THE OBSOLESCENCE OF THE UNITED STATES COURT OF APPEALS:  ROSCOE POUND’S STRUCTURAL SOLUTION

Paul D. Carrington*

Published by the Journal of Law and Politics.

The Commission on Structural Alternatives for the Federal Courts of Appeals has done excellent work and its recommendations with respect to the Ninth Circuit should be adopted forthwith.  Its proposals are, to be sure, a compromise that no one is likely to regard as a perfect outcome, but the report is constructively responsive to the problems the Commission was called to address and no one can reasonably contend that its recommendations threaten harm to any important value.  Particularly given the enormous political inertia to which I will advert below, Congress should act without hesitation before this report, like so many of its predecessors, slides into the oblivion that is the politics of judicial administration.

The Commission has stopped just short of recommending two other steps that should also be taken at once, for they are long overdue.  First, as Erwin Griswold noted over a half century ago,[1] it is unjust to decentralize tax appeals.  Everyone should pay the same taxes and equality can be achieved only if the interpretation of the Internal Revenue Code is placed in the hands of a single federal appellate court.  The Federal Circuit is available for that purpose.  If all tax appeals were centered there, there would be many fewer tax appeals and legal costs to both taxpayers and the government would be reduced; the Supreme Court would be liberated from any responsibility for tax law other than its constitutionality; and we would all pay taxes according to the same law.  The only interests that would be adversely affected are those of taxpayers hoping to avoid a just tax by forum shopping and of lawyers and accountants who profit from the indeterminacy of tax law.  There was perhaps a time when one could reasonably express concern for the access of taxpayers to their neighborhood court of appeals, but that concern has been trivialized by technological advances; the Federal Circuit can hear tax appeals from fifty states without the judges or lawyers leaving their chambers.[2]

Likewise, the proposal of an Article I court for social security cases is undebatably prudent.  The present system merely adds an element of indeterminacy to the administration of social security law.  Social security claims are claims against the United States and there is no significant difference between them and other classes of cases that are tried in the Claims Court subject to the exclusive appellate jurisdiction of the Federal Circuit.  The Claims Court is as accessible as the district courts, and claimants as well as the government would be spared expense and delay by the centralization of those appeals.

Even if the steps proposed by the Commission and these two additional reforms are enacted, the central difficulties that the Commission was summoned to resolve abide.  Those difficulties are secondary consequences of a precipitous increase in the appellate caseload that commenced about 1960 and has only in recent years shown signs of leveling off.  In response to that increase, the United States Courts of Appeals have given diminishing attention to performing the role for which they were established, in order to focus their efforts on a mission that no one has commissioned them to perform and which, for structural reasons, they are increasingly unable to perform.

The intermediate federal courts were established in 1891 to relieve the Supreme Court of its responsibility for correcting the errors of the individual district judges who exercised such extraordinary power over fellow citizens and who were at the time almost unaccountable for their decisions because of the overload on the Supreme Court, and to assure defeated litigants that the judicial power brought to bear on them was the work of more than a single perhaps idiosyncratic individual.[3]  To that duty was later added the responsibility of correcting the errors of federal administrative agencies at risk of politically motivated or bureaucratic mischief.  These traditional functions of the courts of appeals can be denoted as protections of procedural or institutional values.  Their common purpose is to win trust for the federal legal system as a whole institution.

The task to which the courts of appeals have called themselves is that of making the national law as applied to their geographical territories.  Increasingly, they perceive themselves as junior supreme courts providing “doctrinal development for most legal issues”[4] by maintaining the “law of the circuit.”[5]  That aim can be distinguished as substantive because it entails the discernment and refinement of the political substance of the national law.

The substantive law of the circuit that courts of appeals strive to express is transitory.  This is so because the range of legal issues arising within the appellate jurisdiction of the courts of appeals is limited; most of the matters within their reach are not likely to remain settled no matter how lucid and persuasive the opinions written about them.  Courts of appeals do not make constitutional law; that is for the Supreme Court of the United States.  They do not make common law; that is for highest state courts.  What remains is the interpretation of federal legislation and federal administrative regulations that can be and frequently are modified by Congress or administrative agencies or reinterpreted more authoritatively by the Supreme Court. Court of appeals interpretations of transitory texts are therefore rarely of enduring consequence in future cases.

