42 Md. L. Rev. 766 Maryland Law Review 1983 Symposium: The Appellate Judiciary--Its Strengths, Its Woes, and Some Suggestions for Reform *766 THE PROBLEM WITH THE COURTS: BLACK-ROBED BUREAUCRACY, OR COLLEGIALITY UNDER CHALLENGE? Patricia M. Wald [FNa] Copyright 1983 by the Maryland Law Review, Inc.; Patricia M. Wald Much has been written in the past several years about the overload in the federal courts and the "judicial bureaucracy" that has arisen to cope with it. [FN1] The fear is that, like so many over-diagnosed social problems, the public will simply grow weary and accept the ailment as chronic and incurable. The public debate has produced many criticisms but few workable solutions. There are accusations and denials that the courts take too long to issue decisions; that opinions are turgid and murky; that judges write too many concurrences and dissents, leaving no clear idea of what the law is; that judges are delegating too much responsibility to callow clerks. [FN2] Proposals have suggested lopping off chunks of federal jurisdiction, increasing reliance on out-of-court dispute resolution techniques, appointing specialized technical advisers or establishing specialized courts for particularly complex cases. [FN3] One suggestion I find especially unreal is that, like some genetic marvel, we create a new breed of Super-judges who do every speck of their work themselves. (Even Batman had Robin to help out in tight spots.) *767 As with so many other crises, the rhetoric of the "crisis in the courts" may obscure the basic issue: How can the courts, like the rest of society, do their work better in the face of increasing demands on their time and less than adequate resources at their command? A facile labelling of the courts as "bureaucratic" contributes little to the debate. Clarifying the semantics of this debate is important lest we waste our efforts attacking the wrong aspects of court operations because they are associated with the dreaded, but possibly contrived, spectre of bureaucratic justice. I have a difficult time understanding how our courts--warts and all--can be labelled "bureaucratic." Unless we use bureaucracy as a shorthand and pejorative term to cover any large, complex institution with its unavoidable procedures and inefficiences, courts simply do not look or act like typical bureaucratic organizations. Our federal courts hardly operate in the hierarchical fashion of a bureaucracy. Federal district judges have equal and independent decisionmaking authority. Needless to say, we have no power to discipline, demote, or dislodge recalcitrant district judges. The intermediate circuit courts can reverse or remand decisions, but that power is subject to powerful constraints such as "abuse of discretion" and "harmless" and "plain error" doctrines. Last year in our circuit, we reversed only about 18% of appealed district court decisions. That percentage is about the same as for earlier years; it has not grown with the increase in our own numbers. [FN4] Although we may wince at the occasional reversals of our own decisions by the Supreme Court, the Justices last year found time to review only 3.2% of our circuit's decisions. [FN5] Thus I believe it is fair to say that vertically--district to circuit, circuit to Supreme Court--the federal judiciary does not function in the hierarchical fashion of a typical bureaucracy. Within multimember appellate courts as well, there is certainly no hierarchical order of command. To the degree any leadership is exercised, it derives from an individual judge's own abilities and the respect or affection of colleagues. To the extent judges act in conformity, it probably stems from professional collegiality, common professional *768 training, and a shared outlook on their place in the constitutional order. Other key factors of the classic bureaucracy make the label inappropriate for the courts. We have, if anything, too few standard operating procedures or organized routines designed to ensure that we perform technical or repetitive tasks quickly, efficiently, and consistently. There is practically no division of labor on specialized tasks. We have few aggrandizing employees seeking to expand spheres of control or influence. Our "elbow clerks" serve for only a year and are subject to intensive scrutiny. We rely on our informal networks to learn of court operations to a far greater degree than on any formal communications system. So whatever courts are, or are coming to be, they do not look like bureaucracies. Nor do I see them heading in that direction. But even if they do, does it matter if courts do take on some of the features of bureaucracies so long as they continue to achieve fundamental objectives? Our real task is to identify those characteristics of our modern court system that interfere with the achievement of our objectives. It is necessary, then, to identify exactly what we expect the courts to do. At the risk of drawing a charge of sounding bureaucratic, I list here my own "objectives" for a court of appeals. Further, I discuss some of the changed circumstances that inhibit achievement of our objectives, and describe some of the institutional limitations that we must recognize in any search for improvement. Finally, I note my disagreement with several key proposals for dealing with these problems, and suggest some institutional corrections that could prove salutary. I. OBJECTIVES FOR THE FEDERAL JUDICIARY First, we want courts to make correct decisions on the myriad cases and motions they face. Basically, decisions should accurately reflect the facts in the record and existing law on the subject. Ideally, we also should aim to season the logic of our decisions with an understanding of real-world constraints on litigants (who are often government agencies in our court), the public, and the judiciary. Second, the courts' opinions should contain reasoned explanations of their decisions to lend them legitimacy, permit public evaluation, and impose a discipline on judges. Third, courts should produce timely decisions and opinions, meaning, quite candidly, that we should hold our feet to the fire. Fourth, courts should strive for uniform decisions, especially, as in *769 our circuit, when one tribunal is composed of a number of separate panels. Fifth, the courts must bear in mind that, as the only unelected branch of our Constitutional triad, they must act always to preserve and to reinforce public confidence in their integrity. Historically, achievement of this objective has required judges to walk a fine and precarious line: to render decisions based on the facts and the law, resisting personal bias toward individuals or groups, while preserving the values of the judge's own personal reasoning, experience, and ultimately, sense of responsibility. I believe fear of failing to achieve this last objective--maintaining a personalized judiciary that applies impersonalized rules--is at the heart of the alarms that have been sounded about judicial "bureaucracy." The growth in number of cases, judges, and staff has given rise to institutional techniques for decisionmaking that depersonalize the process and result in final decisions that may be less--or more--than the judge's own. Our task is to shape the institution and judges' roles within it to