When a subcommittee of the House Judiciary Committee held a hearing in March 2017 on “ideas for restructuring the Ninth Circuit,” one surprise emerged: there were repeated references to the report and recommendations of the Commission on Structural Alternatives for the Federal Courts of Appeals, generally known as the White Commission after its chairman, the late Supreme Court Justice Byron R. White. The subcommittee chairman, Rep. Darryll Issa (R. Calif.) went so far as to say that the White Commission’s proposals may provide a “middle ground” between those who want to break up the circuit and those who do not. Notwithstanding the complimentary references at the hearing, it would be a great mistake to revive the White Commission’s approach to restructuring the Ninth Circuit. This article explains why. The Commission purports to embrace one of the principal arguments made by opponents of a circuit split: the Ninth Circuit Court of Appeals “should continue to provide the West a single body of federal decisional law.” But the Commission’s plan would not accomplish that objective. Under the Commission’s plan, the Ninth Circuit would remain intact, but the Ninth Circuit Court of Appeals would be divided into three “regionally based adjudicative divisions.” Each regional division would function as a “semi-autonomous” decisional unit. The circuit-wide en banc process would be abolished, and divisional decisions would be binding only within the division. There would also be a “Circuit Division” whose sole responsibility would be to resolve conflicting decisions between the regional units. This “Circuit Division” is the keystone of the Commission’s plan to maintain “desirable circuit-wide uniformity.” But when one digs into the details, it becomes clear that there is less to the Circuit Division than meets the eye. The Circuit Division would resolve only “square” conflicts – a category apparently limited to cases in which one division has explicitly rejected another’s precedent. It would act only upon the request of a party, and it would probably limit itself to issues that affect the operations of multicircuit actors – a circumstance that is the exception rather than the rule. And the Circuit Division would resolve only the issue in conflict, not the entire case. In short, decisions of the Circuit Division would be so infrequent, and their effect on the law of the division so limited, that “the law of the circuit” would shrink to near-insignificance. The Commission plan thus fails to provide the “middle ground” that Chairman Issa hoped for.

The Ninth Circuit Court of Appeals is by far the largest of the federal appellate courts. The court’s cases are ordinarily heard and decided by panels of three judges selected at random from an array of 29 active judges, 12 or more senior judges, and scores of visiting judges. In March 2017, a subcommittee of the House Judiciary Committee held a hearing on “ideas for restructuring the Ninth Circuit.” Proponents of restructuring argued that the size of the circuit, with so many judges responsible for the decisions, results in “the loss of predictability.” One of the Ninth Circuit’s own judges told the subcommittee that “because of its size,” the Ninth Circuit “is not and cannot be a reckonable court.” Are these assertions correct? No one would argue that appellate outcomes can be made entirely predictable. Some kinds of unpredictability are inherent in a common-law system of adjudication. Others, though perhaps not inherent, derive from established practices that, realistically, must be taken as a given. But proponents of restructuring argue that appellate outcomes are more unpredictable in the Ninth Circuit than in other, smaller circuits. This article reports the findings of an empirical study designed to test that assertion. When we look at the decisions of the courts of appeals, we know their actual outcomes. How can we identify the cases whose outcomes were unpredictable ex ante? The answer is that we cannot. What we can do is to analyze a subset of those cases that is representative of the entire group – cases with dissenting opinions. The existence of a dissent tells us that the outcome was attended by at least some degree of contingency. The grounds of dissent enable us to determine whether that contingency can be linked to circuit size. Disagreement within a panel does not implicate size if, for example, the majority and the dissent each rely on a different Supreme Court precedent. Or if the judges disagree over whether the district court’s findings of fact were clearly erroneous. Dissent implicates circuit size only if the disagreement suggests incoherence or inconsistency in the law of the circuit. Drawing on my prior work, I have identified four situations in which disagreement within a panel would signal possible instability in the law of the circuit. In the most acute of these, the majority and the dissent each identify a different in-circuit precedent as being directly on point. If critics of the “jumbo” court are correct, we would expect to find that, in cases with dissents, Ninth Circuit judges argue substantially more often than judges elsewhere that circuit precedent requires a particular outcome. Instead, we find that there is no meaningful difference between the Ninth Circuit and other circuits in the tendency of judges to invoke the compulsion of in-circuit precedent. This finding, reinforced by prior research on intracircuit conflict, suggests that actual or arguable clashes between circuit decisions contribute no more to unpredictability in the Ninth Circuit than they do in smaller circuits. The study also addresses another controversy over the relationship between predictability and appellate structure. Distinguished participants in the legal system have argued that the law is unnecessarily unpredictable because of a dearth of nationally authoritative precedents. To test this assertion, I examined the actual treatment of out-of-circuit precedent in cases with dissents; I then considered how that treatment might altered under a hypothetical regime of national stare decisis. I found that in the overwhelming majority of instances in which the outcome of an appeal is debatable among jurists of reason, the court can decide either way without rejecting any precedent from another circuit. By the same token, a regime of nationwide stare decisis would not alter the element of contingency that those appeals present. The study encompassed all cases with dissenting opinions in calendar year 1993 – a total of 771 cases and 828 distinct issues.

