VIEWS FROM THE LOWER COURT Alvin B. Rubin* · PDF fileVIEWS FROM THE LOWER COURT Alvin B. Rubin* ... opposition to a Supreme Court nominee who ... "We are not final because we are

VIEWS FROM THE LOWER COURT

Alvin B. Rubin*

Trial judges and appellate judges are not adversaries, butthe trial judge who keeps his eyes on what the appellate courtmight do with his decision is somewhat like the tennis player whowatches his opponent instead of the ball: He's likely to make abad shot. Trial judges ought to reach decisions in accordancewith the law and the evidence; they should not be overly concernedabout whether the appellate court is apt to reverse them. They arenot seeded according to their affirmance rate.1

While appellate judges are not opponents of the trial courts,they can hardly be expected to feel overwhelming enthusiasm forthe trial bench; after all, it is trial judges who foment the mass ofappeals that threatens to engulf the appellate courts. Their errors(or at least alleged errors) are the cause of all the work thatburdens the eleven circuit courts and of all the problems dealt within this Symposium. If only trial judges were more learned, wrotebetter findings of fact, gave more cogent reasons for their decisions,and studied the law more thoroughly, the appellate task wouldbe simplified. No one would wish for prescient and infallible trialjudges, for then there would be no reason for appeals. But itwould ease the appellate burden if trial judges were better thanthey are.

Trial judges are understandably wary of appellate judges.Courts of appeals never reverse a lawyer, only a trial judge. It maybe because the trial judge made an error and will honestly admit itupon reflection. But there may be other reasons. A lawyer mayhave failed to cite an authority, 2 or, despite some lip service tothe contrary, an issue may be raised on appeal that was never

* District Judge, United States District Court for the Eastern District ofLouisiana, and an enthusiastic, albeit inept, tennis player. I wish to express ap-preciation to my law clerk, Harold K. Watson, J.D. 1974, Louisiana State Uni-versity, for his assistance.

1 In one exceptional instance, opposition to a Supreme Court nominee whohad served as a trial judge was based in part on what was said to be an egre-giously high reversal rate.

2 In some instances, authority mentioned for the first time on appeal shouldbe considered. Kurdziel v. Pittsburgh Tube Co., 416 F.2d 882, 886 (6th Cir.1969).

448

1976] VIEWS FROM THE LOWER COURT 44Y

brought to the trial judge's attention,8 or the appellate court mayundertake to raise a new theory of the case on its own.' On someoccasions, the "law of the circuit," despite solemn contrary pro-fessions, really is something different when old judges A, B, and C-all confirmed conservatives of advanced years-compose thecourt, compared to what it is when those new and radical ap-pointees, judges M, N, and 0 compose it. Finally, it has beenasserted, and not exclusively by trial judges, that a reversal by anappellate court does not necessarily mean the trial judge waswrong: "We are not final because we are infallible, but we areinfallible only because we are final."5 Alas, none of us likes tobe reversed when he thinks he was right, and covertly we mustconfess that we are not elated even when we were wrong.6

So perhaps there is some natural antagonism between judgeswho sit on high (or midway) and their (sometimes) learnedbrethren below. It may be ungracious for one of those below tosuggest the possibility of improvement to those who monitor hisproceedings. But having been extended an invitation to do so, Ishall try to comment with due regard for the problems thatappellate judges face. Because my experience is in the federalsystem, I use data from the eleven federal courts of appeals.

The statistics on the appellate workload are staggering.7

Some appellate judges must consider and decide two or three fullcases every week-sometimes more." The bar sometimes does less

8 The rule that an appellate court will not review matters not presented tothe lower court is at times honored more in the breach than in the observance.See, e.g., King v. Laborers Int'l, Local 818, 443 F.2d 273, 279 (6th Cir. 1971).

4 The party who benefits by injection of the new theory for the first timeon appeal is not to be blamed because:

[1]t is not one of the parties seeking to advance a new theory, rather, itis this court, in fulfillment of its duty to apply the correct law, that isseeking to put the case back on the right track....

Appellate review does not consist of supine submission to erroneouslegal concepts even though none of the parties declaimed the applicablelaw below .... Our duty is to enunciate the law on the record facts.Neither the parties nor the trial judge, by agreement or passivity, canforce us to abdicate our appellate responsibility.

Empire Life Ins. Co. of America v. Valdak Corp., 468 F.2d 330, 334 (5th Cir.1972) (citations omitted).

5 Brown v. Allen, 344 U.S. 443, 540 (1953) (Jackson, J., concurring).6 Comments made from time to time by appellate judges about the Su-

preme Court's reversal of their decisions lead to the tentative conclusion thatthis is not merely a trial judge's idiosyncrasy.

7 See, e.g., COMMISSION ON REVISION OF THE FEDERAL COURT APPELLATE

SYSTEM, STRUCTURE AND INTERNAL PROCEDURES: RECOMMENDATIONS FOR CHANGE169 (1975) (Table: Selected data on filings and terminations in the Courts ofAppeals) [hereinafter cited as STRUCTURE AND INTERNAL PROCEDURES]. "Filings[from 1970 to 1975] increased by more than 60 percent, yet not a single judge-ship was added." Id. at 1.

