Keynoter? What a daunting assignment before this gathering! I’m reminded of President John F. Kennedy’s remark at a dinner honoring Nobel Prize winners: “This is the most extraordinary collection of talent . . . that has ever been gathered together at the White House, with the possible exception of when Thomas Jefferson dined alone.”[2] As I survey this room, I see the greatest collection of experts on federal courts ever assembled. No exception.

I have no illusions that I can add anything to your knowledge. But possibly, I can spur some additional thinking by posing an overarching question. That question is: What are we worried about? Several possibilities come to mind. The federal court system is too slow; it’s too costly; trials are disappearing; inter-circuit conflicts are not being resolved; an unacceptably large number of people with legal grievances have no access to legal representation; the Supreme Court decides too few cases; too many door-closing techniques are keeping too many meritorious claims out of federal courts; the likely growth of federal court caseloads risks a variety of adverse consequences, not the least of which is the loss of the distinctive nature of the federal court system. Many of you would likely select several of these answers. Some might select all of them. Other causes for concern will no doubt be identified during this event.

Of course, concern about court systems is nothing new. It is worth recalling that nearly a hundred years ago, Learned Hand noted that people were complaining about the inefficiency of the legal process in the third millennium before Christ.[3]

My primary concern about the federal court system—indeed, about all court systems—is excessive delay and cost. By expanding opportunities to litigate a case with thoroughness to achieve fairness,[4] we have unintentionally created a cumbersome process where cases languish before trial and subsequently crawl up the appellate ladder. The delays and attendant escalating costs drive many out of the federal court system and into arbitration or abandonment of claims, leaving an unacceptably high proportion of the population without opportunity to obtain redress of legitimate grievances.

Here is one set of numbers to illustrate delay. In 1995, when the United States Judicial Conference issued the Long Range Plan for the Federal Courts (“Long Range Plan”),[5] the median interval from filing to disposition for civil jury trial cases was eighteen months.[6] In 2018, it was twenty-six months,[7] an increase of 44 percent. Of course, the median numbers for disposition time obscure the large number of cases when the interval was measured in years. In 1995, the number of civil cases pending in the district courts for three years or more was 13,538.[8] By 2018, that figure had risen to 73,938,[9] a five-fold increase.

Was this increase in disposition time due to an increase in the number of civil trials? No. The number of civil trials in 1995 was 10,395;[10] the number in 2018 was 3,193.[11] That’s a decrease of 62 percent. These numbers are only one example from many that demonstrate undue delay.

I should acknowledge, however, that the case for proving undue costs is more debatable. A distinguished authority on the federal court system, Professor Arthur Miller, recently wrote in an important article in the Cardozo Law Review that “time . . . has cast doubt on the veracity”[12] of what he calls “the cost-and-delay-narrative.”[13] Writing in the same issue, Professor Alexander Reinert also disputed the cost aspect of the narrative, pointing out that “for decades, researchers, including those at the Federal Judicial Center, have provided data that undermine the narrative only to see overheated rhetoric about runaway costs take over and motivate reforms.”[14] I do not have solid empirical data on costs to share with you, and I admit that anecdotal evidence and impressions are dubious bases for a conclusion. But for whatever it’s worth, my impression is that our cumbersome federal court system imposes undue costs to some extent. If I may be permitted one anecdote: I did think costs were excessive when one side during protracted discovery in an antitrust case asked me for a protective order only after a senior corporate officer had been deposed for the twenty-eighth time.

You will have two days to discuss what aspects of the federal court system we should be worried about and what improvements you think ought to be made. In these remarks, I want to touch briefly on some structural issues, some jurisdictional issues, and some procedural issues and then use a keynoter’s prerogative to offer a few personal views.

Turning first to structure, I will consider conflicting decisions among circuits. I suspect that some will respond by renewing the Hruska Commission’s proposal to establish a National Court of Appeals as an additional layer of adjudication between the thirteen courts of appeals and the Supreme Court. Such a tribunal could be structured to resolve inter-circuit conflicts on application of a losing litigant; or on referral from a court of appeals; or on certification from a court of appeals, which could be declined, on referral from the Supreme Court; or on some combination of these possibilities.