The time has long passed when the Supreme Court resolved every intercircuit conflict properly brought before it in a petition for certiorari. Is that a problem we should be concerned about? Three decades ago, Congress asked the Federal Judicial Center, the research arm of the federal judiciary, to conduct a study to ascertain “the number and frequency of conflicts among the judicial circuits … that remain unresolved because they are not heard by the Supreme Court.” Congress further requested that the Center determine the extent to which the unresolved conflicts are “intolerable.” The Center asked me to design and conduct the study. The study found that although the number of unresolved conflicts was larger than previous studies had suggested, the numbers alone gave a distorted picture of the phenomenon. Most of the conflicts that the Supreme Court does not hear either do not generate any of the consequences that might make them “intolerable” or do so only for a short period of time. Subsequently, I undertook another empirical study of unresolved intercircuit conflicts. It contained two elements. First, I investigated the later history of the conflicts identified in the initial project. Second and more ambitiously, I carried out a unique program of field research that included individual and group interviews of attorneys, examination of continuing legal education materials, and an in-depth survey of members of two specialty bars. This article reports the results of this later research. The principal conclusion is that the problem of unresolved conflicts exists only if you look for it—and look for it in a certain way. If you concentrate your attention on individual court of appeals decisions that create conflicts and on individual denials of certiorari in conflict cases, you will see (in the words of Chief Justice Rehnquist) “a judicial ‘darkling plain’ where ignorant armies [clash] by night.” But if you look at the conflict issues over a period of time and in context, you will find, if not certitude, a landscape in which courts build upon and reexamine one another's decisions in the untidy but constructive tradition of the common law. The field research also suggests a broader point. The very language that we use to talk about conflicts may convey a misleading picture of what is going on. We say that an “issue” gives rise to a “conflict” that the Supreme Court “resolves” (or does not “resolve”). This language implies that the dimensions of the “issue” remain unchanged from the first decision through consideration by two or more courts of appeals and ultimate resolution by the Supreme Court. The implication is probably correct when the issue is binary and discrete – for example, the “Cartwright issue” (whether mutual fund shares in a decedent’s estate are to be valued at the bid price or the asked price). For other issues, however, the reality will not be so straightforward. Judges and lawyers may continue to frame the issue in the same words, but beneath the verbal identity may lie important differences in the propositions that are accepted as part of the legal landscape and those that are viewed as litigable. Similarly, when the Supreme Court does intervene, its decision may end up reshaping the framework for litigation and counseling about an issue rather than closing a chapter in the particular area of the law. This pattern can be viewed as a departure from the traditional model of lawmaking by appellate courts. In that model, as memorably described by Lord Mansfield, the law “works itself pure from case to case.” In the situation I have just outlined, the law never “works itself pure;” rather, uncertain boundaries and gray areas characterize the legal landscape as long the issue, in some form, remains alive.