retain the special quality of personalized decisionmaking that is the historic strength of our judiciary. II. CHANGED CIRCUMSTANCES Two major changes in the courts' work, occurring gradually over time, may threaten the continued attainment of basic objectives. One is the quantity of work, which has increased enormously over the years. The second is the complexity of the underlying subject matter, whether it be scientific, economic, medical, or some other field of technical expertise. On a court like the District of Columbia Circuit, whose bread and butter work is the review of decisions of specialized government agencies, a generalist judge's struggle to understand these matters, and then to figure out how the fine or complex distinctions among them affect the law, is endless. A. Number of Cases The greatest burden of the growing caseload is probably on the federal district courts, which face a flood of prisoner complaints, habeas corpus petitions, social security appeals, and diversity cases. [FN6] *770 But in recent years, most federal appellate courts have seen their caseload spiral upward as well. In the year ending June 30, 1982, there was a 6.0% rise in federal appeals filed over the previous year; three circuits rose by over 15%. The D.C. Circuit and the Second Circuit were the only ones that saw declines in docketed appeals--by 9.8% and 7.6%, respectively. The D.C. Circuit's decrease in filings resulted primarily from a dramatic decrease in appeals from agency action in the first year of the Reagan Administration, and the Second Circuit saw a significant decline in bankruptcy appeals. [FN7] This year, however, our docket has begun to climb again, resuming the growth that took us from 1,168 appeals filed in statistical year 1972 to 1,611 in statistical year 1981. [FN8] A federal appellate court of course has no control over the total number of docketed cases; every rejected litigant in the district court has the right to appeal a final order once. Even plaintiffs who are denied the right to appeal in forma pauperis can appeal that denial itself, and so, by the back door, obtain a threshold determination on whether their claims have merit. B. Complexity of Subject Matter Another important change in at least the D.C. Circuit's caseload is the increasingly technical, complex nature of the subject matter. Most of these cases involve appeals from the adjudications and rulemakings of federal administrative agencies. The D.C. Circuit hears almost a fifth of the nation's direct agency appeals, [FN9] and they comprise between a third and a half of our workload. [FN10] Many of these time-consuming cases involve records with thousands of pages, multiple issues, and numerous *771 parties. On the average, agency cases take almost a year from completion of the filing of the record to final disposition--two months longer than our average for all cases. [FN11] III. PROPOSED RESPONSES TO CHANGED CIRCUMSTANCES To reduce the stress placed on the judiciary by the increasing number and complexity of cases, several "solutions" have been proposed: limitations on federal jurisdiction, restricting the scope of judicial review, and increased reliance on out-of-court resolution of controversies. [FN12] Although some of them may contain some paydirt so far as decreasing federal caseloads, they also have drawbacks. Most important, however, I do not think they hold any immediate potential for a major cure. A. Narrowing Federal Jurisdiction The requirement that two of three independent appellate judges reviewing the denial of an asserted right by a single district judge must agree that it was a reasonable decision gives substance to our boast that we have a system of laws and not of men. The reality that our court reverses the district court in about 18% of our cases [FN13] is a tribute to the quality of our district judges and not, in my view (and I believe that of most of my colleagues), an argument for restricting the right to appeal in any way. We see just enough cases where a mistake has been made at the trial court level, and a genuine injustice done, to reject any proposals to limit or forfeit any part of our appellate jurisdiction. Some proposals for narrowing federal jurisdiction would undoubtedly affect the number of appeals, but in my view, at an unacceptable cost to our jurisprudence. The swelling influx of cases in large part reflects more legal protections, benefits, and access for those groups that previously only encountered the law as a weapon aimed against them. I am certainly not eager to have the courts turn their backs on these people in order to clear their dockets or to provide time for drafting stylistically more elegant opinions. *772 In actuality, the only big-number candidates for removal from federal jurisdiction are the controversial diversity jurisdiction cases (about 25% of all federal district court civil cases) and state prisoner petitions (about 12%). [FN14] Although their elimination could affect other courts, neither plays any substantial role in the D.C. Circuit's appellate jurisdiction. If we think we can or must contain the flow of cases, we will have to build our levee elsewhere. B. Restricting the Scope of Judicial Review Congress has tended in recent years to commit appeals from agency actions directly to the courts of appeals rather than to follow the older APA model of district court review. There are few who oppose this. On the other hand, a lively debate has ensued about the worth of any appellate review of the substance of complex agency decisions and the courts' institutional competence to perform it. [FN15] Although the Supreme Court has not yet held specifically that review of agency action should be less intrusive than it has been, the tenor of the Court's decisions has sounded forewarning notes. [FN16] Indeed, the D.C. Circuit may have picked up the strains, for it overruled federal agencies in only 8.5% of the appeals in statistical year 1982, as compared to 13.5% in statistical year 1981 and 12.8% in statistical year 1980. [FN17] The winds of change on this issue, however, are not all blowing in the same direction. At least one house of Congress, through adoption of a form of the "Bumpers Amendment," sought to sanction a more searching judicial review of agency actions or rules on the ground they are not "authorized" by the enabling statute. [FN18] *773 When all is said and done, then, I doubt very much that Congress will abolish or even substantially restrict judicial review of administrative agency actions. And even if our scope of review were narrowed by Supreme Court interpretation of existing law, it would be unlikely to result in any real appellate economies. Judicial review would likely still be invoked as frequently and consume as much court effort regardless of the formula used. Thus, narrowing the scope of judicial review in my view would not appreciably lessen the number of cases we hear, nor eliminate the most complex ones from our dockets; rather, it might precipitate a more serious problem--a court system that functions merely as an adjunct to the federal bureaucracy. Those who decry the creeping bureaucratization of the judiciary through the addition of more staff or clerks may be missing the main event. Right now, the closest contact federal