8 Data in the clerk's office of the Fifth Circuit show that each of the 15

active judges was solely responsible for initiating action for final disposition afterbriefing of 125 cases during fiscal year 1975. The average judge wrote 43 signed

m A

450 UCLA LAW REVIEW [Vol. 23: 448

than is desired.9 Dockets mount. 10 Staffs are said to be inade-quate. The problems have been reviewed repeatedly;" they arereal and they are immense. Congress recognized both the problemsand the need to search for solutions when it created the Commis-sion on Revision of the Federal Court System.' 2 Any considera-tion of appellate problems and possible solutions must begin witha thorough reading of the Commission's two reports.'" The firstreport focused on possible circuit boundary realignment, and thesecond report contained a number of suggestions for improvementin appellate procedures. Similar proposals are made elsewhere inthis Symposium. Let us focus on other possible solutions.

opinions, 34 per curian opinions, and initiated the disposition of 48 cases withoutformal opinion. In addition, he participated as a panel member in 250 additionalcases. Even though court records show that a record number of 2,713 cases weredisposed of in fiscal year 1975, dispositions ran substantially behind filings, since3,293 cases were filed this past year. See DIRECTOR OF THE ADMINISTRATIVEOFFICE OF THE UNITED STATES COURTS, ANNUAL REPORT (1974) [hereinaftercited as ANNUAL REPORT].

9 There is an extensive literature on this subject, most of it speeches andarticles by judges. See, e.g., Kaufman, Does the Judge Have a Right to QualifiedCounsel?, 61 A.B.A.J. 569 (1975).

10 At the end of fiscal year 1974 there were 11,470 pending cases, thelargest number in history. In 1968, the total inventory was 6,615, half the 1974number. A review of the data can be found in ANNUAL REPORT, supra note 8,at 179.

11 See, e.g., Cramton, Federal Appellate Justice in 1973, 59 CORNELL L. REV.571 (1974). Dean Cramton's article is an introduction to a collection of paperscontributed to the 1973 Cornell University School of Law Irvine Lecture Series,discussing the quality and structure of federal appellate justice. The lecturerswere Judge Henry J. Friendly, Chief Justice of the United States Court of Appealsfor the Second Circuit; Judge Clement F. Haynsworth, Jr., Chief Judge of theUnited States Court of Appeals for the Fourth Circuit; Professor Maurice Rosen-berg, Columbia University Law School; and Professor Philip B. Kurland, Univer-sity of Chicago Law School.

12 See COMMISSION ON REVISION OF THE FEDERAL COURT APPELLATE SYS-TEM, SENATE COMM. ON THE JUDICIARY, S. REP. No. 930, 92nd Cong., 2d Sess.(1972). The Commission was composed of 16 exceptionally qualified members.Its Executive Director was Leo Levin, a renowned law teacher and a perceptivestudent of judicial processes; he and the staff worked tirelessly.

13 COMMISSION ON REVISION OF THE FEDERAL COURT APPELLATE SYS-TEM, THE GEOGRAPHIC BOUNDARIES OF THE SEVERAL JUDICIAL CIRCUITS: ALTER-NATIVE PROPOSALS (1973); STRUCTURE AND INTERNAL PROCEDURES, supra note 7.The Commission rejected any major circuit realignment, and suggested merelythe division of the two mammoth circuits, the Fifth and the Ninth. Its 22 otherrecommendations are contained in its June 1975 report. Id. at vii-xi. For anoverview of the Commission and the history behind it, see Hruska, The Commis-sion on Revision of the Federal Court Appellate System: A Legislative History,1974 ARiz. ST. L.J. 579. This excellent article goes far beyond a mere historyof the Commission and illustrates some of the arguments that have been made forand against various changes suggested in the past. See also Burdick, FederalCourts of Appeals: Radical Surgery or Conservative Care, 60 Ky. L. REV. 807(1971); Burger, The State of the Federal Judiciary 1970, 56 A.B.A.J. 929, 933(1970); Burger, The State of the Federal Judiciary 1971, 57 A.B.A.J. 855, 859(1971); Lumbard, Current Problems of the Federal Courts of Appeals, 25 N.Y.COUNTY LAw. ASS'N B. BULL. 210 (1967-68).

VIEWS FROM THE LOWER COURT

The first and simplest way to reduce the appellate workload isto decrease the rate of appeal from trial court dispositions. 14

In criminal cases, at present, the rate of appeal is about 90 percent,that is, nine out of ten guilty verdicts are appealed, and the rateis expected soon to reach 100 percent. This is not likely to be re-duced. 15 But lawyers in civil cases usually appeal because theythink there is some chance to change the result. The more cer-tain the rule of the circuit, the more predictable the outcomeof a particular case. Even in criminal cases, while the certainty ofthe applicable rule will not directly reduce appeals, it is likely tolessen the number of appealable decisions. Knowledge that a cer-tain result is preordained tends to lead either to a greater propen-sity on the part of the government to refuse to prosecute, or, wherethe certainty is the other way, to a greater number of voluntarysettlements by defendants-that is, guilty pleas.

Rules are stated and applied in opinions. Any appellate opin-ion worth publishing should not merely give a reasoned disposi-tion of the particular matter; it should, in addition, articulatea standard or a rule that can be applied by lawyers and judges infuture cases.'" The ad hoc, "railway ticket" decision-good onlyfor this day and station-is a betrayal of the appellate function.Yet some appellate judges favor this kind of opinion because itleaves them elbow room to maneuver in future cases. Needless tosay, the more elbow room left, the more counsel will attempt toappeal to the area left vacant."

14 The number of appeals appears to depend on the number of appealable

trial court decisions. Goldman, Federal District Courts and the Appellate Crisis,57 JUDICATURE 211 (1973). The rates of appeal are themselves increasing. Inthe Fifth Circuit, criminal trials in district courts increased 115.5% between 1961and 1972, whereas criminal appeals for the same period increased by 526.9%.In the same period, the number of civil trials increased 155% while civil appealsincreased 362.5%. Some of this can be attributed to the effect of the CriminalJustice Act of 1965, for an indigent is now furnished with paid counsel throughthe appellate process. In civil as well as criminal cases, the cost of an appeal,relative to the total cost of trial, has diminished. There are other causes: Thegreater litigiousness of our society is not the least.

15 If appellate review of sentencing is adopted, the number of appeals islikely to exceed the number of guilty verdicts because sentences in at least someguilty-plea cases will be subject to an appeal. If the sentence can only be re-duced (as most proposals suggest), there is little to be lost by an appeal.