However, I doubt that a fourth layer of adjudication is desirable. The Supreme Court is fully capable of increasing its caseload to resolve major inter-circuit conflicts. Last Term’s total of seventy-six cases decided on the merits[15] is surely not the limit of a court that only recently was deciding 150 cases each year.[16] A few more cases could be reviewed with fewer pages and fewer footnotes per opinion with significant benefit to both the coherence and the clarity of federal law.[17] Since many inter-circuit splits arise from differing interpretations of federal statutes, Congress could help by legislating the interpretation it prefers.

If a fourth layer of adjudication is created, I hope it will not be a permanent court. I would much prefer an ad hoc panel of perhaps seven judges, each drawn by lot from one of seven circuits, which would also be selected by lot. My fallback method of selection would be appointment by the Chief Justice or the Executive Committee of the Judicial Conference. If service for only one individual case is thought undesirable, the judges selected could serve short terms of perhaps three years. The FISA Court offers a familiar model.[18]

The vice of a permanent panel is that its existence could too easily lead to its overuse. Routing a case through four layers of adjudication would be a cumbersome process, which ought to be used sparingly.

Building upon Dean Roscoe Pound’s ideas, Professor Paul Carrington advocates a more far-reaching proposal to avoid inter-circuit conflicts. That proposal would abolish circuit lines and place all circuit judges in a unified national court of appeals.[19] The tradeoff here, of course, is an alleged gain in doctrinal coherence against a clear loss in collegiality, which most of us believe promotes not only a pleasant work environment but also, more importantly, the internal resolution of significant disputes about the language of opinions.

Others have suggested, as a structural response to unresolved circuit splits, increasing the use of specialized appellate courts. With the notable exception of the Federal Circuit, with its jurisdiction over patent appeals and an odd collection of other matters,[20] the United States’ tradition has favored generalist courts. There is a tradeoff between two concerns, often asserted but not always documented: the virtue of the expertise of judges appointed to a specialized court, acquired from their specialized law practice or gained while serving on such a court, on the one hand, with a risk that a specialized court will tilt toward special interests, on the other hand. I have no basis for evaluating either concern.

Another structural variant on the idea of specialized courts is to create Article I courts with limited subject matter jurisdiction. In 1977, the Department of Justice suggested an Article I court for all social security claims.[21]

Conflicts among panels within a circuit present another topic that might merit some structural change. One structural response to intra-circuit conflicts might be using more in banc[22] rehearings, as some have suggested. My Second Circuit bias (we have the fewest in bancs of all circuits) inclines me not to favor more in bancs for two reasons. First, the involvement of all the active judges of a circuit (or even the ten selected by lot, plus the Chief Judge, for an in banc in the Ninth Circuit) precipitates considerable delay in final adjudication. Second, an in banc often yields multiple opinions, which reach a result, but leave circuit law in disarray. Some years ago, Judge Pierce Lively wrote, “[s]ince [the Sixth Circuit] grew to 15 judges, our en banc hearings now typically produce five or six opinions; they clarify virtually nothing.”[23] In banc rehearings would be less burdensome to appellate courts and likely yield fewer concurring opinions if they comprised just seven judges. A smaller in banc court would likely hasten decision of the appeal and the preparation of opinions.

Intra-circuit splits could also be reduced by the use of specialized panels within a circuit court. In the Second Circuit, we recently used that technique to adjudicate the hundreds of motions that arrived in the aftermath of the Supreme Court’s decision in Johnson v. United States.[24] That decision ruled void for vagueness the so-called “residual clause” of the Armed Career Criminal Act.[25] We recognized that similar vagueness claims would immediately be raised in four similar contexts and that prisoners within the Second Circuit would bring these claims to us, seeking leave to file in a district court a second petition for a writ of habeas corpus. So we established four special panels, drawn by lot, each to consider all challenges to the same or similar language in four categories of cases.[26] The vagueness claims were similar, but each category of cases presented slight variations as to the materiality of the challenge. The four panels disposed of 217 motions in just a few weeks. I would not favor permanent specialized panels within a circuit, but our experience with Johnson motions indicates that they can be helpful for certain temporary situations.