From 1971 through 1988, the United States Supreme Court was hearing and deciding about 150 cases each Term. Since the mid-1990s, however, the number of plenary decisions each Term has generally ranged between 75 and 85. How can we explain the shrunken docket? To answer that question, we must identify the changes that have taken place in the cases selected for plenary review and analyze them in a systematic fashion. In this article, I compare the composition of the plenary docket during the three most recent Terms (1993 through 1995) with the docket 10 years earlier. Using these data, I test five hypotheses that have achieved special prominence, including some that have been endorsed by one or more Justices. They are:1. The virtual elimination of the Supreme Court's mandatory appellate jurisdiction allows the Court to deny review in some cases that would have received plenary consideration under the pre-1988 regime.2. After the retirement of its three most liberal Justices, the Court took fewer cases in which lower courts had upheld convictions or rejected civil rights claims.3. Twelve years of Reagan-Bush judicial appointments brought greater homogeneity to the courts of appeals, resulting in fewer intercircuit conflicts that the Supreme Court had to resolve.4. The Federal Government was losing fewer cases in the lower courts and therefore filed fewer applications for review in the Supreme Court.5. The 12 years of Reagan-Bush appointments made the courts of appeals more conservative, resulting in fewer activist decisions of the kind that a conservative Supreme Court would choose to review.None of these hypotheses fully explains the shrunken docket. Rather, the data suggest that the Justices who joined the Court starting in the mid-1980s took a narrower view of the Court’s role than did their predecessors. The result is an “Olympian Court” that is often detached from the work of lower courts – an aloofness that poses a threat to the effective performance of the Court’s role in the American legal system.

For two decades, prominent judges and scholars have expressed concern that numerous intercircuit conflicts remain unresolved because the Supreme Court does not have the capacity to hear them. Other writers have disputed the existence of a problem. Congress, adopting a recommendation of the Federal Courts Study Committee, asked the Federal Judicial Center to conduct an objective study to settle the controversy. The Center in turn commissioned Professor Hellman to undertake the research. In this article, Professor Hellman describes the methods of the study, the underlying jurisprudential issues, and the results. Ne concludes that although the number of unresolved conflicts is larger than previous studies have suggested, the numbers alone give a distorted picture of the phenomenon. Most of the conflicts that the Supreme Court does not hear either do not generate any of the consequences that might make them ''intolerable'' or do so only for a short period of time. Overall, the study points strongly to the conclusion that unresolved intercircuit conflicts do not constitute a problem of serious magnitude in the federal judicial system.

“It is our considered conclusion that the Fifth Circuit is geographically too large and that 15 judges is definitely six too many. … Moreover, it is only natural that intracircuit conflicts multiply when there are 15 active judges. … Jumboism has no place in the Federal Court Appellate System.” That was the view of eight judges of the former (six-state) Fifth Circuit in a statement submitted to the Commission on Revision of the Federal Court Appellate System (Hruska Commission) in 1973. Today, the nine-state Ninth Circuit is much larger geographically than the old Fifth Circuit. And when the Ninth Circuit Court of Appeals is at full strength, it has 29 active judges, almost double the number in the old Fifth Circuit. The Ninth Circuit is thus the paradigm of the “jumboism” that the Fifth Circuit judges warned against. But have intracircuit conflicts multiplied? Is the law of the circuit in disarray? This article addresses the question whether the Ninth Circuit Court of Appeals has succeeded in maintaining consistency in the law of the circuit. The article is divided into four sections. Section I sets forth the underlying premises and describes the procedures adopted by the Ninth Circuit to avoid inconsistencies among panel decisions. Section II develops a theory of intracircuit conflict through extrapolation from basic principles of precedent in a common law system. Section III reports the results of an empirical study applying the theory to measure the extent of inconsistency in the Ninth Circuit. Section IV considers the implications of the study for the workability of the large circuit and poses questions for further research. The theory consists of three sequential propositions, each of which addresses one of the possible relationships between a particular new decision and existing law in the circuit. First, if losing counsel cannot point to relevant circuit precedents that reach results different from the panel’s result in the case being considered, there is no possibility of conflict or uncertainty of the kind that arouses legitimate concern among judges and lawyers. Second, the cases that offer the greatest potential for conflict are those in which the panel distinguishes a circuit precedent that losing counsel has reasonably relied on as requiring (not simply supporting) a different result. Third, to the extent that the distinctions drawn by the later panel are clear and cogent, the potential for disarray is not likely to be