16 K. LLEWELLYN, THE COMMON LAW TRADITION 26 (1960). It wouldbe tiresome to cite this great work repeatedly. It is an indispensable beginningto any study of the appellate process.

17 For a candid example of this type of opinion writing, see Freeman v.

Chevron Oil Co., 517 F.2d 201 (5th Cir. 1975):Cases of the kind presented here are peculiarly dependent for their

resolution upon their factual content. They must therefore be decidedon a case-by-case basis because of the difficulty of formulating a cri-terion applicable to all cases.

Id. at 209.

1976]

UCLA LAW REVIEW

No one will claim that rules alone, however lucid, will set-tle everything. The variations of fact are infinite. The futurecannot be foreseen. But the foundation of our belief in the ruleof law is the conviction that legal rules cannot only be formulated,but can also be stated and applied to govern human decisions.The corollary is that they ought to be stated as comprehensibly aspossible for those who must apply them. When a rule is beingchanged, the opinion ought honestly to say so, and it ought to statethe new rule as lucidly as possible. In these endeavors, hyperboleand metaphor are more frequently a hindrance than a help.' 8

The doctrine known as "the law of the circuit" is that, oncea panel of a circuit formulates a rule of decision in one case, theother judges of the circuit are bound to follow that precedent,right or wrong, in future cases, at least until it is reconsidered bythe entire court sitting en banc.' 9 But a panel of three otherjudges may feel either that the rule is "wrong" or that it shouldnot apply to the case they are considering. So an understandabledesire to decide today's case in accordance with the proclivities ofthe panel now sitting seems to lead to opinions that fail to accordto prior decisions the willing acceptance and wholehearted en-forcement that trial judges are expected to accord appellate de-cisions. Instead, these later opinions seem to distinguish the earlierdecisions on an ad hoc basis.2"

Grudging consideration of earlier decisions by panels in thecircuit leads to several problems. Similar cases are distinguishedon the basis of minute and inconsequential fact variations. The

18 The absence of a citation to an example at this point is intended to beconspicuous. Finger-pointing is impolite.

19 United States v. Olivares-Vega, 495 F.2d 827 (2d Cir. 1974), cert. denied,419 U.S. 1020 (1975); Whatley v. United States, 428 F.2d 806, 807 (5th Cir.1970).

20 Examples of intracircuit conflict are legion. In the Fifth Circuit, onlythe recent decision in Fitzgerald v. Estelle, 505 F.2d 1334 (5th Cir. 1974), re-solved the question whether there is a difference between the standard applied toclaims that a criminal defendant received ineffective assistance from privately re-tained counsel and the standard applied to claims that appointed counsel renderedineffective assistance. Compare Holland v. Henderson, 460 F.2d 978 (5th Cir.1972); Breedlove v. Beto, 404 F.2d 1019 (5th Cir. 1968); Bell v. Alabama, 367F.2d 243 (5th Cir. 1966); and Porter v. United States, 298 F.2d 461 (5th Cir.1962); with Johnson v. Smith, 447 F.2d 985 (5th Cir. 1971); McGriff v. Wain-wright, 431 F.2d 897 (5th Cir. 1970); and Langford v. Alabama, 422 F.2d 760(5th Cir. 1969).

A related problem is the distinguishing of cases on the basis of somethingless than real differences. In the Fifth Circuit, an example of this can be seenin the opinions dealing with the Louisiana Direct Action statute (LA. REv.STAT. ANN. § 22:655 (West 1959)). The Act has been held applicable to acollision between two vessels, but not to a collision involving a vessel and a fixedplatform, an incident equally governed by maritime law. Compare OlympicTowing Corp. v. Nebel Towing Co., 419 F.2d 230 (5th Cir. 1969), with Con-tinental Oil Co. v. London Steam-Ship Owners' Mut. Ins. Ass'n, 417 F.2d 1030(5th Cir. 1969). See 5 J. MARITIME L. & COM. 135 (1973).

452 [Vol. 23: 448

VIEWS FROM THE LOWER COURT

number of appeals increases because, after all, the panel in thisdraw of the lottery might be more liberal or more conservative,more punctilious or less precise than the one that decided the

21earlier case. More trials are also held, further increasing the workof nisi prius courts, because an ability to predict outcomes withreasonable accuracy is one of the major factors enabling lawyersto settle cases. And trial judges may commit what is later viewed aserror because they could not foresee the view that a new panel ofthe appellate court would take.

All appellate courts are aware of the problem. But merelybeing aware of a problem, and even of the solution to it, does notin itself effect a change; most tennis players in my class areaware that we start our backswings too late and that we don'tfollow through-but we continue in the errors of our ways. Howcan greater appellate decisional clarity be achieved? Surely notby a massive series of en banc decisions. But perhaps it can beapproached by more education on what appellate opinions canand should do, by greater efforts on the part of appellate judgesto state and live by a rule of law, by less insistence on personalvirtuosity, and by greater collegiality.

No judge worthy of his office wants merely to dispose ofcases as if he were working on an assembly line. We all seek thejust result. We are all mindful of the adage that no case is decideduntil it is decided right."2 But humility and awareness of the limitsof human wisdom must be coupled with the search for justice.Most of us who have had experience in the law come to doubtthat we can know in every case where right resides.

The rule of law is not the doctrine of perfect decision.2"There must always be stress in the judicial process, between whatEdmond Cahn called the "fundamental classic antitheses of legaltheory," between justice and power, between freedom and order,between security and change.24 The data available indicate that

21 The formula for computing the number of three judge panels that canbe made of any given number of judges is set forth in every standard text on prob-ability theory. See, e.g., W. WEAVER, LADY LUCK: THE THEORY OF PROBABILITY

94 (1963). According to the standard formula, 35 combinations of a seven judgecourt are possible. Nine judges can be combined in 84 different ways; elevenjudges in 165 ways, thirteen judges in 286 ways, and the formidable Fifth canbe put together in 455 combinations without ever using a senior, visiting, or dis-trict judge. The Commission suggests consideration of eighteen judge circuits;that would produce 816 possible combinations-more combinations than the totalnumber of trial and appellate judges in the federal system.