The 1995 Long Range Plan outlined several possible structural changes: (1) increase the number of people authorized to decide appeals, either traditional circuit judges, adjunct officers such as appellate commissioners, or district judges through an appellate division in the district courts; (2) create a new tier of appellate courts between the district and circuit courts coupled with discretionary review in the circuit courts; (3) permit appellate panels of two judges and/or one appellate judge to decide appeals; (4) realign the circuits so that all are of approximately equal size; (5) limit the right to an appeal in some categories of cases.[27] Interestingly, the Long Range Plan did not propose a new appellate court between the circuit courts and the Supreme Court.

I have no enthusiasm for any of these proposals. I am especially opposed to limiting the right of appeal or making appellate review discretionary. District courts do an outstanding job, but confidence in the overall functioning of the federal court system is enhanced if all their decisions are subject to appellate review so that the small number meriting reversal or modification can be examined. The famous merchandiser John Wanamaker is reported to have said, “[h]alf the money I spend on advertising is wasted; the trouble is I don’t know which half.”[28] Same with appellate review. A few really need review, but we don’t know which ones until we look at all of them. Besides, reviewing all cases, preferably with an opportunity for oral argument, adds legitimacy to the court process, and the truly frivolous appeals can be disposed of by merits panels in about the same time that some courts take to screen cases for oral argument.

A recurring theme in comments about the federal court system concerns the perceived adverse consequences of increased volume. As we think about the future of caseloads, the first point to bear in mind is that our ability to make predictions about the state of the federal judiciary in the years ahead is seriously limited. To take one example, the 1995 Long Range Plan predicted that the number of district court civil filings anticipated in 2020 would be 976,500.[29] With the 2018 figure around 283,000,[30] I don’t think I’m going too far out on a limb to say that the Plan’s 2020 prediction of nearly one million cases will turn out to be wildly inaccurate. Incidentally, the planners also predicted that the anticipated volume of filings would require 2,410 district judges and 1,600 appellate judges.[31] The current number of authorized district court judgeships is 677, and for appellate courts, 179.[32] In fairness to the authors of the Long Range Plan, I should acknowledge that their figures were based on straight line projections from the previous fifty-three years, with minor adjustments.[33]

We have been frequently cautioned by Professor Marc Galanter and others that the claimed explosion of federal court caseloads is far from reality.[34] Although civil case filings in the district courts rose about 400 percent from 1960 to 1986, the increase from 1986 to 2015 was just 9 percent.[35] It is also important to recognize that although caseloads have risen, the true burdens on federal courts have not risen proportionately. A major portion of the increased caseloads are pro se cases readily adjudicated. In 2017, 29 percent of district court filings were brought by pro se litigants.[36] In 2017, pro se appeals were 50 percent of all the appeals filed in the federal appellate courts.[37] Additionally, statistics on filings can be somewhat misleading. The Administrative Office total of district court filings double-counts some cases because of reopenings and transfers, and it includes a number of so-called “recovery” cases—government collection of defaulted student loans, for example—which impose little, if any burden, on district courts.[38]

But caseload volume has risen, and with increases in population, not to mention new grounds for litigation, it will continue to increase. In 1995, 248,335 civil cases were filed in the district courts;[39] in 2018, the number was 282,936.[40] That’s an increase of 11.5 percent. The number of authorized district court judgeships in 1995 was 632; in 2018, the number was 663.[41] That’s an increase of just 5 percent. We have continued a pattern that has existed since the First Judiciary Act: the volume of federal cases increases faster than the number of district judges available to handle them.

Whatever the degree of increase will be, rising caseloads have had, and will continue to have, adverse consequences. First, more cases increase the pressure to adopt more shortcuts to handle the volume. A prime example is the marked decrease in the percent of appeals decided without oral argument. Some circuits are now hearing oral argument in fewer than 40 percent of all appeals. I do not dispute that many appeals can be fairly decided without oral argument. But I believe that oral argument is necessary for the legitimacy of the appellate process. When lawyers send in their briefs and receive a disposition months later, they are entitled to ask, “Who really looked at my briefs? Law clerks, staff attorneys, maybe one of the three judges?” With oral argument, the lawyers get to look us in the eyes, speak to our ears, and grasp our attention. Whenever I ask lawyers in our Circuit whether they prefer oral arguments in all cases with short time allotments or lengthy time for oral arguments in a limited number of appeals, they always prefer argument in all cases.

