Copr. © West 1999 No Claim to Orig. U.S. Govt. Works

35 AMULR 909

(Cite as: 35 Am. U. L. Rev. 909)

American University Law Review

Summer, 1986

A Review of Recent Decisions of the United States Court of Appeals for the

Federal Circuit

*909 INTRODUCTION SELECTIVE PUBLICATION OF OPINIONS: ONE JUDGE'S VIEW

The Honorable Philip Nichols, Jr. [FNa]

Copyright 1986 by Washington College of Law of The American University; Philip

Nichols, Jr.

INTRODUCTION

On October 1, 1982, when the United States Court of Appeals for the Federal Circuit (Federal Circuit) came into legal existence under The Federal Courts Improvement Act, [FN1] it adopted as its Rule 18 the following:

(a) Opinions.

All decisions, and opinions accompanying decisions, of this court shall be provided to the parties, shall be public records of the court, and shall be accessible to the public. Disposition of appeals shall be with a published opinion or an unpublished opinion. Opinions which do not add significantly or usefully to the body of law or would not have precedential value will not be published in commercial reports of decisions. Opinions designated as unpublished shall not be employed as precedent by this court, nor may they be cited by counsel as precedent, except in support of a claim of res judicata, collateral estoppel, or law of the case. A party may, on motion, request that an unpublished opinion be reissued as a published opinion, citing reasons therefor. Such motion will be granted or denied by the panel that rendered the decision.

(b) Orders.

The action of the court in granting or denying motions, petitions, and similar actions, shall be reflected in an order of the court. [FN2]

*910 Thereby, the new court joined all the previously existing 'regional' circuit courts of the federal system, which had adopted plans to limit publication at the behest of the 1972 Judicial Conference of the United States. [FN3] In Rule 18, 'will not be published' means will not be published by the court. Since the unpublished opinions are still part of the public records, nothing prevents their publication by others, and it often occurs. In no sense then, is an unpublished opinion concealed. The judges of the new court doubtless felt they had no choice in the matter in view of the position of the Judicial Conference, but it is not of record that they made any effort to change conference policy or secure for themselves an exemption from it. They believed it was the right course for them to follow and did so willingly, despite the fact that the policy had enjoyed bad press in academic critiques. [FN4]

Comments by judges themselves, based on experience, has hitherto been conspicuous by its absence. Among prior statements by several judges, testimony by the late Circuit Judge Robert Sprecher of the Seventh Circuit before a Commission chaired by the late Senator Roman L. Hruska [FN5] has been much overworked; though the judges could at the time have had little practical experience with their circuit's plan, or any other, and were mostly speculating about the future. It has occurred to me that I might usefully discuss the subject from the point of view of a judge who has actually lived with *911 the plans, both as an active and later as a senior judge on the Federal Circuit, and also as a visiting judge [FN6] on other circuits. The opinions expressed are entirely my own. Were Judge Sprecher still alive, he could doubtless do a better job, but alas, we shall never hear from him again.

I.

The academic critiques [FN7] start out with the unstated assumption that the circuit judges are omnipotent regarding publication of their opinions, [FN8] but that is far from being the case. Their unpublished opinions are only unpublished by them. They cannot prevent publication by others, which frequently occurs, [FN9] although restriction on citation limits the economic gain from doing so. When judges desire publication, their means of bringing it about are limited.

I served from 1966 to 1982 on the old Court of Claims. It published official reports using appropriated funds as it had done since soon after its birth, the earliest cases reported in Volume One going back to 1863. It had a 'Reporter of Decisions' who was aided by the clerk of court. From 1867, the Reporter of Decisions was Judge Charles C. Nott; he was Chief Justice 1896-1905, and continued to report the decisions until 1914. [FN10] Afterwards, the Reporter of Decisions was not a judge and no doubt among the judges, only Nott would have been willing to take on this extra and most onerous duty. In Nott's time, these official reports were so serviceable no others were needed. They included summaries of counsel's arguments in important cases: a useful element of old- time reporting for which the modern reader of these old cases is grateful because of their help in understanding the cases and determining for what precedents they are properly citable. The utility of these official reports *912 declined as the years went on, the delays in getting them out increased, and the money cost multiplied. We long had to use the Government Printing Office at a price greatly exceeding that of a private printer. Towards the end, our final publication of reports was in arrears two years or more, [FN11] and it was not even possible to endow them with headnotes. We owe a great deal to the dedication and resourcefulness of our last Reporter of Decisions, Audrey Bernhardt. Through her efforts the problems were overcome and not allowed to frustrate the project entirely. The reports were a constant source of vexation even so, and one of their worst features was the disincentive they provided to desirable private publication of our decisions.