appellate courts have with bureaucratic decisionmaking is in their review of administrative agency orders and rules. Our limited role, the Supreme Court has told us, is to determine whether agency procedures have produced "substantial evidence" for agency findings and whether the decision is rational, i.e., not arbitrary and capricious. [FN19] We have little or no authority to interpret or to apply the law according to our own reading or research; we must approve the decisions made by administrative law judges or lower echelon agency employees--decisions reviewed by politically appointed boards, commissions, or administrators--unless their acts are palpably irrational. Yet these agency decisions very often reflect some of the most disturbing aspects of bureaucratic decisionmaking: inflexible adherence to rules, insensitivity to the equities of individual cases, extra-agency *774 considerations, routine rubberstamping, and imperviousness to realworld consequences. [FN20] Federal appellate judges, nonetheless, are supposed to review and in the overwhelming majority of cases simply to affirm thousands of such decisions each year, without becoming infected. In sum, although the focus of the bureaucratization debate has been on how we do our work, our greatest contribution to bureaucracy may be in what real-life circumstances and narrow standards of review force us to do--to affirm the great majority of decisions made by real bureaucracies. In this regard, we undoubtedly run a risk of functioning basically as an adjunct to the very bureaucracy whose decisionmaking practices we are urged to avoid. Thus, in my view, we should be wary of efforts to limit further our powers of review lest, as watchdogs of the public interest, we become totally toothless. In addition, many judges feel that even though standards for review currently may be strict, the overhanging discipline of judicial review and the extraordinary inconvenience that reversal imposes on an agency have a salutary effect on agencies' compliance with statutory mandates and due process norms. As a veteran Washington-watcher as well as a judge, I am sure of it. C. Out-of-Court Dispute Resolution One suggested approach to reducing caseloads--out-of-court resolution of federal controversies by mediation, conciliation, and settlements--is receiving much attention currently. [FN21] Because the federal government is the plaintiff or defendant in the vast majority of federal cases, significant activity in this area would require a major commitment on the federal government's part. There may indeed be increasingly good prospects for successful preliminary negotiations on at least some troublesome rulemaking issues. [FN22] Even the most enthusiastic proponents of negotiation, however, recognize that it often requires a combination of favorable circumstances *775 to succeed. [FN23] Including agencies as full-fledged participants in the negotiating process may not always be possible or even desirable: the agency's flexibility may be limited by statutory requirements; it may not be feasible to fashion a common position from which to begin negotiations because of intra- or inter-agency squabbles; and there is always the risk of negative publicity about "sweetheart" deals and lax enforcement. Although negotiation--and other promising efforts atarbitration [FN24]--do hold out some long run hope for reduced caseloads, the message still seems clear enough: Neither proposals to cut off federal jurisdiction, narrowing the scope of review, nor out-of-court resolutions offer significant solace in the here and now. Political developments, however, such as our court experienced in 1981-82, [FN25] are capable of making large inroads, at least temporarily. Decreasing government activity, whether in the form of less regulation or less enforcement, quickly will be felt in less federal litigation. But the public's acceptance of this strategy is difficult to predict and its employment is erratic over time. The judiciary certainly is not in a position either to influence it or to use it for planning purposes. IV. JUDICIAL RESPONSES TO CHANGED CIRCUMSTANCES Whatever our yearning for the simpler life, heavier caseloads and increasingly complex subject matter are surely here to stay. Reaction to this fact already has produced circumstances within the courts that may impede their ability to achieve the objectives we have cited. The most notable reaction has been the appointment of more judges--there were about 650 active federal court of appeals and district court judges in 1982 as compared to about 480 in 1975. [FN26] Simply because there is no real hierarchy to order relationships among judges, *776 the increased number of judges already may have strained the traditional collegial decision-making process of the courts. The courts have also expanded their staffs significantly, although they are still small relative to other institutions. [FN27] The increase in judges and staff may exacerbate latent problems in the judicial environment. First, most judges have no training and little interest in organizing and administering a formal decision-making regime (or being subject to one). But with more cases, more complex subject matter, more judges, and more staff, the courts need management techniques to assure relatively accurate, uniform, and timely decisionmaking. [FN28] Despite this need, innovations may be slow in coming for a number of reasons. First, the seniority of many judges and their treasured independence may make them resistant to the institutional changes necessary to cope with new circumstances. Yet the maturity and wisdom gained from long service and the fairness achieved because of judicial independence contribute to the integrity of our judicial system and the respect that it enjoys. Second, the fiercely protected autonomy of judges exacts an additional cost: Once we move beyond the chamber of each judge, the process of decisionmaking is largely ad hoc. We have no consistently enforced system for deciding when to treat cases expansively or summarily, how to deal with related issues in different cases, or even how to reconcile differing treatment of related cases. Every case cannot and should not receive equal time for review, decision, and explanation. But we should have consistent, sensible, and workable rules for managing an infinitely variable caseload. At present, *777 we often allocate our prime resource (judges' time) according to quixotic combinations of tradition, habit, and personal preference. If it is inevitable that a court, like the D.C. Circuit, will face evermore complex cases, and we do not wish to sanction all bureaucratic agency actions with perfunctory review, we must somehow distill the essence of the judicial role--preserve it--and then adjust the institution of the courts to permit judges to use it better. In large part this assignment becomes how to allocate our prime judicial resource--judge's time--to achieve our key objectives. A. In-Chambers Allocation of Time: Use of Staff Most of us imagining appellate judges at work think of them performing their in-chambers tasks: reading briefs, distilling the record, analyzing legal arguments, deciding, and writing opinions. At the heart of judicial accountability is the notion that the