Second, increased volume impairs the deliberative process. When I came to the Second Circuit forty years ago, each member of the three-judge panel, right after oral argument, would send to the other two a voting memo, stating not just a tentative result but also a reasoned explanation. No one read the other judges’ memos until sending out one’s own. A later discussion was enlightening. That was the deliberative process at its best. Now, under the pressure of volume, the voting memo has all but disappeared, replaced too often by a brief discussion.

A third adverse consequence of increased volume is the growing reliance on supporting staff. A bureaucracy of staff attorneys has emerged. These are talented men and women, but they are not Article III judges nominated by a President and confirmed by a Senate. Writing in 1989, I suggested that if some alteration of federal jurisdiction was not adopted, we would have a “bureaucracy where there are not merely 2 or 3 staff counsel per circuit, but 10, 20, 30, or 40.”[42] The number of staff attorneys in the Second Circuit is now twenty-six.

To the extent that increased volume has adverse consequences, one remedy is to curtail the jurisdiction of the district courts. I recognize that this is a touchy subject. Federal courts are the primary adjudicators of federal law, although exclusive federal jurisdiction is rare and state courts have authority to decide most federal questions. Testa v. Katt[43] still lives. But even if all federal question cases are left in district courts, there is one category of cases that I think should be shifted to state courts—those involving diversity of citizenship, which amounted to 28 percent of district court filings in 2017.[44]

Ever since Erie[45] required diversity cases to be decided according to state law, such cases should generally be decided by the final expositors of state law, state court judges. I say “generally” because I recognize that total abolition of diversity jurisdiction is politically unrealistic, and there are sound reasons for keeping a few diversity cases in a federal forum. When I discuss diversity jurisdiction with lawyers, they offer examples of such cases. An occasional case might require the decision of different issues under different state laws, making a federal judge arguably more appropriate than a state judge. Or a large company might be so unpopular or so popular that a jury drawn from a large federal district might be less prone to hostility or favoritism than one drawn from a small jury division in a state court.

So instead of urging an end to all diversity jurisdiction, I suggest a system of discretionary access, requiring plaintiffs to file in a state court, but giving any party the opportunity to remove to a federal court on a showing, just by affidavit, of good reason. There need be no hearing—and no right of appeal on the removal decision—because no substantive rights are being adjudicated. And, if it is thought that busy district judges might be reluctant to add to their caseloads, the removal request could be made to the relevant court of appeals, just like motions to a court of appeals for leave to file a successive writ of habeas corpus.[46]

I do not suggest an amendment of Article III. Discretionary access for most diversity cases could be legislated by amending 28 U.S.C. § 1332. And I would raise the jurisdictional dollar amount substantially. Isn’t it somewhat odd that federal district courts hear diversity cases involving $75,000, but the Texas state courts decided the dispute between Pennzoil and Texaco involving $11 billion?[47]

I acknowledge that leaving most diversity cases to the state courts would impose some burden on them. But I estimate that leaving 28 percent of the district courts’ civil caseload to the state courts would increase their caseload by only 1 percent. That shift would not alter the basic nature or functioning of the state courts but would instead greatly assist preservation of the functioning and nature of the federal court system. Chief Justice William Rehnquist noted in 1989, “[T]here will always be a question . . . whether the precise allocations of jurisdiction . . . which have served to define the two systems in the past are adequate for the future.”[48]

Incidentally, the idea of discretionary access to district courts for diversity cases did not originate with me. It was first proposed in 1945 by Senator Kenneth McKellar and was endorsed in 1998 by Justice Byron White and Sixth Circuit Judge Gilbert Merritt in their statement of additional views in the report of the Commission on Structural Alternatives for the Federal Courts of Appeals.[49]

Discretionary access to district courts could also be used for some federal question cases that involve only state tort claims, like cases under the Federal Employers’ Liability Act (“FELA”).[50] Judge Alvin Rubin estimated that state courts already handle 90 percent of FELA cases,[51] even though they may be brought in district courts.