22 Compare the observation of Mr. Justice Brandeis that "in most mattersit is more important that the applicable rule of law be settled than that it besettled right." Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 406 (1932)(Brandeis, J., dissenting).

23 See K. LLEWELLYN, supra note 16, at 24 (discussion of "One Single Right

Answer"). See also M. FLEMING, THE PRICE OF PERFECT JUSTICE (1974).24 E. CAHN, THE SENSE OF INJUSTICE 2 (1949).

1976]

UCLA LAW REVIEW

the process of affirmance and reversal has not accorded due con-sideration to the thought that in many cases a conscientious deci-sion is as much as can be expected, and that there is no ultimate"right" answer. In 1974, an average of 18.6 percent of all casesdecided by the circuit courts reversed trial court decisions. 5 Butthe Second Circuit reversed in only 13.7 percent of its cases, whilethe Seventh Circuit was reversing 20.8 percent of the decisions ap-pealed there. It is difficult to believe that the trial judges in theSeventh Circuit were half again more fallible than those in theSecond.

The data on 1974 criminal reversals point even more clearlyto the notion that different appellate courts may have differentideas concerning when a litigant has been denied a fair trial be-low. The Second Circuit reversed only 7.5 percent of the criminalappeals; the Ninth Circuit reversed 17.3 percent, over twice as higha rate. A trial judge might be forgiven if he rejected the explana-tion that trial judges in the Ninth Circuit commit reversible errormore than twice as often as their colleagues in the Second Circuit.

The inconsistency in the way circuit courts apply principles oflaw leads not only to lack of a uniform standard among the circuits,but also to more appeals. Unless there is a clear law of the circuit,the decisions of some other circuit always make inviting precedentfor appeal. The remedy is not-I stress because of the risk ofbeing misunderstood-for judges to reduce by concert the reversalrate in criminal or civil appeals; it is for appellate courts to adopta standard that is more certain and whose application is more pre-dictable. When lawyers and trial judges are unable to predictwhat the appellate courts are likely to do, they both must guess-and the losing lawyers are likely to appeal. In their study of theSupreme Court's caseload, Professors Gerhard Casper and RichardA. Posner say the likelihood that review will be sought in theSupreme Court "is a function of the probability that review willbe granted ..... - In similar fashion, the invitation to appealis made more attractive by the probability of reversal. Theproblem of dealing with uncertainty is a professional nemesis; asthe Commission said,

For the judge, uncertainty is the lack of a body of precedentsadequate for confident decision; for the practitioner, it is alack of stability sufficient to provide predictability adequatefor effective service to clients and society. 2T

25 To avoid an infinity of footnotes, the source of the data used from hereon is not given. Unless otherwise indicated, it all comes from MANAGEMENT STA-

TISTICS FOR UNITED STATES COURTS (1974), and the ANNUAL REPORT, supranote 8.

26 Casper & Posner, Study of the Supreme Court's Caseload, 3 J. LE.ALSTUDIES 339, 361 (1974).

27 STRucTuRE AND INTERNAL PROCEDURES, supra note 7, at 13.

454 ,[Vol. 23: 448

VIEWS FROM THE LOWER COURT

The size of our circuit courts, the volume of their work, andthe number of judges on each court have together increased thetendency toward individualization of decision and the use ofopinion writing to display virtuosity. Fewer courts circulate opin-ions in advance. There is less effort to prepare an opinion thatreflects a court view. The author states his decision; others mayjoin or write dissents or concurrences.

But laissez-faire is not a sound approach to writing opinionsthat will create a body of national precedent. The judges of anappellate court ought to constitute a court, not congeries ofjudges who are drawn together in odd groups by chance. Thismight mean, if I may touch on a matter most sensitive, that theymeet together more frequently, live closer to each other, and writetheir opinions in places where they can readily exchange views anddrafts. No one would deny any judge his right to play on centercourt and demonstrate his skills; but when the privilege becomesa habit, litigants and trial judges may think that what appears onpaper to be a court is instead an association of judges. They maythink that the result on appeal will depend on who happens tosit on the panel.

Let me turn from the problems created by decisions that yieldinsufficient guidance for the future to a recent development thatat first glance appears to promise consistency-the institutionaliza-tion of appellate disposition. It becomes increasingly hard to feelthat three appellate judges are really thinking about and decidingeach case that is disposed of by the courts of appeals. Appellatejudges in the federal system now have two law clerks each; chiefjudges have three. The judges of some courts have an extra lawclerk; Fifth Circuit judges, for example, have three each.2 8 Someappellate courts have staff law clerks in addition: The FifthCircuit has a total of nine. 29 The judges of the Fifth Circuit,therefore, have a total of 59 law clerks, a platoon indeed, if nota company;"0 other courts have similar, albeit smaller, numbers ofunseen legal assistants."'

28 In the Fifth Circuit, beginning in fiscal 1975, each judge was permitted to

hire an additional secretary, making a total of two, and an additional law clerk,making a total of three. Prior to that time, each judge had the option to employone or the other. There are nine staff law clerks and a legal director whose dutyis described as "to pre-screen cases for final classification by judges on the oralor summary dockets." Interview with Gilbert Ganucheau, Chief Deputy Clerk,United States Court of Appeals for the Fifth Circuit.