The Court of Customs and Patent Appeals also had two official reports it published, but had the good sense to abandon at least one of them well before the 1982 merger. [FN12] I am sure no judge of the new Federal Circuit ever dreamed of suggesting it should itself publish official reports. The idea of federal circuit courts below the Supreme Court actually themselves publishing their own opinions we may safely regard as dead for the foreseeable future, although the idea survives and works as to certain specialized courts.

This leaves us in the hands of the West Publishing Company. During the life of the Court of Claims, West had begun to publish some of our decisions, and until near the end flatly refused to publish others duly offered to it. Of course, it felt under no obligation while the official reports were (sooner or later) coming out. Its reasons for selective publication were stated to me to be its belief that the Court of Claims was not doing its share to limit the proliferation of judicial opinions, and it relented to the point of publishing everything we tendered for publication only when satisfied we had limitation measures in effect. Among the prolix decisions of the old Court of Claims were conspicuously those with fact findings by trial commissioners which extended to a length and detail not paralleled by any other court. This was corrected by simply withholding the findings from publication. It was clear to me that West had the *913 cause of limiting proliferation of opinions very much at heart and the business reasons for this were and are obvious. The bloating of the current volumes of Federal Reporter and their rabbit-like multiplication on library shelves are, in fact, a menace to the market for the Federal Reporter. [FN13] To complete the picture, however, it is only fair to say that West revealed a special antipathy towards opinions dealing with subject matters West felt were not of interest to the majority of its customers, for example, claims of Indian Tribes under the Indian Claims Commission Act. [FN14]

I am of the view that if the seemingly omnipotent circuit judges of any federal court ever started tendering all their opinions for private publication, they would be able to obtain publication, if at all, only under drastically altered conditions that might not be to their liking, or else all the miseries of the old official reports would have to be revived. Of course, now, if academia were to divert some of its grant money to subsidize 100 percent publication, that might be different! Failing this, it must be remembered that any private publisher for profit has a perfect right to reject material it considers too prolix or repetitious to hold the interest of its readers, and the world is a happier place because, generally, they do so. West actually does much to keep its stable of judicial authors contented, and we can assume it would fail in this only in case of extreme necessity.

II.

Circuit Judge Sprecher has had academia dithering for a dozen years with his forecast that under the limited publication system then being put into effect, judges would have to accumulate and consult files of unpublished opinions. [FN15] It is imagined that counsel are up against the risk of having cases decided against them on the basis of opinions they would normally have no access to, and if they did, could not cite themselves. [FN16] I can, and hereby do, state that I *914 have never used unpublished material of any sort, not of record in the case, as a source of reasoning, or as a precedent. I read the unpublished opinions of my court when issued, rarely recommend that they be published, but usually do nothing with them and forget all about them. There is enough else to remember or check upon, so one is happy there is something one can drop into oblivion. If I had to factor all the unpublished decisions into my thought processes, my task would become too much for me and I would have to quit. I have never, in internal discussions or memo writing in my court, known another judge to mention a past unpublished decision as a reason for doing or not doing anything with a current case. In sitting on another court by designation, I have never been briefed by any active judge or staff employee of that court as to the existence of any unpublished decision that I should be aware of before casting my vote in conference. Of course, they do not expect that I, as a visitor, will know everything about their court that they do, [FN17] and least of all would I be likely to know about their unpublished decisions, so the omission to mention them is, I think, especially significant as to the practice. I have had experience of citation of unpublished opinions by counsel where this is not expressly prohibited. [FN18]

This bugaboo based on Sprecher's testimony ought to be consigned to the same oblivion as the unpublished decisions themselves; hard as it may be for academia to believe, the nonprecedent is really not a precedent, and the rule works as intended.