Finally, I’ll suggest a few procedural changes that might be considered to lessen delays in the district courts. The first concerns depositions. Anyone who has participated in depositions as a witness, or who has been obliged to read them in the course of an appeal, ought to be appalled by the time wasted by dueling lawyers arguing with each other, interrupted occasionally by a question to the witness. And, of course, one of the lawyers, and sometimes both, are being paid by the hour. Depositions are now conducted in lawyers’ offices without supervision by a judicial officer or anyone else authorized to end the protracted squabbling.[52] The problem is the lack of supervision, and the obvious answer is to provide supervision. The supervising officer could be a magistrate judge, a special master, or a member of a panel of experienced lawyers.

In 1937, when the Advisory Committee on the Rules of Civil Procedure was considering Rule 30, which authorizes depositions, some consideration was given to authorizing the presence of a special master.[53] Some members of the Committee thought such an officer was needed to rule on whether questions were eliciting evidence admissible at trial. There is no indication that a supervising officer was thought needed to constrain the lawyers. The proposal was not adopted.

I’ve used this proposal myself. As a district judge, I was about to select a jury for a civil trial one day when the defendant asked for a continuance to take the deposition of an expert witness whom the plaintiff had just identified. I asked where the witness was, and the lawyer told me he was in the courtroom. “Put him on,” I directed. “You mean here? Now?” the lawyer asked. “Yes,” I said, “here, now.” Examination and cross-examination under my watchful eye took twenty minutes. In either lawyer’s office, it would have lasted at least a day, maybe longer.

Another source of delay in a district court that could easily be curbed is voir dire: the practice of permitting lawyers to question prospective jurors. It goes on for hours, sometimes several days. And it is often highly intrusive probing into the private lives of citizens summoned for jury duty. Instead, as a district judge, I asked all the questions, as many trial judges do. I then asked the lawyers if there was anything else they wanted me to ask. Sometimes they had one or two questions, which I was glad to ask. A jury was usually selected in about half an hour.

One last procedural item. Though not a significant source of delay, the availability of a large number of peremptory challenges in criminal cases often has one highly objectionable consequence: the challenges can lead to the empaneling of all-White juries, occasionally in federal courts but more often in state courts.[54] The Supreme Court’s decision in Batson v. Kentucky[55] permits trial courts to reject some racially motivated peremptory challenges, but the ability of prosecutors to validate a peremptory challenge by articulating some pretextual, non-racial basis for their strikes has rendered Batson of limited utility. This is particularly problematic in state courts, where a prosecutor has a large number of peremptory challenges in death penalty cases.[56] With challenges for cause adequately affording counsel an opportunity to reject prospective jurors whose contact with the parties or the case risks partiality, the number of peremptory challenges should be reduced to one, or at most two, per side.

All sorts of procedural changes could contribute to lessening delays. I leave it to your imagination to propose your favorites.

Finally, as I consider the overall functioning of the federal court system, I have two related concerns of a general nature. First, those of us in any one court within the system do not have adequate information about what our counterparts throughout the country are doing. There is no single volume or website where an appellate judge can learn of other circuits’ internal practices for handling caseloads efficiently. The Second Circuit’s sequential voting system, which substantially reduced our exploding volume of immigration cases, uses a procedural device that I learned by happenstance during a phone call on another topic with a judge on the Eleventh Circuit.[57] All federal appellate judges are doing essentially the same thing: deciding appeals. But there are significant differences in the way we do it, and we need to know what’s out there.

Second, we need the help of the bar and especially the academy. Legal scholars need to take the field of court administration seriously. I suspect that some of your law school colleagues regard court operations as less worthy of their intellectual talents than the latest doctrinal issues in the law of the First Amendment or racial or gender discrimination. But the proper functioning of the federal court system presents its own important issues and deserves your attention. We need your ideas for improvement and the collection of data to know more about what we are doing and how we could do better.

The Duke Law School takes court administration seriously and recently granted tenure to a scholar specializing in that field. With the establishment of the Berkeley Judicial Institute, I expect to see the University of California, Berkeley, School of Law join Duke in raising the visibility of court administration as a field worthy of your talents. I hope today and tomorrow mark the start of a valuable alliance between the academy and the federal judiciary.

