Statement of Jon O. Newman,
United States Circuit Judge,
U.S. Court of Appeals for the Second Circuit,
before the Commission on Structural Alternatives
for the Federal Courts of Appeals
U.S. Courthouse, Foley Square, New York, NY
April 24, 1998

Justice White, and members of the Commission -- I appreciate your invitation to appear here today to discuss issues concerning the federal appellate courts. The views I express are entirely my own, and in no sense do I appear for the Court on which I serve. I realize that your immediate focus is the Ninth Circuit and that your mandate also includes consideration of various structural alternatives that might merit consideration as the federal courts of appeals endeavor to handle far more volume than we ever experienced in the past and especially the likely volume we will receive in the future if current trends in appellate filings continue. But with all respect, my basic point to you today is that it is entirely premature to consider structural alternatives until substantial efforts have been made to moderate the volume of appellate caseloads.

Having served as a federal judge for 26 years, as an appellate judge for 18 years, and as chief judge of my circuit for 4 years ending last July, I have become increasingly convinced that unless the caseload volume of the appellate courts -- indeed, of all federal courts -- is moderated, the federal court system will soon lose the distinctive characteristics that have justified our existence for two centuries. It might be possible for district courts to handle increased volume by simply adding judges, without suffering any fundamental loss of essential characteristics. District judges enjoy the luxury of operating in relative isolation, and their roles are not significantly changed by the addition of increased numbers of colleagues down the hall handling additional volume.

The situation is entirely different in appellate courts. We are group courts, transacting our business primarily in panels of three and responsible for maintaining the coherence of law within our circuits. Those two salient characteristics place realistic limits on the size of any one appellate court. For judges to work effectively in panels of three, month after month, year after year, they must sit with each other with some frequency. We use the word collegiality, but that term does not begin to capture the subtle elements of respect, trust, cooperation, and accommodation that characterize members of a group court at work. At some point, if volume continues to grow and more and more judgeships are created, a court of appeals expands beyond the size at which its judges can work effectively. And for the court of appeals as an entity, undue size poses an entirely unacceptable threat to maintaining the coherence of circuit law.

Structural alternatives are not being considered because they have intrinsic worth. They are being proposed as one way of enabling the federal appellate courts to handle their growing volume. The reality is, however, that no matter what structural alternatives are implemented, a restructured appellate court system endeavoring to handle unchecked volume will not be able to maintain the distinctive features of federal courts. These features are quality of appointments, individualized attention to cases, judicial craftsmanship, and a relative absence of bureaucracy.

All of these values are threatened by permitting the size of appellate courts to expand. We have already seen the price being paid as volumes rise and the number of judges grows, usually a decade or more after the need has arisen. Oral argument, once a characteristic of all federal appeals, is available in some circuits in less than half of the cases. Our circuit is the last one in the nation that still offers oral argument to parties, including unincarcerated pro se litigants. Published opinions decline to as low as 12 percent of all decisions. And central staff increases with ever-growing roles in the processing, screening, and preliminary decision of cases. These consequences of growth have already begun to erode the distinctiveness of the federal appellate court system. If volume is unchecked, the consequences for the future are ominous.

There is no ideal solution to the problem of growing caseload volume. What is required is a sensible use of all of the court resources of our country, both state and federal. It is not a matter of natural law that there should be a dual court system in the United States. Today, we allocate about 97 percent of all trials and appeals to the state courts. That basic fact should make it clear that no matter what we do about federal jurisdiction, the inevitable growth in caseloads resulting from growth in population will primarily be felt in the state courts. These courts are already high volume courts, and they will be obliged to handle even more volume. But it is not inevitable that the federal courts will be obliged to handle increased volume. Whether such volume is channeled to federal courts depends entirely on the decisions to be made by Congress, advised by expert commissions such as this one, as to the proper allocation of jurisdiction between state and federal courts.

It is my basic view that some reallocation of jurisdiction to the state courts is warranted because it will not alter their essential nature, but it will save the essential nature of the federal courts. And it will do so far more effectively than any of the proposals I have seen for structural alteration of the federal appellate courts.

I realize that it is hardly novel to suggest the reallocation of some cases from federal to state courts. What might be of interest to this Commission and ultimately to Congress, however, is a way of accomplishing such reallocation that has not been previously considered. Up until now, suggestions for reallocation have involved the placement into state courts of all cases within designated categories of current federal jurisdiction. The prime example of such a category is the diversity jurisdiction, especially "in-state plaintiff" diversity jurisdiction.

The premise of my rather different approach is that within every category of cases that has been proposed for reallocation, there are some individual cases with a legitimate claim to a federal forum. The suggestion I make, based on this premise, is a system of discretionary access to federal courts.

What I envisage is, first, a determination by Congress that certain categories of federal jurisdiction cases include a large number of cases that can appropriately be handled in state courts, and, second, the authorization of federal judges to entertain petitions for particular cases within the designated categories to be litigated in federal courts. Diversity cases are an obvious category for discretionary access, but the technique would also be applicable to some designated area of federal question cases.