III.

The above accusation perhaps implies that some hard work has gone into unpublished decisions that we cannot bear to let go to waste afterwards. The opposite charge is also voiced, that is, because it is to be unpublished, a decision will not receive enough work, the writing will be sloppy, and real difficulties with a position taken will be passed over in silence. [FN19] Some of the critics have read unpublished opinions and produce anecdotal instances where they are thus inadequate in the critic's eyes. [FN20] Of course, some intermediate *915 appellate opinions will be badly made, published or not published, else we would not need a Supreme Court. It is hard to say such bad opinions are better off published and precedential than unpublished and unprecedential, unless the object is to expose the incompetence of the circuit judges to a maximum readership. [FN21]

The idea that the commentators, the bar, etc., are 'consumers' who will point out mistakes they know about by reading published opinions is unrealistic. What a judge gets from these 'consumers' is silence or a big yawn, as a rule. After hundreds of published opinions, I have had no experience of 'consumer' feedback in most cases, and when it does occur, it is not half as devastating as that emanating from colleagues on the court itself. [FN22] In all my opinions, published or unpublished, meeting their standards is to my mind my most difficult burden.

The real question here is, however, whether the decision not to publish has contributed to the inadequacy. Judges are quoted as saying that they sometimes start on opinions that 'will not write' and the difficulty in putting it on paper is finally realized to be due to a weakness in the decision itself, as tentatively arrived at after oral argument. [FN23] I have experienced this, other judges say they have, and there can be no doubt it occurs. The supposition of the critic apparently is, the more lengthy, learned, and elaborate the mode of writing, the more likely such enlightenment is to occur and thus the more probable the final product will be correct. It is assumed, and rightly assumed, that a decision not to publish will tend towards simplicity, brevity, and directness. The critic's error is, I think, in supposing that the writing difficulty that generates the judicial epiphany occurs in the portions of a published opinion that would not be found in one written for nonpublication. That is not my experience in which the instructive writing difficulty is found most often in writing decisions for nonpublication. I have also observed, among colleagues, that a writing for nonpublication often comes out as plainly signaling its qualification to be published. When you see an unpublished decision run on for several pages, the question if it should be published at once comes to the mind. In that event, the original decision not to publish is not set in concrete: the decision is *916 published. In which case, it is usually not necessary to change it very much; the idea that published and unpublished opinions are totally different is just as incorrect as the notion that unpublished opinions are somehow a lode of unmined precious metal.

In one respect, briefing and opinion writing are irrevocably changed by selective publication, even if correctly done. That is, arguments based on frequency of citation cannot be used. One cannot say: A v. B has been cited and followed twenty times, therefore it should not be overruled or should be broadly construed; C v. D has never been cited or followed, therefore it has scant value as a precedent. This type of argument was perhaps most often useful to a court when the question was to assess the value of a citation from a sister circuit in the federal system. Sometimes a rule as stated in an opinion often cited could be shown to be yet dicta that never actually governed the disposition even of the case where it was uttered. If there is a conflict of precedents, it may help decide which should be followed. A problem that would have cut down the value of this technique of dealing with case precedent is the practice of 'string citation.' When a case appears in a string of precedents, one cannot tell how much it has really influenced the court that merely cited it. On the whole, I do not see the justification for retaining the whole ponderous apparatus of unselective publication merely to save the occasional use of this usually unpersuasive argument. On the other hand, we are now enabled to argue that if a circuit selects a decision for publication, it must itself consider it has created an important precedent. One could not say that with nonselective publication.

The true reason behind the selective publication policy is that it is wrong to ask publishers to publish, libraries to collect, and scholars to read opinions that merely labor the obvious, so far as they deal with the law at all, rehashing conclusions already reached in authoritative decisions of the same court or the Supreme Court.

IV.