In 1995, the Long Range Plan warned, “Unless a distinctive role for the Federal court system is preserved, there is no sound justification for having two parallel justice systems.”[58] How well the federal courts function will determine whether their distinctive role is preserved. I look forward to your contributions—today, tomorrow, and in the days ahead.

DOI: https://doi.org/10.15779/Z38GB1XH71.

Copyright © 2020 Jon O. Newman, Senior Judge, United States Court of Appeals for the Second Circuit. This article is adapted from Judge Newman’s keynote address at the Berkeley Judicial Institute/California Law Review Symposium on “Charting a Path for Federal Judiciary Reform,” held on April 12, 2019, at the University of California, Berkeley, School of Law.

2. John F. Kennedy, Remarks at a Dinner Honoring Nobel Prize Winners of the Western Hemisphere, The Am. Presidency Project (Apr. 29, 1962), https://www.presidency.ucsb.edu/documents/remarks-dinner-honoring-nobel-prize-winners-the-western-hemisphere [https://perma.cc/RU68-VPH8]. ↑

3. See Learned Hand, The Deficiencies of Trials to Reach the Heart of the Matter, 3 Lectures on Legal Topics 87, 89 (1921–1922). ↑

4. See Jon O. Newman, Rethinking Fairness: Perspectives on the Litigation Process, 94 Yale L.J. 1643 (1985) (Cardozo Lecture), reprinted in 40 Rec. Ass’n B. N.Y. 12 (1985). ↑

5. Judicial Conf. of the U.S., Long Range Plan for the Federal Courts (1995) [hereinafter Long Range Plan]. ↑

6. Admin. Office of the U.S. Courts, United States District Courts—National Judicial Caseload Profile 167 (1997), http://www.uscourts.gov/sites/default/files/statistics_import_dir/District_FCMS_Sep_1997.pdf/ [https://perma.cc/ENH7-ND7U]. ↑

7. See Admin. Office of the U.S. Courts, United States District Courts—National Judicial Caseload Profile (2018), https://www.uscourts.gov/sites/default/files/data_tables/fcms_na_distprofile0331.2018.pdf/ [https://perma.cc/AE48-HHNG]. ↑

8. See Admin. Office of the U.S. Courts (1997), supra note 5. ↑

9. Admin. Office of the U.S. Courts (2018), supra note 6. ↑

10. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts (2017), https://www.uscourts.gov/sites/default/files/data_tables/jff_6.4_0930.2017.pdf/ [https://perma.cc/PL6M-ME9U]. ↑

11. Id. ↑

12. Arthur R. Miller, Widening the Lens: Refocusing the Litigation Cost-and-Delay Narrative, 40 Cardozo L. Rev. 57, 62 (2018). ↑

13. Id. at 61. ↑

14. Alexander A. Reinert, The Narrative of Costs, the Cost of Narrative, 40 Cardozo L. Rev. 121, 128 (2018); see also Elizabeth G. Thornburg, Giving the “Halves” A Little More: Considering the 1998 Discovery Proposals, 52 S.M.U. L. Rev. 229, 246–48 (1999) (summarizing numerous studies discounting prevalence of high discovery costs). ↑

15. U.S. Supreme Court, Opinions of the Court – 2017, https://www.supremecourt.gov/opinions/slipopinion/17 [https://perma.cc/A92T-XMH2]. ↑

16. Ryan J. Owens & David A. Simon, Explaining the Supreme Court’s Shrinking Docket, 53 Wm. & Mary L. Rev. 1219, 1229 (2012). ↑

17. Justice Byron White reported that he had dissented from the denial of a petition for certiorari in about two hundred cases where he thought there was a circuit conflict. Byron R. White, Enlarging the Capacity of the Supreme Court, in The Federal Appellate Judiciary in the 21st Century 145, 146 (Cynthia Harrison & Russell R. Wheeler eds., 1989) [hereinafter FJC Report]. ↑

18. See 50 U.S.C. § 1803 (2018) (stating that the FISA Court consists of seven appointed district court judges who serve for a maximum of seven years). ↑