judges' decisions, the ultimate exercise of the power and discretion conferred on them by the Constitution, are in fact their own decisions, supported by their own reasoning. Because some commentators now assert this work has been usurped by our clerks, [FN29] it provides a good place to start an analysis of how a judge should allocate his or her time. The harsh reality is that federal appellate judges typically sit on 130- 150 cases each term, and as many motions; participate in six to ten en banc cases; author up to forty full-fledged opinions and another twenty to thirty memorandum opinions; consider 100 petitions for rehearing; serve on several Judicial Council Committees; [FN30] and all the while strive to keep abreast with the world at large. Were a judge to research and to write every word in every opinion, it would mean that he or she could write only a dozen or so opinions a term; the rest would be consigned to a boilerplate order or a two-paragraph per curiam. With our workload, uncomfortable choices have to be made--choices that were not required in the primeval days when I was a solo circuit law clerk. Every judge has to decide what core function she alone must perform. Personally, I feel that so long as the judge reads all the briefs (and key portions of the record), listens to the oral argument and engages in face-to-face dialogue with the counsel, participates in the conferences with her colleagues, and controls the formulation and expression of the ideas and analysis that go into her opinion, she has *778 done the job. Whether the judge writes the first draft for the clerks to critique and to flesh out, or the clerk writes the first draft for the judge to revise and to challenge is not dispositive of whether the judge is still in charge. We all know, of course, about extremes where judges give law clerks too much discretion and too little supervision in fashioning their decisions. The check on those instances usually lies in the rapid awareness of colleagues that a particular judge's drafts need special scrutiny to assure they contain no unexpected twists of reasoning or shifts of logic. The court's distinctive role in our constitutional framework requires that a judge's opinions reflect the judge's thinking, familiarity with the facts and arguments, legal analysis, and literary style. On the other hand, the decisions we have to make as federal judges are complicated, and involve concepts and raw material with which we have no innate familiarity. Counsel spur the dialectic process in the courtroom, but a judge needs to keep the momentum going back in chambers. Good judges cannot go off in the desert like St. Francis and meditate until the right decision descends upon them. Judges do not have the luxury of discussing their cases with anyone but law clerks and other judges, and other judges have time-sharing problems of their own. Very simply, the most seasoned opinions benefit from a continuing dialogue between judge and law clerk. The vast majority of judges operate within the range of reasonableness between drafting their own opinions, and scrupulously reviewing and editing the drafts of their clerks. The latter mode of course is a familiar one for senior partners, high government officials, and professors. Why should it be out-of-bounds for judges? The role of law clerk to a federal judge, however, is unique. The clerk usually serves only one year, but in an intensely intimate relationship. The balance between the boundless energy and imagination of the clerks and the intuitions, experience, and perspective of the judge has to be worked out anew each year, all year, if the relationship is to stay in control. Ideally, the recycling of new law clerks each year means a continuing legal education course for the judge as well as the clerk. But vigilance over the final work product is a price that the judge cannot discount. There is a practice in some courts of relying on a central staff of lawyers to draft orders and memorandum opinions for the judges to accept, to revise, or to reject on a variety of routine cases that may be decided on the papers without oral argument. The process is reductive, but I fear this "completed staff work" mode needs careful watching lest *779 it result in too ready approval by overworked judges of the work of court staff who are less likely than the elbow clerks to reflect the same intense familiarity with the judges' style and thinking. Terse orders and memoranda may be the only possible judicial response to a large number of cases in an overcrowded system, but we must be careful that our decisions do not deteriorate into standardized forms. If a choice must be made, I personally would prefer expanding the judge's in-chambers staff, because it is far more accountable to the judge, or relying more on brief memorandum opinions so long as they are still authored by judges. Individuality and care can shine through even in a few paragraphs. B. Scheduling Help Nevertheless, central court staff can play an important support role by scheduling cases to assure both efficient presentation and the matching of interdependent cases. When a court "hears" some 1,000 cases a year, employing eleven active judges, three senior judges sitting parttime, and a shifting corps of visiting judges, effective case management can make an enormous difference. The scheduling job actually begins far in advance of the actual packaging of cases and panels. So long as there is no sizeable backlog of cases ready for argument, the pending cases in the pipeline must be monitored to insure that a sufficient pool exists for each session. Thus extensions of time for filing of briefs and records or transcripts cannot be given routinely without calculating whether they will upset the measured flow of cases. Scheduling also should shape more efficiently the oral and written presentation. Our court's staff counsel meets with the lawyers in complex cases to plan the presentation of the case; to avoid duplication of briefs and oral argument; to identify threshold issues, such as jurisdiction and finality, which could mean early disposition of the cases; and to consolidate appeals with similar issues. The staff counsel also skims all briefs in order to "weight" a case so that each day's package of cases is roughly equivalent; judges should not be overwhelmed one day and underchallenged the next. She identifies cases that seem on the surface (sometimes, deceivingly so) to be so clear as not to merit oral argument, and recommends their disposition on the papers. Any judge on the assigned panel can, and often does, ask for oral argument. And, when we build up a backlog of middle-range cases--not overly complex but serious enough to require oral argument--she schedules supplementary panels. Most such cases eventually can be disposed of with a short memorandum opinion. *780 An efficient scheduling team must work closely with the chief judge of the court, and the bar must know that the chief judge backs up the staff counsel's recommendations for allocating counsel time on argument or elimination of duplicate briefs. At all stages, the chief judge needs to place priority on moving the cases along. Consistent movement is important not only because it assures timely decisions, but also because, as time