Most circuits do not allow citation of unpublished cases and will not cite them in their opinions, with three circuits an exception if no better precedents are available. Apparently, the Tenth Circuit, though resorting to unpublished decisions, will freely cite them in unrelated cases and will allow counsel to do so. Others have no firm prohibition. [FN24] The Eleventh Circuit has held one of its panels *917 'bound' by a prior decision of another panel, a 'Local Rule 25 affirmance without opinion.' [FN25] This frequently appears at least partly commendable to commentators, who often find noncitation even worse than nonpublication. [FN26] The reason appears to be: if citation is allowed, a feasible way is found to make so-called unpublished decisions available to the bar. [FN27] In effect, they are published a different way and none are truly unpublished.

In the final half dozen years of the Court of Claims, we had what were in substance unpublished opinions, called 'Orders,' treated in a manner somewhat analogous to the Tenth and Eleventh Circuits' unpublished, but citable, opinions. They were destined for eventual publication in the rear parts of each of our official hard cover volumes, but these were so inordinately delayed, that always a large mass of these 'Orders' had not yet appeared. Subscribers to our slip opinions did not get them. There was never a suggestion that West should publish them, rather their existence was one of the things that satisfied West that we were cooperating in reducing the bloat of their volumes. There was no fixed standard for selection to publish and no special push to exclude precedential analysis from orders or get precedential orders published. As West was until near the end refusing to publish all the opinions we thought to be precedential as stated above, [FN28] it was futile to think of asking them to publish Orders. Copies of these Orders could be purchased at moderate prices from the clerk, and supplemental efforts were made, but were unsuccessful, to encourage private publication to a limited subscription list.

The government was defendant in all our cases and, of course, government attorneys were well acquainted with, and had ready access to, all as yet unpublished 'Orders.' There was no restriction on either side citing them, but it is of course not necessary to state who cited them most. The court itself frequently cited these orders also. Government counsel, when citing, furnished copies to their adversaries, and some of their more scrupulous attorneys even furnished to claimants' attorneys copies of 'Orders' that favored claimants and so were not likely to be relied on by the government.

I do not think any of us were wholly satisfied with the fairness of *918 this arrangement. Some of our number resisted proposals to limit citation of these 'Orders,' stoutly arguing that it was wrong to hamper the seeker for truth in making use of anything that could be discovered that might throw light. Efforts continued sporadically during the remaining life of the old Court of Claims to find a better way, but on the abolition of that court on October 1, 1982, the better way was yet unfound. There was, I think, much relief in putting that unhappy chapter of our publishing history behind us with the new Rule 18 and Standard Operating Procedure (SOP) No. 7. [FN29]

It is certainly arguable that, short of ex parte contacts and extrajudicial investigations, a judge is entitled to employ the procedures most helpful to him, in his eyes, to decide cases rightly, regardless of how equitably they operate towards the concerns of counsel. The contrary argument commits the judge to umpire equally over the contests of counsel, perhaps at the expense of his anxiety that his decision may not be the best. You might, I suppose, argue that a judge should not even read a published court opinion not cited by counsel without first telling them he intends to do so and entertaining objections, if any. Some scrupulous trial judges do this before taking judicial notice of regulations not called to their attention first by counsel and not in the Code of Federal Regulations. The sensible compromise must be found and is much aided by not relegating to nonpublication any decision of potential value for citation. The Eleventh Circuit in the instance referred to above has committed what seems the absurdity of declaring a decision has no precedential value and then citing it as a binding precedent. This compromise may have to be differently framed in different courts. The Federal Circuit still has a large proportion of cases in which the government is a party, and this is, I think, a factor imposing on the Federal Circuit a special duty not to allow counsel advantages rising out of superior knowledge of decisions unpublished by the court's election.

It is, however, I think another hobgoblin that counsel of an 'inside' group, knowing of unpublished decisions, may quote their arguments before the court without his opponent even knowing what is going on. [FN30] This scenario continues that the judges will smile in recognition of their own work, and hold in favor of the attorney who quotes them without attribution, he having, presumably, displayed his better taste. Those who read and believed the opening parts of this paper will know that the judges will probably not even remember *919 or recognize their own words in former unpublished decisions but, if they do, they will also remember their own announcement that the unpublished decisions will not be precedents, and resent the attempt to make them so. Also, they will wonder why counsel did not quote the surely extant published material equally on point. I suggest this kind of approach in an appellate argument is a sure key to defeat. Judges, reading such imaginary scenarios on the uses of unpublished opinions, often will ask themselves, what kind of people do they think we are?