realized. To estimate the extent of inconsistency in the Ninth Circuit, I applied this theory to two random samples of published opinions of the Ninth Circuit Court of Appeals, one from 1983, the other from 1986. The sample encompassed more than 20% of the court’s precedential output in the two years. For each case that was not excluded at the first step of the analysis, I attempted to trace the subsequent history of the common legal issue to determine if the coexistence of arguably inconsistent decisions had created confusion or uncertainty. Indicia of confusion were apparent inconsistencies in the later decisions, disagreement within panels, disagreement between appellate panels and district courts, and frequent litigation. If one or more of these indicia were present, I examined the cases further to determine the extent to which the confusion was produced by the coexistence of the apparently conflicting decisions rather than by other factors. I was particularly interested in searching for conflicts created by a panel's outright failure to mention relevant precedents that reached contrary results. The research was completed in 1988. Based on this research, it does not appear that intracircuit inconsistency is as much of a problem as many lawyers think. Head-on conflicts are quite rare. Disarray caused by the existence of a large number of precedents pointing in different directions is more common, but seldom involves issues that directly affect primary activity.

More than a decade before he was appointed to the Supreme Court, Felix Frankfurter summarized the conditions that he believed "indispensable to a seasoned, collective judgment" by that tribunal. Four of these are of particular importance: 1. Encouragement of oral argument; discouragement of oratory. The Socratic method is applied; questioning, in which the whole Court freely engage, clarifies the minds of the Justices as to the issues and guides the course of argument through real difficulties. 2. Consideration of every matter, be it an important case or merely a minor motion, by every Justice before conference, and action at fixed, frequent, and long conferences of the Court. This assures responsible deliberation and decision by the whole Court. 3. Assignment by the Chief Justice of cases for opinion writing to the different Justices after discussion and vote at conference. Flexible use is thus made of the talents and energies of the Justices, and the writer of the opinion enters upon the task not only with the knowledge of the conclusions of his associates, but with the benefit of their suggestions made at the conference. 4. Distribution of draft opinions in print, for consideration of them by

During the last decade, the Supreme Court has been deciding 65 to 70 cases a Term after oral argument. That represents a sharp decline from the 1970s and 1980s, the era of the Burger Court, when the Court was deciding about 150 cases a Term. The Burger Court’s docket, in turn, reflected a shift from the 1960s, when the docket was smaller. In short, what is “normal” for the plenary docket varies from one era to another. The period of the Burger Court retains a special interest in that regard because that was the only period after World War II in which the plenary docket reached the 150-case level. This article provides a detailed examination of the plenary docket during first six Terms of the full Burger Court, with comparisons to the two preceding six-Term periods. It thus includes ten Terms under the leadership of Chief Justice Warren. The analysis is structured by reference to the four major functions that the Court performs in the life and law of America: delineating the limits of governmental authority against claims of individual liberty (civil rights); defining the boundaries between state and national power and among the branches of the national government (federalism and separation of powers); interpreting and applying the body of statutes and regulations through which the national government exercises its sovereign powers and regulates activity in the private sector (general federal law); and supervising the operation of the federal courts (jurisdiction and procedure). The study found that the expansion of the plenary docket could be attributed largely to growth in civil rights cases. During the first twelve Terms of the study, the annual total of civil rights cases was gradually increasing, while the size of the plenary docket remained essentially the same. The result was that, as the Court moved through the 1960s, civil rights cases were steadily displacing litigation involving other issues of federal law. By the 1970 Term, all cases primarily involving issues other than civil rights occupied little more than one-third of the plenary docket; their number had been reduced by about forty percent from what it had been a decade earlier. Then, in 1971, the Court suddenly expanded the size of the plenary docket. The consequences were twofold. First, there would be no further displacement of cases that did not involve civil rights issues. Second, as long as the Justices were willing to accept the heavier workload, they could continue to hear a large number of cases that did involve civil rights. It thus appears that the Court expanded the size of the plenary docket in 1971 because the Justices were unwilling to cut back on the number of civil rights cases they were deciding, but recognized that there was a need for a greater number of authoritative precedents in other areas of federal law. The only way they could satisfy both desires was to increase the total number of cases given plenary consideration.

