory of law, formalism, mechanical jurisprudence, or the gramophone theory of adjudication.
The most articulate contemporary advocate of this approach is Ronald Dworkin. Dworkin's thesis is that the law does indeed dictate a unique solution to all but the rarest cases, and that judges accordingly do not have what he terms "strong discretion." n21 Dworkin accepts, of course, the pre-theoretical data of ambiguous laws and judicial disagreement previously mentioned. Indeed, these data are important vehicles for his theorizing about the law. n22 Dworkin, nonetheless, argues that these disagreements are about the law, and that the judicial search is a search after the single right meaning of the law, when we add to the clear legal rules a background of institutional arrangement and political morality. n23
[*1632] I believe Dworkin's critics have the better of this argument. n24 But for the purpose of this paper, a weaker point suffices. Dworkin's judge clearly undertakes a substantial amount of moral and political reasoning to reach the unique legal solution which presents the law in its best possible way. Dworkin's definition of "law" may be broad enough to regard all this as a search within the law, but I think even Dworkin will concede that the kinds of arguments he presents for his judges to make are not usually encompassed within legal competence. A person may pass law school and the bar examination and be rated well qualified by his professional peers without ever being examined or evaluated on the kinds of arguments Dworkin has his judges make constantly. I believe Dworkin is right in arguing that at times judges make or should make these kinds of arguments, but I believe Dworkin is extending the accepted meaning of legal competence; this extension should be reflected in criteria for selection and accountability of judges.
Moreover, the answers Dworkin advocates for some complex constitutional issues are often controversial. Even if he is right that his answer is the correct one, those who disagree with him may resent the court which makes that decision, and may respond with an attack on the legitimacy of the court's activity. Thus, even if judges, in fact, do not have discretion, there is a part of their activity which is likely to raise controversy and to seem of questionable legitimacy. If one of the desiderata of judges and courts is to seek to minimize controversy and loss of judicial legitimacy, we need to know more of what judges do in such cases, whether it is described as creativity beyond the law or as a search within it.
In fact, possible implications to the selection of judges help in rejecting Dworkin's view on this issue. If Dworkin had been correct, judges' own values and politics should not make a difference. Dworkin's analysis of controversial cases is clearly that of a person with certain political commitments. Scholars with other commitments are likely to put forward a different constitutional analysis. Therefore, the commitments are relevant, after all.
On the other extreme are those who say that the norms never bind the judge and never dictate a unique answer. The realists, who were [*1633] among those who made this claim, stressed the ambiguity of key concepts in statutory laws, and the almost endless manipulability of precedent. n25 A contemporary scholar, Michael Moore, argues that norms never dictate answers because of semantic analysis. n26 According to these scholars, judges always exercise creative discretion, and all cases are hard cases.
Most theories of adjudication concede that there are cases, indeed most of the litigated cases, in which the law is clear enough to determine an answer for all practical purposes. They also concede that there are cases in which the law is ambiguous or unclear or apparently silent, n27 and in which the judicial decision is indeed an interpretation of the norm, but the language and the function permit more than one right answer. The decision is, in such cases, at the discretion of the judge. n28
It is boring to join the majority, but I believe they are right. The logical point of those who claim that all cases are hard may be correct, but it is a misleading claim from the practical point of view. It is true, logically, that whenever judges "apply" the ordinary language of a statute, or the accepted meaning of a precedent, they make a judgment that no moral or other reasons suggest that they should do otherwise. Nonetheless, saying that in such cases judges indeed applied the norm to cases is justified. These cases are easy ones.
B. THE EXTENT OF JUDICIAL DISCRETION
The extent of the need to go beyond the law in adjudication is relevant to the relative importance that may be given to this feature. If enquiring into the politics of judges has disadvantages, and the extent to [*1634] which such values may affect judicial performance is small and infrequent, we may opt for not allowing such enquiries at all. If, on the other hand, hard cases abound, and it is likely that judges' personal values will color many of their decisions, our willingness to do so may be reduced, and justifying it may become more difficult.