V.

If all the appeals filed in any intermediate federal court ought to be there, the court would have no need for a selective publication policy. The ones that should not be there create the need. The theory of appellate review in the past was that losers in trial courts would exercise self-censorship and forego appeals unless competent counsel determined that legal errors of the trial court could be pointed out or, if not certainly errors, at least rulings as to which competent judges might differ. Thus, the appellate judges, handling a moderate volume of appeals, could devote collegial attention to them, act without haste or pressure, and set down the reasons at length for every decision. We all know this does not describe why appeals are filed today.

This appellate judge, at least, believes that practically the entire lay population, and a substantial minority of the lawyers, think federal judges at all levels will do anything they feel like doing with every case. One who believes this will not see any appeal as hopeless. Judges must at least share the blame for creating this misapprehension, but that is water over the dam; the point is it exists, and many appeals are filed under its influence. With them, the assignment of arguable errors is only ostensibly, not really, the grounds of appeal. In many instances, too, counsel hope for delay, or to harass the other side, or for flustered and confused appellate judges. Some of these appeals can be and increasingly are dealt with as frivolous, but to so label them presupposes a want of subjective good faith not usually capable of proof. The appellants I am discussing here are not dishonest, merely misinformed as to what we still hope are the rules of the game. Losing litigants have a right of access to the federal appellate tribunals that cannot be denied them; they can demand that courts at least look at their cases. What they cannot demand, and in fact do not ask, is that every appeal they take, whether taken reasonably or otherwise, will result in loading up the *920 law publishers and the libraries with useless dissertations. They cannot require that forests be destroyed to disseminate and perpetuate statements such as this, from a recent unpublished decision:

Although petitioner seems unaware of it, we do not have jurisdiction to review the initial decision of February 5, 1985, which became final March 12, 1985, because he failed to appeal within the period set by statute, 5 U.S.C. § 7703(b)(1), and that portion of the appeal must be dismissed.

The fifth and Eleventh Circuits have frequently used the practice of not only not publishing, but not writing opinions in certain cases of affirmance. The fifth Circuit's entire decision reads: 'AFFIRMED. See Loc.R. 47.6.' [FN31] Not only is this in bad odor with those who criticize nonpublication of opinions that are at least written, [FN32] but even judges of other circuits who employ selective publication are inclined to disown this variant. [FN33] The rule itself leaves the litigant to guess which of four stated reasons is why an opinion was not written in his case. None of the four is what I feel, after experience with this procedure as a visiting judge, to be the real reason. That is usually, if not always, that the case presents no genuine appealable issue and the parties who initiated the appeal should have known this and probably did. Invocation of the rule is thus a rebuke for misuse of the appellate process, but one administered with true Southern courtesy. I rather like the procedure and so stated to Senator Hruska's Commission, [FN34] and would be happy to see it in use in our court, provided it is limited to the meaning suggested above, of course in more diplomatic language. An obvious *921 improvement, easily made, would be to modify the standard decision language to refer to the portion of the rule that is pertinent, not the whole of it.

VI.

To avoid misunderstanding, I should add that by my work ethic, no opinions, published or not, should be written primarily with an eye to making a precedent. The court has had a problem set before it; its task is to arrive at and explain a just, reasonable, and lawful solution. How good or bad the precedent will be respecting other problems is at most a matter for precautionary consideration unless, of course, one is a Justice of the United States Supreme Court. When an opinion is a valuable precedent, it is so because it contains material for precedent not to be found in previous published opinions of the same court, or the Supreme Court, and not always according to the intentions of its author. Very often we find our opinions doing duty as precedents in ways quite other than we expected, and this is to the good. Our surprise may be due to not having written as accurately as we intended, but most often it is because we have failed to anticipate the problems our opinions will be used to solve. That is all right: we should not use our power to select opinions for publication so as to prevent that from occurring. The occasional rejection for publication of an opinion that afterwards appears to be a valuable precedent therefore must occur, and it is foolish to pretend otherwise, but it does not justify total disregard of a system that has proved to be of value. It does justify alertness to reduce instances of this and to deal with them rationally when they occur.