29 The Ninth Circuit is also authorized nine staff law clerks.30 This includes four senior circuit judges, three of whom have one law

clerk, and one of whom has two.31 It is difficult to determine exactly how many people are working as law

clerks because some positions classified as clerical may in fact be filled by people

who are doing law clerk work. However, the number of law clerk positions au-

1976]

UCLA LAW REVIEW

What are these able, intelligent, mostly young people doing?Surely not merely running citations in Shepard's and shelving thejudge's law books. They are, in many situations, "para-judges."In some instances, it is to be feared, they are indeed invisiblejudges, for there are appellate judges whose literary style apearsto change annually. Even those who write every word of everyopinion themselves must be depending on these lawyers for some-thing important.

Indeed, who would not? The work is there to be done. Itsits like a glacier before the appellate judge. It advances on himinexorably. The short-term solution is to get more staff, delegatemore work, accept the inevitable." But that surely means dilutionof appellate responsibility as we have come to know it. It will inthe long run lead to less respect for appellate decisions, a greaternumber of appeals, and a constantly growing problem. Nor willit really achieve any consistency. Bureaucratic anonymity willnot craft abiding legal rules. Witness what has happened in theNational Labor Relations Board and other administrative agencies.So the proliferation of staff should not only stop; pruning shouldbegin. If what we need is more courts and more judges, then weought to hold out for that. Let us not lessen the quality of justicein the meanwhile.

Delay in the appellate process is a constant peril. The mediantime from filing to disposition of a civil case in the federal districtcourts is 9 months; for a criminal case, it is 3.8 months. Sincethis includes cases disposed of by settlement and dismissal, it doesnot give much insight on how long a litigant who desires a trialmust wait. The median time from the joinder of issue to trial is11 months. If we assume that issue is joined in a month, a mostsanguine estimate, then it takes half of the cases more than a yearto come to trial. Conversely, half of the cases are actually triedwithin a year.

The median time for disposition of an appeal after a com-plete record is filed is 6.8 months. Since this includes criminalthorized for the 97 circuit judges as of June 30, 1975, was, according to the Ad-ministrative Office budget, 272. Of these, 228 were assigned to active judges;44 to senior judges.

32 The Commission recommends more "central staff." SRucrURE AND IN-TERNAL PROCEDURES, supra note 7, at 54. It adds the admonitions:

In our view, central staff attorneys should not draft opinions, nor shouldthey identify cases for disposition without oral argument. Duties ap-propriate for central staff include research, preparation of memoranda,and the management and monitoring of appeals to assure that cases movetoward disposition with minimum delay.

Id. It recognizes that this solution is not "altogether free of risk." My view isthat appellate courts with their present staffs already dilute the judge's participationin decision making, and that further increases in staff create not merely the riskbut the virtual certainty of further institutionalization of decision making.

456 [Vol. 23: 448

457VIEWS FROM THE LOWER COURT

cases, now expedited, the time to dispose of civil cases is muchlonger. For example, in the Fifth Circuit a civil case requiringoral argument is not likely to be heard until 16 months after the

appeal is filed. The median time to decision thereafter is 1.4

months. We must add the time required to complete the record.

With that time lag included, it is evident that more than half the

civil cases filed in the Fifth Circuit will not be decided on appeal

until more than two years after the trial has been completed unless

they are placed on the summary calendar. The median time when

cases on the summary calendar are considered is 7.8 months. Theyear required to reach and complete trial is, at least in part, busy

time. But while a case is waiting to be heard on appeal, no plead-

ings are written, no depositions are taken, and no evidence is heard.

All that is done is to prepare a transcript, write briefs, hold oralargument, and render decision.

The only way to tackle delay is to tackle it at each stage.

Another part of this Symposium deals with the relatively minor

gains that can be made by expediting completion of the record,

speeding up opinion writing, and the like.33 These efforts are to be

commended and encouraged. However, since appellate judges

by and large are now working at the peak of human capacity for

the kind of work they must do, the only true solution in the long

run is to add more decision makers.After looking at the political, jurisprudential, and practical

problems of true circuit realignment and enlargement of appellate

capacity, the Commission finally decided that it would suffice to

split the Fifth and Ninth Circuits, and add an "adequate number"of additional judges. One-half of the old Fifth would become the

Eleventh Circuit, and half of the Ninth would become the

Twelfth. 4 The Commission's two reports give all the good rea-

sons why it rejected a major infusion of new judgepower and a

national circuit realignment. In this regard, I think, the Commis-sion failed adequately to face the problem and the only viable

long term solution. The Fifth Circuit has been a master of ap-

pellate speed-up. It has tried a host of procedural changes. It

disposes of 52 percent of its cases on a summary calendar, by

33 See, e.g., Lay, Reconciling Tradition with Reality: The Expedited Ap-

peal, 23 UCLA L. REV. 419 (1976).34 A bill which would split the Fifth Circuit into two divisions is S. 729,

94th Cong., 1st Sess. (1975). The Eastern Division would consist of Florida,

Georgia, Alabama, Mississippi, and the Canal Zone. Three judges would be

added to the Eastern Division, making a total of 12. The Western Division would

consist of Louisiana and Texas; 6 judges would be added making the total 11.

The Ninth Circuit would be split into two divisions: the Southern Division con-

sisting of Arizona, Nevada, and the Southern and Central Districts of California;

and the Northern Division consisting of the rest of the present circuit. Seven

additional judges would be added to bring the number of judges to 11 in the

Southern Division and 9 in the Northern Division.

1976]

UCLA LAW REVIEW

screening and one-sentence opinions. Yet the East-Fifth and theWest-Fifth would require eleven and twelve judges respectively,just to handle the current volume of business.