The law of the circuit is not only transitory, it is largely illusory.  The courts of appeals are severely limited in their ability to make law, in the sense that the Supreme Court or highest state courts do, for the reason that they sit in small, randomly selected groups.  As Roscoe Pound observed long ago, the three-judge panel was designed to be and is an instrument for correcting error and diffusing the responsibility for decisions made by individual judges and administrative agencies.[6]  By sitting in panels, an appellate court can substantially increase the number of appeals it can handle for those basic institutional purposes.  But the small panel is not suited to the function of making law.

This is so because of constraints that inhere in the concept of judge-made law.  The utterances of those highest courts that almost always sit en banc are regarded as law because they are a reasonably coherent and reliable forecast of what the same judges will do in similar cases.  As Karl Llewellyn has emphasized, lawyers reading such opinions can read them with a knowledge of the persons who write them and with the knowledge that there is likely to be substantially the same membership of the court over a significant period of time.[7]  The reality underlying this observation is that words gain meaning from particular knowledge of their users.  Among family members, a few words can transmit much meaning; among strangers, many words may transmit little.

An opinion of a small, randomly selected panel is, in contrast, a mere sample of the reactions of judges, and speaks unconvincingly as a voice of the next randomly selected panel.  The perpetual rotation from one case to the next in the memberships of panels makes the judges strangers to those whom they address.  Try, as they do by means of en banc proceedings[8] to bring one another into line with group thought and group values,[9] panels cannot be relied upon by the bar to fully understand and to heed one another.   Adherence to circuit precedent can be confidently reckoned only if the issue resolved has slight political content, i.e., one about which the judges on the second panel have no strong predilections.  If circuit precedent seems to a panel of three judges to lead to a disfavored result, it is generally feasible for the sitting judges to distinguish the former decision even if the distinction is one that the previous panel would not have been likely to make.  And there is also now often available the opportunity to decide a case without opinion, obviating any need to deal with unwelcome precedent.[10]

Obviously, as judgeships are added to a court of appeals to deal with the caseload problem, the law of the circuit becomes ever more illusory.  En banc procedure becomes ever more cumbersome and ever less effective.  Lawyers increasingly see the appeal as a game of chance, the outcome to depend on the luck of the draw in the assignment of judges to the panel to hear their cases.  And district judges share that perception,[11] with the inevitable result that they sense increasing autonomy and discretion.  If the outcome of an appeal will depend on the identity of the members of the panel assigned to decide it, and their identities cannot be known at the time the trial judge must decide, even the most deferential trial judge cannot decide according to the law of the circuit.

Because the law of the circuit is transitory and illusory, it has limited marginal value. If the Federal Reporter were to cease publication, the loss to the legal profession in its ability to serve clients in planning transactions or arguing a legal point in the district courts would be only marginally greater than if the leading student-edited law reviews were to fall under the sea.  This observation is confirmed by the facts that

(1)   there is virtually no secondary literature attempting to restate or synthesize the law of any circuit with respect to any topic of the national law;

(2)    lawyers seldom if ever plan transactions on the basis of the law of the circuit; and

(3)   lawyers and their clients infrequently base a decision to appeal or not, or to settle a pending appeal, on the basis of precedent established in a single opinion of a three-judge panel of the circuit to which the appeal is taken, at least if there are plausible reasons why a different panel might be willing to distinguish or disregard it.

I have no empirical data to validate these observations, but the appellate lawyers with whom I converse confirm my impressions, and the last of the three is confirmed by the current movement to compel federal government lawyers by law to “acquiesce” in circuit precedent.[12]  An essential premise of that proposal, which is supported by the American Bar Association and the Judicial Conference of the United States, is that government lawyers do not always believe what they read in the Federal Reporter.  We know that government lawyers are generally more cautious in taking appeals than are private lawyers; [13] if they do not seem to be convinced by the utterances set forth in the Federal Reporter, we can be sure that lawyers paid by the hour to represent private interests do not. One who doubts my observation is encouraged to examine the mint condition of the spines of the volumes of the Federal Reporters resting on the shelves of law firms, or count the “hits” on courts of appeals opinions by lawyers using Lexis or Westlaw.  

The opinions of the courts of appeals are, of course, used by lawyers writing briefs and their availability for that use occasions and justifies many billable hours.  But if there were many fewer of them, the briefs would still be written, and the world would go on very much the same.  There would be less judicial rhetoric to imbibe, so perhaps the briefs would be a bit shorter and take less time to write.