VII.

While a considerable amount of muttering about selective publication still occurs, it appears that judges like it and feel at home with it, that no better method of checking proliferation of opinions is 'practical politics' at this time, and that the press, the public, and Congress do not view selective publication as one of the major problems concerning the judiciary. It has not been on the agenda of recent sessions of the Judicial Conference of the United States. If the academic world or the bar really feels that serious harm may be occurring from unwise or improper selection of opinions for publication according to their anticipated precedential value, I would think it would not be alleviated by more self-serving statements by the judges themselves. With the unselected opinions being of public *922 record and open to all, any disinterested, nongovernmental foundation or law center could make a study of them by an appropriate statistical method. Perhaps, they should be so examined. Absent this, the attention turns thus from whether to how--how shall selection of opinions for publication or nonpublication be made?

The Federal Circuit apparently surpasses all the circuits in the low ratio of published opinions to all opinions; by my mathematics, of 946 opinions issued in court year ended October 1, 1985, only 223 were published, which I presume means published in the court's 'chosen instrument,' the Federal Reporter. Some others may have been published or abstracted elsewhere. This, I judge, is due to differences in the nature of the Federal Circuit's business; for instance, the high ratio of pro se appeals, and the high ratio of cases already the subject of published opinions at the trial level. A recent report by Donna Stienstra of the Federal Judicial Center indicates for 1981-84 that, in the other circuits, the unpublished decisions ran slightly over 50 percent; the maximum was for the Third Circuit in 1984, 77.1 percent. [FN35] I take this as superseding inconsistent earlier data.

Despite differences in the way different circuits state their plans, I think it probable that a like case would usually be published, or not published, similarly in all circuits. The judges, as time has passed, have gained a better understanding of how to select opinions for publication than their original rules might suggest. Choice as to selection for publication is something you do better with practice. Ms. Stienstra has found that practice does not always conform to 'written policy' and attributes this to practice responding faster to changing events. [FN36]

Stienstra did not have access to the text of the Federal Circuit plan. It is hidden, on the principle of Poe's story The Purloined Letter, by placing it in plain sight. It is in this court's SOP's, called 'Internal Operating Procedures' by some other courts, that set forth the court's routine of paper shuffling, of no possible interest except to those who must shuffle the papers. Cases are selected for publication according to the following criteria:

5. The court's policy is to limit publication to opinions meeting one or more of these criteria:

(a) The case is a test case.

(b) An issue of first impression is treated.

(c) A new rule of law is established.

*923 (d) An existing rule of law is criticized, clarified, altered, or modified.

(e) An existing rule of law is applied to facts significantly different from those to which that rule has been previously applied.

(f) An actual or apparent conflict in or with past holdings of this or other courts is created, resolved, or continued.

(g) A legal issue of substantial public interest, which the court has not sufficiently treated recently, is resolved.

(h) A significantly new factual situation, likely to be of interest to a wide spectrum of persons other than the parties to the case, is set forth.

(i) A new interpretation of a Supreme Court decision, or of a statute, is set forth.

(j) A new constitutional or statutory issue is treated.

(k) A previously overlooked rule of law is treated.

(l) Procedural errors, or errors in the conduct of the judicial process, are corrected, whether by remand with instructions or otherwise.

(m) The case has been returned by the U.S. Supreme Court for disposition by action of this court other than ministerial obedience to directions of the Court.

(n) A panel desires to adopt as precedent in this court an opinion of a lower tribunal, in whole or in part.

6. The election to employ an unpublished opinion shall be unanimous among the Judges of the panel. [FN37]

It will be seen that several new criteria for publication exist that are not stated in the plans of all older courts. These include: (e) application of an old rule to a new fact situation; (k) knowing conflict with decisions of this or other courts; and (m) cases on remand from the Supreme Court. These close significant gaps that have been shown to exist in prior plans. [FN38] I do not believe that a case often occurs that is not publishable under this plan that I would say should be published. It should be noted, however, that the plan does not expressly require publication of any opinion, and this, I think, is a mistake. It also would be better to add examples of what it is the policy of the court not to publish, for example, cases where the outcome would have appeared inevitable to a disinterested, *924 qualified lawyer, at the latest on completion of the briefs, because of the absence of any seriously arguable issue of law.