There appears to be a consensus that the average appellatejudge in the federal system cannot handle any greater volume ofcases than he is now deciding. While there has been a substantialincrease, the plateau appears to have been reached. In 1969, theaverage appellate judge participated in 93 dispositions. In 1973,the average output was 156 dispositions. In 1974, this average'appellate judge participated in 159 terminations, including con-solidated cases and cross appeals. The appellate judges and theirobservers agree that it will be virtually impossible to sustain ahigher annual rate than this, approximately 160 cases a year.35

The Commission is right in observing that it will not do simplyto add judgeships to burdened circuits. Nor will a National Courtof Appeals, deciding perhaps 150 cases a year, dent the workload,though it may ease intercircuit conflicts.

Nine judges have often been referred to as the optimal num-ber of judges for a circuit.3a I would opt for seven. This is in partbased on experience, part on hunch. Seven judges can meet ina room that need not be an assembly hall, can get to know eachother intimately, can dine together easily, and make only 35 pos-sible combinations. Nine judges add only two more people but49 more possible combinations. There is no empirical way to finda magic number, but large circuit courts present almost insuper-able management and jurisprudential problems.IT

At present, each of the circuit courts differs in workload, size,and problems. But the three judges of the First, who terminate

35 The average appellate judge wrote 33 per curiam or other unsigned opin-ions in 1974, about the average since 1970. He wrote 33 signed opinions, againabout the average since 1969. So 'there appears to be a plateau here, too.

s6 STRUCTURE AND INTERNAL PROCEDURES, supra note 7, at 57.87 Chief Judge John R. Brown of the Fifth Circuit believes:[A]ny attempt to divide the states to achieve the ideal of nine-judgecourts throughout the nation will be thwarted by the increasing volumeof judicial business which will require the attention of more than ninejudges within only a few years.Brown, The Fifth Circuit: To Be or Not to Be?, 20 LOYOLA (NEW ORLEANS) L.

REv. 415, 416 (1974). Whether this is correct depends on whether adherence tohistoric circuit boundaries is considered imperative. Judge Brown considered thatcourts composed of judges drawn from a number of states under a wide geographiccompass have a greater "federalizing role," and that this sort of court compositionassures "a nonprovincial approach." Id. at 416-17. But while one may not forother reasons consider the Court of Appeals for the District of Columbia or theSecond Circuit exemplary, neither can be considered either provincial or anti-feder-alist. Nor do I think the charge can fairly be made with respect to most of thefederal trial judges, a majority of whom were born, went to law school, and prac-ticed in the state where they now sit. By way of aside, the Fifth Circuit's confer-ence "table," with space for all fifteen judges, is a cabinet maker's marvel. Its de-sign evokes memories of celebrated peace conference tables.

[Vol. 23: 448

VIEWS FROM THE LOWER COURT

a total of 420 cases a year, 140 per judgeship, all by the samejudges, are certainly a different kind of circuit court than the mam-moth Fifth, sitting in innumerable panels, terminating 2,713 casesand holding 2,760 sittings. There is no Gertrude Stein of juris-prudence who can truly say of these that a circuit is a circuit is acircuit.

It is time to face the real problem. If we are not to abandonthe tradition of "one appeal as of right," and if we are to make thisa true appeal in the traditional sense--one to be heard and decidedby judges-we need both more judges and more circuits. Thecourts ought to be fairly uniform in size and workload. This meanstheir geographical jurisdiction will be radically altered. That juris-diction may be a sacred cow, but it has outgrown its primitive use.About twenty circuits of seven judges each or seventeen of ninejudges each would do nicely in providing the appellate capacitywe need for the next two decades.18

More circuit courts would forbode more applications to theSupreme Court, and more possibilities of intercircuit conflict. Butthe capacity of our federal judicial system to handle the mattersof which it disposes must not be limited by the ability of theSupreme Court to handle, in present fashion, review of an in-creased number of applications for writs. That problem shouldbe faced directly. As we all know after the Freund Study Group, 9

the American Bar Association Committee,4 ° and the Hruska Com-mission 41 reports, there are profound differences concerning wheth-er there is a problem, the nature of that problem if it exists, and howa solution may be achieved. The proposed National Court of Ap-peals is an effort to cope with part of the problem. My own views

38 For present purposes, the figure need not be computed with mathe-

matical accuracy. One hundred sixty terminations per appellate judgeship is morethan the 1974 average, and more than the Commission considers the NationalCourt could handle. On this basis, it would take slightly over 100 judgeships tokeep pace with current filings. This would not erode the backlog, nor keep pacewith future growth. The Judicial Conference thought 11 additional circuit judge-ships were needed in 1972. Added to 97 present judgeships, this would make thecomplement 108. The Fifth Circuit needs, but simply could not accommodate, 8more judges. This brings us to 116. So if we start with 120 judges we wouldhardly be over-supplied. Increasing the number of judgeships will not be effective,as the Commission says, if new judges are merely added to administratively un-manageable courts.

39 FEDERAL JUDICIAL CENTER, REPORT OF THE STUDY GROUP ON THE CASE-

LOAD OF THE SUPREME COURT (1972) (Paul A. Freund, Chairman).40 ABA, REPORT OF THE SPECIAL COMMITTEE ON COORDINATION OF JUDICIAL

IMPROVEMENT (1974) (C. Frank Reifsnyder, Chairman).41 STRUCTURE AND INTERNAL PROCEDURES, supra note 7 (Sen. Roman L.

Hruska, Chairman). See generally Owens, The Hruska Commission's ProposedNational Court of Appeals, 23 UCLA L. REv. 580 (1976).

4591976]

460 UCLA LAW REVIEW [Vol. 23: 448

with respect to that proposal were expressed at the Hruska Com-mission hearings:42 I think it a worthwhile experiment, but I am notconvinced that it is a long-term solution. Perhaps sooner or laterwe will have to choose between the doctrine that every dis-appointed suitor is entitled to one appeal 4I and the manifestdifficulties of a four-level judicial structure, or we may haveto resort to a system of institutionalized case processing. That cer-tainly involves political considerations, but no solution will everbe purely jurisprudential. Even those considerations that appearto be based solely on efficient judicial administration will havesome potential impact on the outcome of the cases. Therefore,any change in the present structure radical enough to promise re-lief will be inherently controversial. However, the only non-controversial article a trial judge can write in 1976 is one thatpraises the dedication of all judges, trial and appellate, commendsa slight reduction in the reversal rate, and urges a sharp increasein judicial salary. I am an advocate of all of these, and my en-thusiasm for them waxes in the order listed.