19. See Paul D. Carrington, U.S. Courts of Appeals and U.S. District Courts: Relationships in the Future, in FJC Report, supra note 16, at 71, 83–85, Appx. A, 227–30. ↑

20. See, e.g., 28 U.S.C. § 1295(a)(8) (2018) (jurisdiction over cases arising under the Plant Variety Protection Act, 7 U.S.C. § 2461 (2018)). ↑

21. Dep’t of Justice Comm. on Revision of the Fed. Judicial Sys., The Needs of the Federal Courts 9 (January 1977). ↑

22. I prefer the spelling “in banc” to the faux Old French “en banc.” The relevant statute uses “in banc,” 28 U.S.C. § 46(c), although the Federal Rules of Appellate Procedure uses “en banc.” See Fed. R. App. P. 35; Jon O. Newman, In Banc Practice in the Second Circuit, 1984-1988, 55 Brook. L. Rev. 355, 355 n.1 (1989). ↑

23. Pierce Lively, A Long-Range View, in FJC Report, supra note 16, at 153, 154. ↑

24. 135 S. Ct. 2551, 2563 (2015). ↑

25. 18 U.S.C. § 924(e)(2)(B) (2018). ↑

26. The categories were claims under 18 U.S.C. §§ 924(e)(2)(B), 924(c), and the Sentencing Guidelines, and a group of miscellaneous cases. ↑

27. Long Range Plan, supra note 4, at 131, 132–33. ↑

28. Adam F. Simon, Mass Informed Consent : Evidence on Upgrading Democracy with Polls and New Media 160 (2011) (quoting John Wanamaker). ↑

29. Long Range Plan, supra note 4, at 18 tbl.7. ↑

30. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the U.S. Courts tbl.C-1 (2018), https://www.uscourts.gov/sites/default/files/data_tables/jb_c1_0930.2018.pdf [https://perma.cc/X4AQ-97WH]. The exact number was 282,936. Id. ↑

31. Long Range Plan, supra note 4, at 18 tbl.7. In 1981, Judge J. Clifford Wallace predicted that by 2000 the number of federal appellate court filings would be 75,000 and that we would have 289 federal appellate judges. See J. Clifford Wallace, Working Paper—Future of the Judiciary, 94 F.R.D. 215, 228 (1981). Eighteen years after 2000, in 2018, federal appellate filings were 49,276 and there were 247 federal appellate judges. See Admin. Office of the U.S. Courts, Report of the Director: Judicial Business of the United States Courts tbl.B (2018), https://www.uscourts.gov/sites/default/files/data_tables/jb_b_0930.2018.pdf [https://perma.cc/9BCC-VFHP]; Admin. Office of the U.S. Courts, Judicial Facts and Figures tbl.1.1 (2017), https://www.uscourts.gov/sites/default/files/data_tables/jff_1.1_0930.2017.pdf [https://perma.cc/7JHN-WFZ3]. ↑

32. Admin. Office of the U.S. Courts, Judicial Facts and Figures tbl1.1 (2017), supra note 30. ↑

33. See Long Range Plan, supra note 4, at 18. ↑

34. See Marc Galanter, An Oil Strike in Hell: Contemporary Legends About the Civil Justice System, 40 Ariz. L. Rev. 717, 734–737 (1998); Marc Galanter, The Day After the Litigation Explosion, 46 Md. L. Rev. 3, 5–6 (1986). ↑

35. See Patricia W. Hatamyar Moore, The Civil Caseload of the Federal District Courts, 2015 U. Ill. L. Rev. 1177, 1182–83 (2015). ↑

36. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts tbl.C-13 (2017), https://www.uscourts.gov/sites/default/files/data_tables/jb_c13_0930.2017.pdf [https://perma.cc/7UV6-4EQT]. ↑

37. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts tbl.B-9 (2017), https://www.uscourts.gov/sites/default/files/data_tables/jb_b9_0930.2017.pdf [https://perma.cc/WCV2-RT9D]. ↑

38. See Moore, supra note 34, at 1184–85. ↑

39. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts tbl.4, https://www.uscourts.gov/sites/default/files/statistics_import_dir/Table408_4.pdf [https://perma.cc/9QEJ-EV2C]. ↑