At recent hearings on proposals to restructure the Ninth Circuit Court of Appeals, two witnesses (a Ninth Circuit judge and a law professor) expressed some support for a realignment that would divide the state of California between two judicial circuits. This article explores the legal problems that might arise if such a reorganization were to be enacted, and it considers how those problems might be dealt with. It concludes that the problems are far from intractable and that they can be addressed through use or adaptation of familiar mechanisms for avoiding or resolving conflicts between decisions of different courts. Almost 50 years ago, the Commission on Revision of the Federal Court Appellate System (Hruska Commission) recommended that the two largest federal judicial circuits of that era – the Fifth and the Ninth – should each be divided into two new circuits. The Fifth Circuit was then composed of six states, three on the east and three on the west, with approximately equal caseloads. The Commission recommended a three-three split, and six years later Congress enacted legislation implementing the proposed realignment. The Ninth Circuit was a tougher nut to crack. The Ninth Circuit is composed of nine states, but one of those states is California. Then as now, California accounted for about two-thirds of the Ninth Circuit’s caseload. And the Commission was adamant in rejecting the idea of putting California in a circuit by itself. “The creation of … a one-state circuit,” the Commission said, “invites the loss of important elements of our federalism.” Among other things, “[t]here is reason to believe that judges from different states reinforce one another’s perceptions that they are judges of a national court.” Instead, the Commission recommended that Congress divide the Ninth Circuit into two new circuits with the boundaries drawn so that two of the federal judicial districts of California would be placed in a southern circuit (with Arizona and Nevada), and two in a northern circuit (with Hawaii and the northwestern states). Although a bill was introduced in Congress to implement the recommendation, it did not go far. Since then, attention has focused on other proposed realignments, in large part because the idea of dividing California between circuits has generated intense opposition from California’s legal and political community. The idea has not disappeared from view, however, and for good reason. The Hruska Commission was on sound ground in arguing that a one-state circuit would be at a severe disadvantage in performing the functions of a national appellate court. Indeed, a later study group, the Commission on Structural Alternatives for the Federal Courts of Appeals (White Commission) went further, concluding that three states are “the minimum necessary for units of the intermediate tier of a federal system to serve an appropriate federalizing function.” No realignment that satisfies the White Commission’s criterion would come as close as the Hruska Commission’s proposal to an equal division of caseload between the two new circuits. The question, then, is whether the concerns underlying opposition to the Hruska Commission proposal justify excluding it from consideration. The detailed analysis in this article, based on a memorandum written for the Commission, suggests that the answer is “No.” Two classes of cases have loomed large in discussions of the Commission proposal: diversity cases and cases involving challenges to the validity of a California state statute or regulation. With respect to the first, the article finds that notwithstanding the attention paid to diversity cases, dividing California between circuits would have little effect on the litigation of state-law claims. Suits challenging the validity of state laws do present potential problems, but several mechanisms are available to avoid or resolve conflicts between decisions of the northern and southern circuits. These include transfers under 28 U.S.C. § 1404(a) or § 1407 (the multi-district litigation statute) and review by a limited en banc panel similar to the one used by the Ninth Circuit today, but drawn from judges from both of the new circuits. A recurring theme in the analysis is that none of the conflicts likely to arise in the divided-state situation are unique. A judicial system that can handle the delicate problems raised by federal injunctions against state-court proceedings and the logistical problems raised by multitudinous suits in related antitrust cases should be equal to the task of preserving harmony between two federal appellate courts sitting within one state.