There are various alternative procedural devices for implementing a system of discretionary access, but the one that seems most promising is to permit the case to be filed initially in state court, subject to a removal petition if either party can demonstrate grounds for the need for a federal forum for that particular case. Some in-state plaintiff diversity cases have a substantial claim to a federal forum, but there is no need to place all such cases in federal courts just because a few belong there. Discretionary access can provide a politically palatable way of reallocating jurisdiction between state and federal courts so that federal caseloads can be moderated, and the essential attributes of the federal courts can be preserved.

Any reallocation of jurisdiction to the state courts should be accompanied by some resources to help the states undertake additional burdens, but the federal budget impact will be far less if future caseload growth is handled by helping the states handle reallocated cases than by constantly expanding the size of the federal judiciary.

Let me now consider briefly your more immediate concerns. On the issue of the Ninth Circuit, it is my view that the decision whether to split that circuit should be largely determined by the judges of that court. If I were serving on a court of 28 judges, with a pending request for ten additional judges, I would probably favor a split. But resident judges know what works in their courts, and I would not wish to intrude my preferences into their matters.

Turning to some of the principal suggestions for structural changes, I think the most useful and least injurious would be experimentation with two-judge appellate panels. Such panels would enable the existing complement of appellate judges to handle increased volume without the immediate need for more judges. Two qualifications should be noted. First, reducing panel size from three to two for all or at least most appeals, thereby creating 50 percent more panels, would not necessarily result in the ability of these panels to handle 50 percent more volume. The reason is that the writing assignment is shared three ways on three-judge panels but only two-ways on two-judge panels. Thus, if the expected 50 percent increase in dispositions were to be achieved, each judge would have to write 50 percent more opinions. Conversely, if each judge writes only the same number of opinions as the judge wrote when serving on a three-judge panel, the reassignment to two-judge panels would achieve no increase in dispositions at all. If two-judge panels were used, I think it is realistic to expect that the number of dispositions would rise because the writing assignments would increase somewhat, but the increase in dispositions would be somewhat less than 50 percent.

The second qualification concerns the procedure to be used in the event the two-judge panel is evenly divided. Many suggest that the tie should be broken by adding a third judge to the panel. I think it would be preferable to refer such cases to a mini-in banc court of five judges. Since the dissent rate in federal appellate courts is extremely low, my guess is that the need for such rehearing panels would rarely arise.

It might also be appropriate to experiment with even one-judge appellate panels. I think we could entrust every judge of the federal appellate courts to make the decision whether an appeal is so lacking in merit that a single judge could affirm, or whether it presents issues of sufficient substance to merit consideration by a three- or two-judge panel.

The other principal proposals I have seen strike me as extremely dubious. I do not favor discretionary appellate jurisdiction for two reasons. First, every litigant should have the right to one level of appeal, even if before a panel of only two judges, or even of one judge. Second, a certiorari practice for all federal appeals would turn out to yield at best only a modest increase in the rate of dispositions. The reason is that judges would now have to serve both on certiorari panels and merits panels. Though time savings would result whenever certiorari was denied, time would be added for those cases where certiorari was granted, even if the same judges accepted certiorari and ruled on the merits.

Nor do I favor specialized courts. The great strength of the federal appellate system is that it consists of general courts, manned by judges attracted to the appellate bench because of the range of issues presented. Specialized courts might be appropriate for very narrow categories such as patent appeals and perhaps tax appeals, but extending the concept more broadly risks courts with ideological biases, as appointing authorities succumb to the temptation to select judges primarily for their anticipated voting patterns in the narrow jurisdiction of each specialized court.

The proposal to abolish circuit lines altogether and have all appellate judges function on panels of one national court strikes me as singularly ill-advised. Such a plan fundamentally ignores how appellate judges function in practice. They are effective only to the extent that they can work cooperatively with each other. They are not fungible cogs in some gigantic national appellate machinery.

Nor do I favor the various schemes to combine the circuits into a small number of large circuits. This technique would create on a slightly smaller scale the shortcomings of the proposal for one national court of appeals.

If the number of appellate judges continues to grow, and the only choice is between combining circuits or subdividing them, I would prefer to see them sub-divided. The functioning of appellate courts would be maintained by keeping each circuit to a manageable size, and the slightly increased risk of intercircuit conflicts could be alleviated by an intercircuit tribunal, composed of perhaps seven judges drawn by lot from a panel to which each circuit would designate one of its number. Such a panel would not be a permanent court, but would function only as the need arose.

My fundamental point remains, however. Before any of these structural alternatives are seriously considered, we should be sure that we have fully explored the far preferable solution of moderating federal caseloads by a reallocation of jurisdiction between state and federal courts. The federal court system is worth preserving, but only if it retains its essential characteristics. Tinkering with its structure will accomplish little and will create the illusion of progress, even as growing caseload volume causes the federal system to deteriorate, no matter what structural alternatives are adopted. If you want to preserve the quality of federal justice, especially at the appellate level, the best course is treat the nation's litigation system as an integrated whole, and assign to the federal component of that system only those cases that truly need a federal forum.