Intuitively, there is a difference in the answer to this question between different tribunals: judges in higher courts are probably asked more often to go beyond the law. This difference between tribunals may suggest that criteria for selection should be different according to the post the candidate is designed to fill. n29
The controversy in the literature, however, has been primarily logical rather than empirical or psychological. Most scholars believe that the extent of creativity depends on a number of factors. One of the more important factors is the manner in which norms are drafted. The more general and abstract the norm is, the greater the need for creativity in judicial application. Thus, the legislature has some measure of control over the extent to which judicial creativity is needed. In different contexts, these scholars will emphasize the predictability and certainty of judicial decision-making, or the uncertainty and the flexibility of adjudication. n30
As mentioned above, this seems to me the correct view for all practical purposes. The extent of judicial discretion will depend on contingent [*1635] factors within each system. Yet there are universal features of legal regulation which will guarantee that in all systems there will be some discretion. In all systems we use general terms, and their application may raise problems in the "penumbra." In all systems, power is granted in a general manner, and the extent of judicial review depends on the interpretation by courts. In all systems, questions will arise which the legislature had not addressed and which precedents have not decided either way. In most systems, the legislature has opted for some principles of general applicability (such as that of negligence or good faith) which require creative instantiations by the courts. And finally, in most legal systems, laws occasionally use non-legal categories, or refer to general moral values which do not enjoy the same consensus as more technical legal terms.
Another factor which should be remembered is that the highest courts have discretion in the sense of finality. Consequently, in many cases, the scope of judicial discretion exercised effectively might be quite different from that envisaged by the legislature. Judicial review in the
Furthermore, judges have some control over the amount of "judicial legislation," especially in public law, since the court can avoid deciding cases which the court thinks it is not wise or proper for it to decide. n31
Yet, even for those who accept that in many cases the law does dictate an answer, it seems that the cases in which it does not are not marginal or exceptional, and do not belong to one area of the law. Thus, the question is not the number of cases in which going beyond the law is required, but the extent to which deciding them may require going beyond the law. I am not familiar with a systematic attempt to provide an answer to this question. The myth of mechanical jurisprudence is misleading, however, for all courts and at all levels. There may be more discretion in supreme courts. Decisions of supreme courts may be more strongly binding, and have greater political effects than decisions of lower courts. But adjudication requires discretion as an immanent part of fulfilling the job. n32
[*1636] C. HOW JUDGES DECIDE (HARD) CASES
The way judges decide cases in fact may reveal that they invoke considerations which are not predetermined by the law and thus individualistic and creative. If these affect the decisions, we may wish to control them or examine these considerations as part of the selection process. We may want to hold judges accountable for the values they in fact invoke. Knowing how judges decide cases is crucial to predict their decisions, and to identify the factors which may be relevant for such prediction.
Descriptive decision theory is not very advanced in contexts other than legal. A judicial decision is a type of decision, and we have no more general knowledge about it than we do about other decisions. Some scholars draw the distinction between the context of discovery, the way judges reach their decisions, and the context of justification, the way they seek to justify it. n33 We have almost no direct data on the first, which is why so much is made of the justification. But even if for legal dogmatics the justification is all that matters, discovery is at least as important for selection of judges. Here theories offer a denial that judges reach their decisions by justifying them (although at times they may modify their intuition if they fail to justify it persuasively). n34 Positively, the process is described as one of intuition or of "reflective equilibrium" of sorts. n35 No model tells us how a judge's values and biases affect that judge's decisions. Rather, we are left with the common sense statement that they do. n36 The distinction between easy cases and hard cases is relevant because in the first group, we usually think that it is unproblematic to say that judges in fact do what they ought to do: apply the law. Even in such cases, experienced judges may reach the decision without explicit deliberation and consultation with "the law." Their legal competence [*1637] will probably make them "know" what the law is on the question without such deduction from the law. The possibility of a semi-deductive justification will present itself when judges write their opinions. Nonetheless, it is fair to say that in such cases, legal doctrine does in fact guarantee impartiality and neutrality, and enables predictability of judicial decisions. n37
The detailed models which seek to describe how judges in fact decide hard cases are put forward both as descriptions of what judges do in fact and as prescriptions for what they should do. One of the reasons for this happy coincidence is that the models are derived from judicial rhetoric, in the absence of much alternative data on which to base descriptive claims. n38 Moreover, these models seem to argue that the decision-making mechanism is the same for all judges. I am not certain this latter hypothesis is even presumptively attractive as a descriptive model, but no tools for a more differentiated analysis exist.