for the hearing approaches, the pressure for out-of-court settlement grows. Litigants who may have counted on delay have to come to grips with the merits of an appeal precipitously or improvidently filed. C. Dividing Time Between Motions and Merits Work Very often the manner in which a case is decided in an appellate court depends on the label that attaches to the cover sheet of the docket. For instance, the D.C. Circuit has separate procedures for deciding motions and merits cases; [FN31] whether these differences reflect accurately the attention we should allocate to each is sometimes problematic. Many of the twenty weekly motions, especially those arising from executive actions of federal agencies, may be just as complex and have as far-reaching implications as our twenty monthly merits cases. Judge Leventhal once described the range of motions matters before the court in his day. They included "arguments on stay of publication of the Pentagon Papers by The Washington Post; stay of the Amchitka detonation; and on stay of the encampment on the Mall of the Vietnam Veterans against the War." [FN32] The list gives a sense of deja vu. In a recent two-month period, we had an eve-of-trial appeal from the denial of prosecutorial immunity in a section 1983 action by the McSurelys, the Kentucky civil rights workers; [FN33] an emergency appeal from a judgment requiring the Interior Department to act on 17,000 Indian claims before the end of the year; [FN34] a petition to bar interim directors for the *781 Legal Services Corporation; [FN35] a petition to allow the videotaping of an FDA hearing on a long-term contraceptive drug; [FN36] and--adaptation on Judge Leventhal's account--a stay of a lower court ruling denying demonstrators in Lafayette Park the right to sleep overnight to symbolize the plight of the homeless. [FN37] Yet judges too often regard motions practice as an "add-on" to regular merits-panel assignments. We have decided, without actually considering why, to allocate to motions less of that precious resource, judges' time. Oral argument is rare; even rarer is a full-scale opinion written by an individual judge; most motions are decided by order or by a short per curiam opinion. Motions work is sometimes regarded as an annoying diversion of judges' priority time; the motions assignment is not offset by a reduction of merits assignments, creating a strong incentive to minimize involvement. I have no pat solution, but a few suggestions. We might take judges off merits work, or reduce their merits assignments while on motions. This shift would concentrate judges' focus on the motions work and encourage extra efforts on those motions deserving special attention. We also might move the motions clerks in-chambers while judges serve on motions, to approximate more nearly the symbiotic relation-ship of judge and clerk. At a minimum, the division of time among cases should not depend on the label they bear or the track they come in on. The tight compartmentalization of motions and merits work too often tilts toward that result. D. Summary or Extended Treatment Within either category of cases--merits or motions--judges must make discriminating choices about how to allocate their time. A key element in that decision is the type of opinion that accompanies a decision. Although every litigant would like an ultimate exposition of reasons for a decision, the truth is that judges should allocate more effort to the opinions in hard cases in which courts must "develop and declare the law." We sometimes seem to do so, however, only coincidentally, and whether the present "system" for identifying those special cases makes sense is questionable. Many courts, including ours, currently decide around half of their *782 cases by boilerplate order or short per curiam memorandum opinions. [FN38] There is no uniformly practiced rule or guideline as to which cases will be accorded what treatment. The decisional mode often depends on particular judges' feelings about the cases, their views as to the significance of specific issues, and their individual beliefs on the importance of giving every litigant their fully developed reasons for the result. My own guiding principle is that virtually every appellate decision requires some statement of reasons. The discipline of writing even a few sentences or paragraphs explaining the basis for the judgment insures a level of thought and scrutiny by the court that a bare signal of affirmance, dismissal, or reversal does not. A minimum opinion need not be unduly time consuming to write. Abridged memorandum decisions can be addressed primarily to the parties, not to the public. They can identify the issue for decision, the court's disposition, and the principle basis for the ruling. They can normally forego an exposition of the facts and procedural history. [FN39] By their very nature, they may have less precedential significance than more extensively reasoned opinions. [FN40] This memorandum treatment, however, should be reserved for cases in which the issues involve the application of existing rules of law to a standard fact situation, and in which there is no reason for qualifying the settled rules. The memorandum approach should not be used when the court develops or modifies a rule of law, resolves a conflict or apparent conflict between panels or its subordinate courts, faces an issue of special public interest, or is not unanimous in disposition of the matter. We have no consistent enforcement of rules or criteria on when to issue per curiam opinions. Sometimes we do so because a case is thought to be relatively simple. Sometimes it is because a case is so complex that the judges split the drafting task among them and the resulting patchwork is called a per curiam opinion. A court also may use a per curiam to share the responsibility for a controversial decision--or to add the weight of three judges to the words of one. In motions practice, it seems to be traditional that all decisions-- regardless *783 of length or complexity--are per curiam because they always have been. Some of these practices need to be reevaluated. The courts' accountability and the public's respect for the institution are in most instances furthered by a signed published opinion, but there is a place for short per curiam opinions proclaiming the court's decisions in cases that are so complex or controversial that they provoke a series of separate opinions, or are so cut-and-dried that they genuinely do not need more than a few paragraphs explanation. I believe we should try to hold the line there. [FN41] *784 E. En Banc Hearings The development of a more uniform practice for rehearings en banc may also be necessary in response to a rising caseload. Such rehearings serve important purposes: rectifying an error made in a panel opinion--so that it will not confound us in the future or have to be overturned by the Supreme Court--or, conversely, placing the court's full weight behind a controversial panel ruling. Yet we do not have the capacity to rehear all or even a fraction of the cases that some or even most of the judges think were wrongly decided. Some judges vote routinely for rehearings en banc on all cases with which they disagree; others hardly ever vote for rehearing because of the untidiness