42 2 Hearings Before the Commission on Revision of the Federal Court Ap-pellate System, Second Phase, 1035-49 (1975).

43 The issue is open to debate only if the right to one appeal is not amatter of due process, a question on which opinions differ and with respect towhich there is no authoritative decision. Having never considered fully the con-stitutional issues, I withhold present expression of opinion. But I add a few ob-servations about other aspects of the problem.

In the past, "one appeal of right" largely meant that those aggrieved suitorswho could afford to pay for an appeal had the right to do so. Money wasneeded for counsel's fees, printed briefs, and court costs. Those who lackedmoney naturally could not afford an appeal. Hence, expense served as a filterof appeals and limited their number. We have properly removed most of thisbarrier by more liberal interpretation of in forma pauperis statutes and by provid-ing appointed counsel to persons who cannot afford to pay lawyers. In addition,the escalation of the cost of trial in the lower court has been so great that, even forthose who must pay for an appeal, its cost is not a significant addition to thetotal cost of litigation. As data set forth in note 14, supra, indicate, the rate ofappeal has therefore escalated enormously in the past decade. It is just andproper that we have made the ability to pay for an appeal irrelevant to the rightto appeal, and that we have made an appeal less financially onerous even for thosewho are able to pay. However, no appellate system that we now envision couldhandle the volume of work involved if an appeal were taken on every case. Itis unlikely that we can rely merely on a losing litigant or his lawyer beingwilling to accept defeat if an appeal can be had at little cost. There is a prag-matic nccd for some way to limit the kind of issues or cases that may be appealedif cost is no longer a barrier.

One possibility that appears never to have been fully considered and debatedwould be conversion of the courts of appeals to handling matters only on applica-tion for writ. While I am not yet prepared to advocate this method, I do seesome attractive possibilities to it. Whether or not up to 5,000 applications forwrits per year constitute an impossible load, the Supreme Court has demonstratedthat a court can readily consider many more applications for writ than it canhandle appeals, even by screening. The approach of the applicant, as well as thework required of the court, are different. It appears possible that a courtswamped by 2,000 appeals might be able to achieve the appellate function better ifit considered 2,000 applications for writ and allowed appeals in, say, 500 cases.

1976] VIEWS FROM THE LOWER COURT

The suggestions I have made from the sidelines are neithereasily followed nor infallible cures. They will be difficult to imple-ment, and it may be a long time before any results are seen. Butif the weaknesses in the appellate system were easy to eliminate,they would long since have been remedied. Others have struggledwith the problem long and hard. A trial judge is not likely to hiton panaceas. Indeed, it is evident that the problems are acute, thejudiciary and other observers are aware of them, and the com-ments in this Article are not original perceptions visible only fromthe underside of the bench.

APPENDIX

COMMENTS FROM THE JUDICIARY

Judge Ben Cushing Duniway:

When I came on the court [in 1961], I had time to not only readall of the briefs in every case I heard myself, which I still do, and allthe motion papers in every motion that I was called upon to pass upon,which I still do, but I could also go back to the record and I could takethe time as I went along to pull books off the shelves and look at them.And then I had time, when I was assigned a case, to write. And occa-sionally I could do what I call "thinking", which was to put my feet onthe desk and look at the ceiling and scratch my head and say, "Howshould this thing be handled?"

I always thought the [primary] business of a Judge-and particularlyof an appellate judge- . . . is to declare the laws not just for the case

he's deciding, but [to declare] the law for whatever the jurisdiction of hisCourt may be, and he ought to have time enough to give the questionthe study that it deserves and to produce an opinion which will be con-vincing and which will become an authority within the area of theCourt's jurisdiction.

Today the situation is quite different.

I have a strong feeling and I know many of my brothers and sisterson the Court have the same feeling-that we are no longer able to giveto the cases that ought to have careful attention the time and attentionthat they deserve.

Quoted in Hruska, The Commission on Revision of the Federal Court

Appellate System: A Legislative History, 1974 ARIZ. ST. L.J. 579,583 n.14.

Judge Henry Friendly:

I think [the prestige of the courts of appeals] would seriously decline ifthe increase [in number of circuits] were to say 20. To alter slightly thewords of Professor Geoffrey Hazard, "what were once authoritative ap-pellate tribunals, subject to occasional review by the Supreme Court ...would have been converted into a judicial Tower of Babel." The pro-liferation of utterances could divest any one of these courts of significantauthority."

H. FRIENDLY, FEDERAL JURISDICTION: A GENERAL VIEW 40 (1973).

462 UCLA LAW REVIEW [Vol. 23: 448

The Judicial Conference of the United States:[N]ine is the maximum number of active judgeship positions which canbe allotted to a court of appeals without impairing the efficiency of itsoperation and its unity as a judicial institution.

REPORT OF THE PROCEEDINGS OF THE JUDICIAL CONFERENCE OF THE

UNITED STATES 15 (1964).

Judge Charles M. Merrill:I am concerned about what is going to happen to the Courts of

Appeal, and to the Ninth Circuit in particular, if sensible and reasonablesize is not imposed and recognized. As an extreme, one can conceive ofone single Court of Appeals headquartered in Washington with up to ahundred members, scattered all over the country, sitting together whenand where and over what, as some computer may decide. Now, that, tome, is not [a] Court. It is an aggregation of judges, and the only courtthat you would have there to which a judge might feel associated or con-nected would be the fleeting and ephemeral grouping of three in whichhe happens to sit at one time or during one week. And I am afraid theNinth Circuit is starting now to drift in that direction, and if the onlysolution to our problems . . . is to increase the number of judges on theCourt-and that, of course, is one solution [-] then I am afraid weare faced with a court of 15 or 18, and perhaps in a couple of years oneof 27, and as far as I can see, the court is gone.