Given the history and the limited consequentiality of their work product, why should circuit judges be so preoccupied with writing for the Federal Reporter?   I have heard some unkind speculations.  A Freudian psychiatrist might detect in the judges a deep-seated territorial instinct not far different from that manifested by district judges and other professionals.  Perhaps meaner is the suggestion that the Federal Reporter has become a vanity press, like academic journals, a mere outlet for expression by literate men and women wishing to exhibit their talents to their peers. A more charitable explanation is that circuit judges learned in law school from studying cases that appellate judges make law.  It is therefore their duty to make some, notwithstanding the structural impediments they face.  Most circuit judges sought and accepted their judicial commissions in the expectation that as appellate judges they would influence the shape and content of the national law, and few of them are likely to be disabused of that illusion by their experience on the bench or by the present utterance.  Moreover, they are reinforced in their impulses by their law clerks, who come to them fresh from the study of judicial opinions and ready to participate in lawmaking; judges who do not make law are disadvantaged in the competition to recruit law clerks.  They are also reinforced by their peers, whose sense of status is in some measure linked to the shared perception that they are making important law.

Because of the presence of the law clerks and the peer pressure of other judges, it would be difficult to dissuade circuit judges from their impulse to engage in lawmaking even without the influence of the Judicial Conference of the United States.  However, that body was created in this century to reflect and strengthen the professionalization of the judiciary being effected by Progressives in the early decades of this century.[14]  The Conference is a sort of professional association (I do not say a labor union) for federal judges.  As Stephen Yeazell has recently observed, it is highly resistant to any proposals, or even perceptions, that might diminish the professional gratification or social status of members of the federal judiciary.[15]  Like a bar association or an academic faculty, or any other professional group, it is locked into a world view centered on the belief that whatever its members have been doing must be continued and enhanced.  There can never be too much judicial discretion to suit the Conference.  The preferred habits of such an organized profession do change, but usually only in response to powerful external forces such as a market.

Increased caseload of the courts of appeals is of course such an external force, but it has been accommodated without jeopardy to the desires of circuit judges to make law by publishing learned opinions.  One response has been to delegate much of the more onerous responsibility to law clerks and other staff.  Particularly on the criminal side, it is increasingly difficult for an appellant to attract the notice of a real Article III federal judge.  A second response has been to dispense with arguments and signed opinions, because they are the stages of the appellate process requiring the first person presence and participation of an actual judge.  This leaves the circuit judges free to concentrate on working with their elbow clerks on learned opinions stating how the national law should be interpreted.  From the perspective of the litigant, the federal appeal is an increasingly invisible process.

Because the institutional concerns animating the creation of the courts of appeals are everyone’s business, they are no one’s special concern, and there is no political constituency for reform facing squarely the erosion of the quality of the federal appeal as an assurance that district judges and administrative agencies are accountable for the exercise of their vast powers. The institution having the most influence with Congress on these matters is likely the Judicial Conference.  It is also the interest group having the least receptivity to concern for the procedural values of 1891. There is no special political interest on the other side to convince the emperors that they have no clothes. While there are many lawyers who regret the lack of oral argument and who suspect that their appeals are not receiving the sober attention of Article III judges, their regrets are episodic and those sharing them are not a political constituency.

Given this political context, the Commission faced a Herculean political challenge.  It is one that many others have confronted without significant effect.  Perhaps the first person to call attention to the present problem of structural alternatives was Erwin Griswold.[16]  For a time, the Tax Section of the American Bar Association agitated for attention to the structural problem as he identified it, and as the Commission noted, it has been revived on several occasions.

In 1965, I was appointed to direct an American Bar Foundation study of The Business of the United States Court of Appeals.  We reported in 1968 on the problem to which the present commission was addressed, and reviewed alternative solutions.[17]  It is fair to say that no entirely new ideas have surfaced in the intervening decades, not because my work was so admirably complete, but because the dimensions of the problem are limited and obvious.