of the en banc format and its inevitable production of multiple opinions. We receive over 100 petitions for rehearing en banc a year; less than a dozen typically are granted. [FN42] In some circuits, the figure is much lower. It sometimes seems as though we avoid the most disturbing and far-reaching cases and focus on the small, albeit annoying, mistakes. If so, it is a waste of the court's time and energies. We need to develop more specific and enforced guidelines that press judges to weigh the clearness of the error and the importance of the case so that en bancs are limited to those cases that truly have an impact on the circuit's jurisprudence. F. Ordered Collegiality Evaluating the use of judges' time is impossible without considering how they interact as a court. A court is a collegial system--unwieldy, hard to direct, impossible to command. More than most systems, our collegial process takes a disproportionate amount of its members' time to operate, partially because judges write and rewrite for others as well as for themselves. We rightly accept these costs as the price of protection against the arbitrariness of any single judge. Nevertheless, I think we can take steps to make the collegial system work better. One thing courts need is deadlines. Judges' output can slip because they work alone so much of the time without external controls. Litigants have a right to expect cases to be disposed of in a reasonable time, and colleagues deserve prompt action on their drafts. We impose deadlines on lawyers; we should suffer some ourselves. By taking an unconscionably long time to decide, we put litigants in an intolerable position--frustrated but too polite to complain. *785 At present, our court has an internal working rule that bars a judge from hearing cases in a new term if he has not circulated draft opinions on more than three cases argued at least six months previously. The Supreme Court has an even stricter rule--a July 1 deadline for all opinions in cases argued that term. We also have recently adopted a rule, similar to that in several other circuits, requiring a judge to indicate his response to a circulated draft opinion within seven days, and permitting the judge who wrote the panel opinion to release it if he does not receive a dissent within thirty days. We need still another rule to require recirculation of a revised draft within a certain time after receiving comments or dissents. The larger our court gets, and the greater our caseload becomes, the more we need such rules. Group norms do matter to judges, even if the sanctions we impose for noncompliance are more moral than punitive. Another target for improvement is the techniques for communication among judges in areas of the law in which the court does not speak with one voice, or where fuller explanation is needed. As it is now, except with each panel, judges learn of each others' views only through circulated written opinions which, in the court's pressured work environment, often gain more dust than readership. It might make sense for the judges of the court to meet occasionally to discuss areas of law in the circuit that may need clarification, or have been left a bit murky. The purpose of sharing views on such topics would not be to establish a fixed agenda for action and definitely not to decide abstract issues. Rather, its purpose would be to make us more sensitive to our colleagues' interests and views, and perhaps to establish a general aura of agreement on our responsibility, as a court, to identify and to elucidate particular subjects. We might try such sessions once or twice a year, perhaps at the start of a new term. Our judicial conferences may have been originally designed for some such purpose, but now they are primarily social functions. Our monthly judges' meetings and Judicial Councils take up only administrative business. V. CONCLUSION Judges are in the business of making decisions, accurately and promptly. An increasingly complex society means more cases of increasing complexity. No magic solution to that is on the horizon. Judges legitimately need help--law clerks and central staff--in performing their judicial tasks. They must take care, however, that their reactions to a stepped-up work load do not corrode the essence of their judicial functions--reading and listening to the arguments of the parties, *786 being familiar with the record, making and explaining their decisions, generally for publication. Taking shortcuts in any of these areas is a danger signal for the process; on the other hand, paying more attention to scheduling practices and setting timetables for the judges may be necessary to make the collegial process work under pressure. Undramatic as such internal tinkering may seem, it holds more promise for preserving the essentially personal nature of judicial decisionmaking than the more global, but largely unproductive debate, on the bureaucratization of the judiciary. Ironically, the judiciary remains one of the few checks on the excesses of real bureaucracies, and it is vital that we retain our capability to perform that task. [FNa] Judge, United States Court of Appeals for the District of Columbia Circuit; B.A. 1948, Connecticut College; LL.B. 1951, Yale Law School. The author is deeply grateful to her law clerk, Robert Zoellick, for his hard work and thoughtful contributions to this article. [FN1]. See Hoffman, The Bureaucratic Spectre: Newest Challenge to the Courts, 66 JUDICATURE 60 (1982); McCree, Bureaucratic Justice: An Early Warning, 129 U. PA. L. REV. 777 (1981); Rubin, Bureaucratization of the Federal Courts: The Tension Between Justice and Efficiency, 55 NOTRE DAME LAW. 648 (1980); Vining, Justice, Bureaucracy, and Legal Method, 80 MICH. L. REV. 248 (1981); Edwards, A Judge's View on Justice, Bureaucracy, and Legal Method, 80 MICH. L. REV. 259 (1981). [FN2]. One innovative op-ed writer has suggested that judges should cope with the workload by firing their clerks. See Barone, Our Overworked Justices Should Fire Some Law Clerks, Wash. Post, Nov. 24, 1982, at A17, col. 1. [FN3]. See, e.g., Meador, The Federal Judiciary--Inflation, Malfunction, and a Proposed Course of Action, 1981 B.Y.U.L. REV. 617, 625-37 (providing a history of proposals over the last 20 years to restructure the federal judiciary), Powell, Are the Federal Courts Becoming Bureaucracies?, 68 A.B.A.J. 1370 (1982) (discussing possible areas in which to cut back federal jurisdiction); Stevens, Some Thoughts on Judicial Restraint, 66 JUDICATURE 177 (1982) (suggesting increased reliance on the doctrine of judicial restraint to avoid unnecessary lawmaking); Waxman, Alternatives to Litigation, 7 DISTRICT LAW. 29 (1983) (discussing alternatives to litigation for resolving disputes); Yellin, High Technology and the Courts: Nuclear Power and the Need for Institutional Reform, 94 HARV. L. REV. 489 (1981) (suggesting institutional reform of the judicial process, including a specialized body of experts within the judiciary, to aid in the resolution of nuclear power cases). [FN4]. This figure is for statistical year (SY) 1982 (July 1, 1981-June 30, 1982). These and other statistics for the United States Court of Appeals for the District of Columbia Circuit are compiled by the Clerk's Office and the Circuit Executive [hereinafter cited as D.C. Circuit Statistics]. [FN5]. Id. The District of Columbia Circuit disposed of 616 cases after argument or submission in SY 1982. During that same time period, the Supreme Court granted writs of certiorari in 20 cases from the circuit. [FN6]. In SY 1982, a total of 206, 193 civil cases were filed in U.S. district courts, a 14.2% increase over the previous year and a 57.9% increase over SY 1977. Criminal case filings in U.S. district courts in SY 1982 increased by 4.2% to 31,623. This was the second year the criminal caseload increased, after declines in the previous three years. The single biggest factor in the increase in civil litigation in the U.S. district courts was the surge in actions for recovery of overpayments and enforcement of judgments. Most of these cases involve the U.S. government's efforts to recover defaulted student loans and overpayments of veterans' benefits. See ANNUAL REPORT OF THE DIRECTOR OF THE ADMINISTRATIVE OFFICE OF THE UNITED STATES COURTS 92-132 (1982) [hereinafter cited as REPORT OF THE COURTS]. [FN7]. A total of 27,946 appeals were filed in the U.S. courts of appeals in SY 1982, as compared to 26,362 for the previous year. The D.C. Circuit's filings declined from 1,604 in SY 1981, to 1,447 in SY 1982, with a decrease in new administrative agency appeals of 40.8%. The Second Circuit's filings fell from 3,061 in SY 1981, to 2,827 in SY 1982, led by a decline in bankruptcy appeals of 36.4%. The three circuits that had filings increase over 15% were the First (15.3%), the Fourth (18.0%), and the Eighth (16.7%). Id. at 76-79. [FN8]. As of April 1983, total filings in the D.C. Circuit for SY 1983 were up 7% over SY 1982. See D.C. Circuit Statistics, supra note 4. [FN9]. In SY 1982, there were 504 appeals involving administrative agency decisions filed in the D.C. Circuit, out of a total of 3,118 administrative agency appeals filed in the 12 circuits. If the 838 NLRB appeals (22 in the D.C. Circuit) are removed from this calculation, the percentage of administrative agency appeals filed in the D.C. Circuit increases from 16.1% to 21% of the total. See REPORT OF THE COURTS, supra note 6, at 79-80. [FN10]. In the D.C. Circuit, filings of appeals from administrative decisions amounted to 34.4% of total filings in SY 1982. This was an abnormally low year. See supra notes 7-8. The average percentage figure for the five years from SY 1977 to SY 1981 was 45.3%. See D.C. Circuit Statistics, supra note 4. [FN11]. D.C. Circuit Statistics, supra note 4. [FN12]. See supra note 3. I have discussed elsewhere two alternatives intended to cope with the increased complexity of cases--the creation of more courts to deal solely with specialized subject matter, and reliance on expert advisers who would serve within the judiciary or as its employees. See Wald, Making "Informed" Decisions on the District of Columbia Circuit, 50 GEO. WASH. L. REV. 135, 149-54 (1982). [FN13]. See D.C. Circuit Statistics, supra note 4. [FN14]. In SY 1982, there were 50,555 diversity cases and 24,975 petitions by state prisoners among the 206,193 new civil cases filed in U.S. district courts. See REPORT OF THE COURTS, supra note 6, at 98-103. [FN15]. See, e.g., Shapiro, On Predicting the Future of Administrative Law, REGULATION May-June 1982, at 18, 23-25. [FN16]. The Court has recently gone to great pains to reemphasize that courts should be deferential in their review of agencies' interpretations of statutes. See, e.g., Federal Election Comm'n v. Democratic Senatorial Campaign Comm., 454 U.S. 27, 36-39 (1981); Federal Communications Comm'n v. WNCN Listeners Guild, 450 U.S. 582, 598 (1981). At least one commentator has also suggested that the Court's decision in Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519 (1978), may be followed by a decision of equal intensity dealing with standards of substantive review. See Verkuil, Judicial Review of Informal Rulemaking: Waiting for Vermont Yankee II, 55 TUL. L. REV. 418, 419, 424 (1981). But cf. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Ins. Co., 51 U.S.L.W. 4953, 4958 (U.S. June 24, 1983) (Vermont Yankee should not be invoked as "a talisman under which any agency decision is by definition unimpeachable"). [FN17]. See D.C. Circuit Statistics, supra note 4. [FN18]. S. 1080, 97th Cong., 2d Sess.
706(c), 128 CONG. REC. S2713, 2717-18 (daily ed. Mar. 24, 1982), as passed by the Senate on March 24, 1982, included a compromise between Senator Bumpers and the Administration on the scope of judicial review. The section stated: (c) In making determinations concerning statutory jurisdiction or authority under subsection (a)(2)(C) of this section [dealing with agency action in excess of statutory authority], the court shall require that action by the agency is within the scope of the agency jurisdiction or authority on the basis of the language of the statute or, in the event of ambiguity, other evidence of ascertainable legislative intent. In making determinations on other questions of law, the court shall not accord any presumption in favor of or against agency action, but in reaching its independent judgment concerning an agency's interpretation of a statutory provision, the court shall give the agency interpretation such weight as it warrants, taking into account the discretionary authority provided to the agency by law. See also H.R. 746, 97th Cong., 1st Sess.
706 (1981) (as introduced by Congressman Danielson with 27 other sponsors). [FN19]. See, e.g., Camp v. Pitts, 411 U.S. 138 (1973) (per curiam); Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 414-17 (1971) (construing the standards of review defined in 5 U.S.C.
706(2)). [FN20]. See generally A. DOWNS, INSIDE BUREAUCRACY (1967) (describing bureaucracies and their performance in the course of developing a theory of bureaucratic decisionmaking); A SOCIOLOGICAL READER ON COMPLEX ORGANIZATIONS (A. ETZIONI & E. LEHMANNNN 3d ed. 1980) (presenting essays on organizational analysis); C. LINDBLOM, POLITICS AND MARKETS 27-28 (1977) (providing a list of eight characteristics of a bureaucratic organization). [FN21]. See, e.g., A. TALBOT, SETTLING THINGS: SIX CASE STUDIES IN ENVIRONMENTAL MEDIATION (1983); Mosher, EPA, Looking for Better Way to Settle Rules Disputes, Tries Some Mediation, 15 NAT'L J. 504, 504-06 (Mar. 5, 1983); Negotiated Environmental Progress, Wash. Post (editorial), Mar. 7, 1983, at A12, col. 1. [FN22]. See Harter, Negotiating Regulations: A Cure for Malaise, 71 GEO. L.J. 1 (1982). [FN23]. Id. at 42-52. Harter notes nine predictive criteria on the likelihood of successful negotiations: countervailing power; limited number of parties; mature issues (subject matter should be a concrete issue and the parties' political jockeying should have settled down); inevitability of decision; mutual opportunity for gain; the dispute cannot concern fundamental values of the competing parties; multiple issues to trade; research would not be determinative of the outcome; implementation of an agreement is likely. [FN24]. See Nejelski & Zeldin, Court-Annexed Arbitration in the Federal Courts: The Philadelphia Story, 42 MD. L. REV. 787 (1983). [FN25]. See text accompanying note 7. [FN26]. In SY 1982, there were 132 authorized U.S. courts of appeals judgeships, 7 vacancies, and 54 