40. Admin. Office of the U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts tbl.C-1 (2018), https://www.uscourts.gov/sites/default/files/data_tables/jb_c1_0930.2018.pdf [https://perma.cc/X4AQ-97WH]. ↑

41. Admin. Office of the U.S. Courts, Authorized Judgeships (2019) uscourts.gov/sites/default/files/allauth.pdf [https://perma.cc/XB8G-EJXB]. ↑

42. Jon O. Newman, Discretionary Access to the Federal Courts, in FJC Report, supra note 16, at 67. ↑

43. 330 U.S. 386 (1947). ↑

44. Admin. Office of U.S. Courts, Annual Report of the Director: Judicial Business of the United States Courts tbl.C-2 (2017) (75,552 of 267,769 cases), https://www.uscourts.gov/sites/default/files/data_tables/jb_c2_0930.2017.pdf [https://perma.cc/43AN-37PP]. ↑

45. Erie R.R. v. Tompkins, 304 U.S. 64 (1938). ↑

46. See 28 U.S.C. § 2244(b)(3)(A) (2018). (“Before a second or successive application permitted by this section is filed in the district court, the applicant shall move in the appropriate court of appeals for an order authorizing the district court to consider the application.”). ↑

47. See Pennzoil Co. v. Texaco, Inc., 481 U.S. 1 (1987) (rejecting federal court challenge to state court litigation). ↑

48. William H. Rehnquist, Introduction, in FJC Report, supra note 16, at 11, 13. ↑

49. See Comm. on Structural Alt. for the Fed. Court of Appeals, Final Report 77, 78–79 (1998), https://library.unt.edu/gpo/csafca/final/appstruc.pdf [https://perma.cc/8HR4-UFEG]; see also David L. Shapiro, Jurisdiction and Discretion, 60 N.Y.U. L. Rev. 543 (1985) (endorsing the necessity of judicial discretion within the concept of jurisdiction, even in the absence of statutes authorizing such discretion). ↑

50. 45 U.S.C. § 51 et seq. (2018) ↑

51. Alvin B. Rubin, Reallocation: A Two-Way Street, in FJC Report, supra note 16, at 124. ↑

52. Although Rule 28 of the Federal Rules of Civil Procedure calls for depositions before an “officer,” that officer is merely a person authorized to administer oaths, see Fed. R. Civ. P. 28(a)(1)(A), and the officer’s duties are only to administer oaths and record testimony, see id. 30(b)(5)(A), (c)(1). ↑

53. See generally Ezra Siller, The Origins of the Oral Deposition in the Federal Rules: Who’s in Charge?, 10 Seton Hall Cir. Rev. 43 (2013) (providing a comprehensive history of the development of Rule 30). ↑

54. I always capitalize “White” and “Black” when referring to Caucasians or African-Americans. These words are proxies for racial groups, not just identifications of skin color. I explained my practice in an early district court opinion. See Moss v. Stamford Bd. of Educ., 350 F. Supp. 879, 880 n.2 (D. Conn. 1972). ↑

55. 476 U.S. 79 (1986). ↑

56. See, e.g., Ala. R. Crim. P. 18.4(f)(1)(i) (2012) (thirty-six challenges for each side where penalty is death); Miss. R. Crim. P. 18.3(c)(1)(A)(i) (2017) (twelve challenges for each side where punishment may be death or life imprisonment); N.C. Gen. Stat. Ann. § 15A-2117(a) (West 1977) (prosecutor allowed fourteen challenges for each defendant in death penalty case); Tex. Code Crim. Proc. Ann. art. 35-15(a) (West 1991) (fifteen challenges for each side where penalty is death). ↑

57. See Jon O. Newman, The Second Circuit’s Expedited Adjudication of Asylum Cases: A Case Study of a Judicial Response to an Unprecedented Problem of Caseload Management, 74 Brook. L. Rev. 429, 434–35 (2009). The number of pending agency cases, most of which were petitions to review decisions of the Board of Immigration Appeals denying asylum claims, declined from 5,299 on September 30, 2005, to 962 on September 30, 2019. ↑

58. Long Range Plan, supra note 4, at 21. ↑