I shall follow John Bell in dubbing these models the consensus or social mores model, the rights model, and the model of interstitial legislation. n39 Bell, who does make a systematic effort to see what judges are in fact doing and to go beyond the way in which they describe what they are doing, concludes that the closest model to reality is that of the interstitial legislature. n40 He concedes, however, that all three models have problems as adequate descriptions. n41
The consensus model tells us that when a judge is faced with the need to go beyond the clear mandate of the law, he will decide the case by the positive morality or the social mores of his society. n42 The search is not for the judge's own standards. The standards invoked enjoy the acceptance of the community as a whole.
[*1638] The rights model tells us that the judge will complement the clear legal material with the rights which can be derived from institutional arrangements and background political morality. n43 Here, too, the search is for norms which are not the judge's own. At least according to Dworkin's analysis, the search is not only for positive norms, it is for norms within the "law" broadly speaking.
The interstitial legislature model is the one which has the courts invoke, within the systemic constraints, their own vision of the desirable solution. n44 In some systems, the laws themselves direct judges to act in this way. n45 In others, judges invoke the norms of political morality as they see them.
Some theorists argue that the two first models are not simply wrong, that they are illusory in the sense that judges cannot follow them, even if they try. These theorists argue that we cannot talk of one "social mores," since there are a number of divergent ones in most modern societies. When judges choose between competing mores, they will naturally choose their own, unless they are aware that their own beliefs are minority views. n46 Similarly, the argument is that complex constitutional issues which are supposedly decided by background rights are governed by norms which are so open textured, and by background arrangements that are themselves so open textured and often ambiguous, that the only way judges can decide, and are likely to decide, is by invoking their own moral and political intuitions and commitments. Furthermore, it is claimed that the rights model makes unrealistic presuppositions about the coherence of the political morality of the society in question. Judges invoking the rights model are bound to reconstruct the political morality of their society rather than accept it as given. This reconstruction will be based on the reconstructor's own values. n47
[*1639] There are apparent implications to the choice of a model for selection of judges. n48 The first test (social mores) requires the ability to search for sociological data, namely the positive morality of the relevant society. Of course this does not mean that judges should be sociologists, or that they should conduct opinion polls. It does require that judges be knowledgeable about the moral climate of society, aware of the plurality of approaches to certain issues and of the limits and non-representativeness of an individual's approach, and care enough to be willing to concede that their opinions may be elitist positions unshared by the majority of the population. If this is the way judicial decisions are made, there are implications both to the requirement of representativeness of the court and to special considerations in allotting judges to cases. There are implications to the relevance of evidence, and to the nature of the proceedings. Usually, judicial notice might not be enough. It is important that an explicit attempt to identify social mores is made. This is of particular importance since one of the advantages of this model is its sensitivity to the legitimacy issue. If, under the name of social mores, judges merely apply their own sectarian values, this may contribute to a feeling of alienation and loss of legitimacy.
The second test (rights) requires familiarity with political arrangements other than legal norms, and with the basic justifications for legal norms and institutions. It also requires a familiarity with the history of social and legal institutions and with the current versions of political morality.
The third model requires that judges be persons of valid and wise moral judgment and integrity who will know to identify the cases that should be decided within the law and those cases in which it is justified to legislate.
Needless to say, there may be controversies concerning the content of social mores or political morality and concerning the status and validity of moral judgments. These controversies make the seemingly simple implications of the models quite intricate. I shall have more to say, below, on the relevance of ethical and meta-ethical controversies, in the context of the third model, to these issues.
I tend to accept the superiority of the third model as a description of what judges can and will do. The practical differences between the models may be diminished by the fact that judges' moral judgments are likely [*1640] to be those of their society, and these moral intuitions are likely to correspond well with the institutional arrangement of the society in which the judge serves. n49 But it is clear that these coincidences are likely to be more difficult to maintain on issues which are controversial, or on issues which are strongly affected by status, education, richness, or social background. Judges usually do not belong to the majority of their communities in many respects, and in some areas differences between their views and social mores or even background rights may be likely.
D. HOW JUDGES SHOULD DECIDE (HARD) CASES
As I said, for most scholars there is a happy coincidence between the way in which they think judges in fact decide hard cases and the way in which they should do so. n50