My work was soon followed by that of a committee appointed by the Chief Justice and headed by Paul Freund; their attention was directed to the workload of the Supreme Court of the United States, but they quickly perceived that the problem of that Court’s docket is part of the larger problem of the structure of the federal judiciary.[18] There was then the Advisory Council on Appellate Justice, an organ of the Federal Judicial Center funded by the Law Enforcement Assistance Administration, which sat for some years in the early seventies under the leadership of the late Maurice Rosenberg and conducted a national conference in 1975 at which the issues were again reviewed, this time in relation to the comparable problems of state courts, and with four volumes of readings for conferees.[19]  A secondary product of that organization was published in 1976 as Justice on Appeal, a work of which Professor Meador and I were co-authors along with Rosenberg.[20]   Coming on the heels of that initiative was the report of the Hruska Commission on Revision of the Federal Appellate Court System, a body created by Congress and well staffed to study the same universe of related problems.  It published a good report in 1977.[21]  Leo Levin, the director of the Federal Judicial Center renewed the commission’s recommendation a few years later,[22] and so did Chief Justice Warren Burger[23] and Justice Byron White.[24] Then came the Federal Courts Study Committee, also created by Congress and chaired by Judge Joseph Weis; they made a fine report in 1990.[25]  In 1993, as directed by the Study Committee, the Federal Judicial Center completed and published another comprehensive study of appellate court structure.[26]  Professor Thomas Baker, a participant in these events, published a learned work in 1994[27] again reviewing the options and urging action. His work was followed in 1995 by the Long-Range Planning Report of the Judicial Conference of the United States.[28]  In 1996 came yet another insightful observation by Martha Dragich.[29]  All of these groups and individuals acknowledged the structural problem; many of them made reasonable suggestions responding to them.  All of them attracted a measure of outspoken resistance going to the details of specific proposals, but few indeed have denied the problem they addressed.

It would not be accurate to say that nothing has come of their efforts.  Here and there, changes have been made.  But mostly Congress has been content to continue to create additional judgeships and to massively enlarge the number of staff personnel to whom much of the work of the appellate courts has been delegated, reforms that were sought by the Judicial Conference and were congruent with the general drift away from the correction of error to the illusory law-making function.  The most important structural change was the formation of the United States Court of Appeals for the Federal Circuit,[30] which was the result of efforts of Professor Meador when he was serving as Assistant Attorney General of the United States.

On the available evidence, one would have to conclude that the structural problems of the federal judiciary are insoluble for political reasons, at least until some extraordinary crisis compels fundamental re-thinking, and it is hard to imagine what such a crisis might look like.  Meanwhile, we will drift toward increasing discretion for both circuit and district judges,[31] as both will be less accountable to systemic constraints.  As I noted a decade ago, one is moved to think of the Supreme Court of the United States and the Congress as ever more closely resembling Napoleon at Borodino, as described by Tolstoi:

[I]t was not Napoleon who directed the course of the battle, for none of his order was executed and during the battle he did not know what was going on before him . . . [The battle] occurred independently of him, in accord with the will of hundreds of thousands of people who took part in the common action.  It only seemed to Napoleon that it all took place by his will.[32]

A tertiary consequence of this increasing elasticity in the system is that the national law is less “reckonable,” and hence less a restraint on private actors and less reliable by those who plan transactions.  At the margins, there will be more conduct that the law intends to deter as mischief.  Also at the margins, some transactions that depend heavily on the assurances that stable law provides will not be consummated.  But these erosions proceed at so glacial a pace and their effects are so intermixed with other causes that they never have and perhaps never will be visible or even measurable.  Accordingly, they may never present a crisis sufficiently intense to evoke the political response needed for effective correction.

While the Commission proposes to do little more than tinker with the structural disabilities, among the other possibilities that it has considered and not rejected is a scheme first proposed by Roscoe Pound, the most notable judicial law reformer of this century.  Pound, unlike the present author, did not write from ignorance of reality.  From 1901 to 1903, he served as a part-time commissioner for the Supreme Court of Nebraska.[33]  He heard and decided by opinion over 250 cases.[34]  A variation on his scheme was propounded by Shirley Hufstedler in 1967;[35] she, too, is a person of substantial experience on both trial and appellate benches.[36]  She and her husband, Seth, then proposed a further variation for use in the California judicial system;[37] their proposal was discussed[38] but not adopted.  I endorsed the Hufstedler plan as a suitable response to the federal problems in 1987.[39]

Pound emphasized that the appeal, if it does nothing else, is necessary to assure the litigants and the public that the judicial power is not vested in a single individual, but is exercised only by a larger institution.  This emphasis on diffusing the responsibility for applying the lash of the judicial power was in keeping with the premises of those who enacted the 1891 legislation creating the courts of appeals.  It is the hope of attaining that purpose in the federal courts that animated Shirley Hufstedler thirty years ago.  So far are we from the attainment of that purpose that we now hear serious proposals to make even the first level of review discretionary,[40] an idea that is rightly resisted by the present Commission.

To provide for the essential assurance, both Pound and the Hufstedlers proposed that the appeal be consolidated with the motion for new trial, with a hearing to be conducted before a three-judge panel sitting in the district court.[41]  One effect of this scheme is to eliminate the need to take a case from one court to another; in form, the structure might resemble that of the Supreme Court of New York with its trial and appellate divisions.  One of the three judges sitting on the review panel might be a circuit judge who performs almost exclusively the function of review; certainly the three should not be a stable group sharing the same office address as the trial judge because the purpose, indeed the only purpose of the proceeding, is to correct the errors of the judge conducting the trial.