senior judges. There were also 515 authorized U.S. district court judgeships, 20 vacancies, and 163 senior judges. In SY 1975, there were 97 authorized U.S. courts of appeals judgeships, 1 vacancy, and 47 senior judges. In that year, there were also 400 authorized U.S. district court judgeships, 13 vacancies, and 102 senior judges. See REPORT OF THE COURTS, supra note 6, at 3. [FN27]. Federal civilian employment in the legislative branch as of March 1982, totaled almost 39,000. The executive branch included over 2.8 million civilian employees. All the federal courts throughout the nation had about 15,700 employees, not much more than the Department of Housing and Urban Development's 15,000 people and a good deal fewer than the Department of Justice's 55,000 employees. BUREAU OF THE CENSUS, U.S. DEP'T COM., STATISTICAL ABSTRACT U.S. 26 (103 ed. 1982). [FN28]. Numerous authors have addressed various facets of court management. See, e.g., P. CARRINGTON, D. MEADOR & M. ROSENBERG, JUSTICE ON APPEAL (1976) [hereinafter cited as CARRINGTON]; J. MCDERMOTT & S. FLANDERS, THE IMPACT OF THE CIRCUIT EXECUTIVE ACT (1979); Nihan & Wheeler, Using Technology to Improve the Administration of Justice in the Federal Courts, 1981 B.Y.U. L. REV. 659; Solomon, The Training of Court Managers, 1981 B.Y.U. L. REV. 683. Some authors worry that the management movement, especially at the trial court level, has undermined the adversarial system. See, e.g., Resnik, Managerial Judges, 96 HARV. L. REV. 376, 424-31 (1982). My focus on management is directed at how the judicial decision-making process works at the appellate level. Such a process currently exists; I urge that we recognize and adjust it to achieve our objectives. [FN29]. See e.g., McCree, supra note 1, at 785-87; Vining, supra note 1, at 252-53; Barone, supra note 2. [FN30]. D.C. Circuit Statistics, supra note 4. [FN31]. All motions in our circuit--including stays, motions to dismiss, motions for summary affirmance, or reversals--go to a motions panel of two judges (with a third "on-call" to serve as a tie-breaker). The judges rotate motion panel duty, each panel of three serving for two months. Many of the motions must be decided within a finite--usually telescoped--period, if the decision is not to be made by default, e.g., FOIA or merger cases. The preparatory work on all such cases is not done by chambers law clerks, but by a pool of six court law clerks, chosen for one-year terms by a committee of judges. On average, the panels decide 10-20 motions each week. [FN32]. Leventhal, Appellate Procedures: Design, Patchwork, and Managed Flexibility, 23. UCLA L. REV. 432, 442 (1976) (footnotes omitted). [FN33]. McSurely v. McClellan, 697 F.2d 309 (1982)(per curiam). [FN34]. Covelo Indian Community v. Watt, No. 82-2377 (D.C. Cir. Dec. 21, 1982) (per curiam), vacated as moot, Feb. 1, 1983. [FN35]. McCalpin v. Dana, No. 82-2318 (D.C. Cir. order denying emergency motion for injunction pending appeal Dec. 16, 1982) (per curiam). [FN36]. Branan v. Hayes, No. 83-1035 (D.C. Cir. order denying injunctive relief Jan. 10, 1983) (per curiam). [FN37]. Community for Creative Non-Violence v. Watt, 703 F.2d 586 (1983) (en banc) (per curiam). [FN38]. In SY 1982, 51.2% of the D.C. Circuit's 627 decisions were disposed of by order or judgment with only a brief unpublished statement. The figures for SY 1981 and SY 1980 were 45.9% and 46.8%, respectively. See D.C. Circuit Statistics, supra note 4. [FN39]. See CARRINGTON, supra note 28, at 31-35 for an explanation of the usefulness and characteristics of such memorandum opinions. The authors provide examples of several possible styles of such abridged decisions in Appendix B. Id. at 243-53. [FN40]. CARRINGTON, supra note 28, at 39-40. [FN41]. After the court has decided whether to write an opinion, and how to sign it, the judges must determine whether it should be published. Our circuit, like others, has a plan for publication of opinions that lists criteria that are supposed to govern which opinions we publish. See Plan for Publication of Opinions (Apr. 17, 1973); see also D.C. CIR. R. 8(f), 13(c); see generally Reynolds & Richman, The Non-Precedential Precedent--Limited Publication and No-Citation Rules in the United States Courts of Appeals, 78 COLUM. L. REV. 1167 (1978) (examining the merits of courts' rules on publication and citation of opinions). But the "publish or not publish" debate has come to the fore again recently. The New York Times reported that in the Second Circuit, "[t]he judges say they cannot draft a well-reasoned opinion in every case, so they save time in writing opinions that are not intended for publication." Chambers, U.S. Appeals Court Restricts Use of Opinions by Lawyers N.Y. Times, Feb. 21, 1983, at B1, col. 1. According to the article, the New York City Bar Association is starkly critical of this practice, which results in placing two-thirds of the court decisions "outside of the normal reach of the bar." Id. at col. 2. These attorneys are unmoved by the argument that publication of all decisions "creates tons of useless print and overwhelms law libraries, law students, lawyers and judges alike." Id. The bar understandably deplores unpublished decisions because they cannot be cited as precedent or even brought to the attention of the judges. And they are asserted to have a discriminatory impact; while "unpublished opinons" are practically inaccessible for many lawyers, they can be obtained from the clerk's office, and large firms, the government, and private subscription services routinely collect and index them for background. It is argued that the option to decide which of those opinions will not be published permits judges too much discretion to decide which opinions will become the law of the circuit. The practice masks and therefore increases the likelihood of nonuniform dispositions, impedes informed client counseling, and may leave counsel groping in darkness. Yet 55% of all criminal cases and 63% of all civil cases decided in the federal circuit courts were unpublished in SY 1979. Id., at B5, col. 2. See generally Reynolds & Richman, An Evaluation of Limited Publication in the United States Courts of Appeals: The Price of Reform, 48 U. CHI. L. REV. 573 (1981) (presenting an empirical assessment of the workings of the publication plans of the 11 U.S. courts of appeals during SY 1978-79). The practice has an obvious bearing on judicial accountability. Although unpublished opinions may indeed save time, they limit the public's ability to evaluate the correctness of judicial actions and give rise to uncertainties about the integrity of the courts. For these reasons, I believe cases should be published and collected in an accessible form. It need not be the same as the Federal Digest System, which could be reserved for longer, more elaborate opinions. The short memorandum opinion I discussed above would be entirely appropriate for inclusion in some kind of summary and, I would hope, inexpensive reporting system. Lawyers could cite these decisions, but judges could reject them as binding precedent where they were not supported by sufficient reasoning. This argument is developed in greater detail in CARRINGTON, supra note 28, at 35-41. [FN42]. See D.C. Circuit Statistics, supra note 4. END OF DOCUMENT