. What has been referred to . . . as collegiality of the Court . . . isgone.

There is not, it seems to me, an institution to which a judge canattach himself, with which he can feel a sense of identity, and to me thatis very important. I felt it on this Court when it was a Court of nineand when our work was being done by the judges of our Court and theirseniors, the senior judges.

This is not the case now. One third or thirty percent of our workis being done by judges who are not members of our Court, and theCourt is larger and is scattered more widely, and the communication andcontact between the judges is lost to a very substantial degree.

I feel this. It is a hard thing to express. I am sure that judgesknow what I am talking about, but I want to be a part of the Court. Idon't want to be just one judge in an aggregation of judges who occa-sionally sit together and do their work-that is the point I wish particu-larly to make.

Quoted in Hruska, The Commission on Revision of the Federal CourtAppellate System: A Legislative History, 1974 ARIZ. ST. L.J. 579, 584n.17.

1976] VIEWS FROM THE LOWER COURT 463

COMMENTS FROM OBSERVERSAmerican Bar Foundation:

The basic problem already being created by the continuing growth in

the number of appeals and of judges is the diminishing effectiveness of

existing devices for holding the federal judiciary together as a single

enterprise. In particular, the function of the Supreme Court as an

agency for harmonizing and unifying federal law is spread ever morethinly.

ABA, ACCOMMODATING THE WORKLOAD OF THE UNITED STATES

COURTS OF APPEALS, REPORT OF RECOMMENDATION 6 (1968).

Professor Paul Carrington:The pressure created by trying to stay abreast of a congested docket,

however, threatens the successful performance of the law-making func-

tion of the federal courts of appeals. The law-making role requires a

delicate balance between the importance of flexibility in the national law

and the importance of stability of doctrine. Busy judges will find it

difficult to evaluate these needs in the cases that come before them, andthey may be more tempted to take the shortcuts which can be made byignoring one of the interests.

Congestion also threatens the effectiveness of review. Pressure oftime may create a tendency to give greater deference to primary decision

makers. In addition, there is a paradoxical prospect that increased ap-

pellate activity may impair the very quality of the process that justifies its

use. There is a danger that courts of appeals' decisions may themselvesbecome more responsive to the personal values of individual circuit

judges, and less responsive to the general and idealized values of the

system as a thing apart from the individuals momentarily operating it.

* * Even the most industrious, detached, and selfless judge is

bound to respond to the pressures of congestion, indeed perhaps the

more so for his sensitivity. If meritorious cases are being delayed, or

receiving too little attention, the conscientious judge must respond to

his impulses to cut short his deliberation over matters that seem less

promising. As he does so, he is unmistakably transforming the character

of his decision and the process by which it is made.

Carrington, Crowded Dockets and the Courts of Appeals: The Threatto the Function of Review and the National Law, 82 HARV. L. REv.

542, 554-55 (1969).

Dean Roger C. Cramton:[T]he federal courts of appeals and the Supreme Court are faced with

massive and growing workloads that cannot be handled adequately bypresent manpower and techniques under existing jurisdictional rules.

There is agreement that a major current problem is undue delay in the

disposition of appeals. The time required for disposition of civil appeals

has grown to nearly two years in several courts of appeals. Althoughthe record is better in criminal cases, growing backlogs and resultant de-

lays are being encountered here also. The purposes of the legal sys-

tem are hampered and a litigant's right to appeal threatened by suchundue delay.

A second and widely shared concern involves the quality of the

decisional process as the courts of appeals and the Supreme Court adopt

assembly-line techniques to deal with massive case loads. . ..

A third major concern involves the current methods of handling

criminal appeals and prisoner petitions challenging state and federal con-

victions. Finality and dispatch in criminal cases have been bled out of

UCLA LAW REVIEW [Vol. 23: 448

the federal appellate process, resulting in a large, repetitive, and oftenfrivolous body of appeals crowding the dockets ...

There is dispute concerning the significance of a fourth problem-lack of stability and evenhandedness in federal law. . . . Litigants aretreated differently by different panels of the same courts of appeals andby courts of appeals sitting in different parts of the country. Uncertaintyconcerning important questions of law also encourages relitigation of thesame question in different courts and increases the burdens at all courtlevels by generating a continuous flow of new lawsuits throughout thecountry. Professor Rosenberg and Judge Haynsworth conclude that theincreasing instability of federal law warrants major structural revisions ofthe federal appellate system, including the creation of a new tribunalunder the Supreme Court. Judge Friendly, on the other hand, whileagreeing that the Supreme Court is unable to resolve all intercircuitconflicts, concludes that structural revision would not accomplish thepurposes sought and is objectionable because it would create specializedtribunals and result in jurisdictional bickering between the new court andexisting courts.

Cramton, Federal Appellate Justice in 1973, 59 CORNELL L. REV. 571,572-73 (1974).

Professor Charles Wright:If there are to be more judges in the overworked circuits, while retaininga limit of nine judges per court, the only alternative, as the Special Com-mittee recognized, is to split existing courts and create more circuits.It is doubtful whether [circuit splitting] would be feasible [as a methodof adding additional judges to a region without creating a bench of morethan nine judges] for the Second Circuit, since the day is not far offwhen business arising out of the state of New York will alone amountto more than the [workload] which a nine-judge court can . . . handle.How the District of Columbia Circuit can be split defies imagination.

Wright, The Overloaded Fifth Circuit: A Crisis in Judicial Administra-tion, 42 TEXAS L. REv. 949, 973 (1964).