It was also a feature of Pound’s scheme that this post-trial proceeding be conducted orally.[42]  For him, oral argument was not just a formality, but was also an instrument of accountability for both lawyers and judges.  There are many legal arguments that would not be advanced at all if the advocates knew that they would soon be required to appear in a courtroom before judges, stand, and make an oral presentation of their transparently vulnerable contention. And a judge who asks sensible questions on oral argument exhibits the reality that he or she has read and listened, which are the indispensable activities of judging.  Pound also hoped by this means to shorten the time within which litigation was brought to termination, thereby reducing the use of the appeal as a dilatory device. 

Modern electronic transmissions have made the sort of proceeding he envisioned much more feasible; it can be conducted without either lawyers or judges leaving their desks.[43]  Two lawyers and three judges in five different locations can share the same computer screen.  Given the available technology, it might be best in most cases to have the oral argument first, and call for written briefs only if the appellate panel is uneasy about deciding a case “from the bench” (i.e. on the computer monitor) in prompt response to the argument.

Thus, Pound was skeptical regarding the value of published opinions.  He advocated the use of oral opinions.  These can be presented seriatim in the traditional English manner, a manner that is employed in many other legal cultures.  The purpose of such an oral opinion is not self-consciously to “make law,” but is to confirm to the parties that each of the judges was attentive to their arguments and has been intellectually engaged in deciding their case.  The oral responses might be followed by a brief written disposition that could include a citation of authorities and, if there is a remand, some words of advice to the judge as how the reversible error might be best corrected.  In performing this role, the judges would have much less use for law clerks than they do in writing opinions for publication in the Federal Reporter.

Having thus assured the procedural values of the appeal by the consolidated hearing in the trial court, Pound would join those who now insist that the right of appeal to the court of appeals should be abrogated.  One review is enough; further effort to correct errors is an unjustified burden on the parties.  He acknowledged the risk that there would be inconsistencies in the results reached by different three-judge panels.  He did not dismiss the desire for perfect consistency as a hobgoblin of small minds, but he did conclude that some indeterminacy in the administration of the law is inevitable.  Indeterminacy is best endured if we do not call attention to it by publishing learned opinions that are inconsistent.  And there will be less inconsistency if judicial decisions are not the work of single sometimes idiosyncratic judges.

Of course, Pound would allow for further review in the Supreme Court to respond to substantial and important questions of national law raised in the appeal-new trial proceeding.  These questions might be certified by the three-judge panel, or might be considered on petition of counsel.  If the Supreme Court cannot answer all the questions presented to it, then perhaps it needs a delegate, a genuine alternative forum with a national jurisdiction of the sort proposed by the several commissions and committees previously considering the matter, or perhaps other national institutions resembling the present Federal Circuit. 

But it seems unlikely that Pound or anyone else would as a fresh question suggest a need for a regional structure resembling the present courts of appeals.  If the basic function of the appeal were performed in the proceeding envisioned by Pound and outlined here, there would be no reason to retain the present courts of appeals.  It would be apparent that their continued existence served chiefly to provide professional satisfaction for circuit judges and law clerks.  The United States Courts of Appeals could go the way of the United States Circuit Courts, which were abolished in 1912 because they were essentially redundant.[44]

Although the Commission’s proposals, by linking groups of district judges more closely to groups of circuit judges, move us a step closer to the Pound solution, his vision seems very remote at the present time.  It may become less so as experience with electronic transmission accumulates.  When the implications of that reality are fully assimilated, some variation on his proposals may become obviously the thing that must be done to provide speedy and efficient justice. 

Whether even that development could suffice, given the instinctively ossifying effect of the politics of judicial administration, is doubtful.  It is hard to imagine that the Judicial Conference of the United States would suffer peaceably the abolition of any existing judicial institution.  Had it existed in 1910, it seems not unlikely that we would still have not only the United States Circuit Courts, but also the Commerce Court whose sole jurisdiction was to review decisions of the Interstate Commerce Commission.  One might be tempted to call upon Congress itself to devote some serious attention to the structural problem, but alas it has been some decades since either Judiciary Committee has given sustained, intelligent attention to any serious problem of judicial administration. 

Hopefully, these are merely the musings of an elderly observer who has seen too many reports to too many judiciary committees.  Optimism in the affairs of our Republic has thus far proved to be justified.