The requirement of unanimity for nonpublication is, however, found in only two other plans, [FN39] and is a safeguard against injudicious failure to publish, if not so much so against injudicious publication. The low ratio of published to unpublished opinions, as compared to figures for other courts, would suggest that additional safeguards against failure to publish would not do any harm. Some appellate judges like to see their own deathless prose in published format, while others much prefer the unpublished mode, and are perfectly happy with assignment to put out decisions for nonpublication by the dozens. This causes variations in the ratios among individual authoring judges and, doubtless, it takes both kinds to make an efficient court. I think erroneous failure to publish and erroneous election to publish both have occurred. I am conscious of having been guilty of one flagrant instance of the former myself, but wild horses could not drag its identity out of me, and no one else has divulged it.

I do not think such errors are anywhere near numerous enough to justify abandoning the selective publication system itself, but they do justify giving more thought to the selections than sometimes they receive. The desire of some judges to publish must be curbed, as must that of others to run on at length under the shield of nonpublication. Selective publication adds another class of issues where even the judge who tries hardest will sometimes call the case wrong, and no brain is great enough to be right all the time. [FN40]

When the decision to publish or not publish is correctly made, the question of the so-called nonprecedential precedent will not arise. If that nonprecedential precedent were published and made ostensibly a precedent, it would never be cited and never be a precedent in reality. When the decision to publish vel non is incorrectly made, the potential nonprecedential precedent is lurking there and may be a source of future trouble for the court. It is a loose cannon.

VIII.

Some circuits publish if there is a dissenting opinion, [FN41] but the *925 Federal Circuit does so only if the dissenting judge insists on it. If dissenting myself, I would never so insist: it would result in making the decision I objected to precedential instead of nonprecedential, under Rule 18, and I would be 'bound' by it afterwards. I have not been aware of much such insistence by other of our judges when dissenting. To some minds, publication is the only logical course: how can you argue there is no issue of law in the case on which reasonable legal minds could differ, if the dissent establishes that either they do differ, or the dissenter is being unreasonable? But the dissent may be based on something other than a difference of opinion about the law, for example, whether a fact finding has the support of substantial evidence. I like to feel free to dissent or concur separately and, because I do, my silent adherence to a larger number of majority opinions by others is shown to be willing and unconstrained. I think tying up the question of dissenting with publication may work adversely on the dissenter, constraining him not to dissent, and this I would avoid.

I should add, whether I am writing for nonpublication a majority opinion or a dissent, the feeling is that because it is written, though first for the panel members, after them primarily for the parties, I can say things to or about them or others I would not shout to the world. [FN42] Also, the right to say what cannot be flung in one's face as a precedent afterwards could easily be misused, but in its proper place it is valuable. Of course, a judge has got to remember that nonpublication is far removed from assurance that what he writes is private. For the judiciary, the written word and privacy are opposite and incompatible concepts. It is not just us: no one in any official position, executive, legislative, or judicial, is free to write anything he would not wish to read in his favorite newspaper the next day.

IX.

Closely related to the rule of some circuits that if a dissenting opinion is filed in a case both opinions must be published, is the rule *926 that a panel which reverses (including to vacate or modify) a decision below must publish. [FN43] Thus, nonpublication is effectively limited to affirmances. If I am right in thinking that whatever the rules say, the primary reason for nonpublication is that the appeal had no reasonable foundation in the first place, evidently the reversal is an example of the opposite. It is a case where the appeal was vindicated.

Whatever the rules say, the proportion of reversed decisions left unpublished ought to be and usually is small. However, some circuits seem to take joy in not publishing a very high proportion of their reversals, notably the Sixth. [FN44] Yet to me it is rather a wooden rule to say that all reversals must be published. The circumstances that usually relate nonpublication to affirmances sometimes work in reverse. However, I think a panel which refuses to publish a reversal should realize it is doing something anomalous and should consider carefully whether it is doing the right thing.

X.

If the decision below by the trial court is published, that provides another factor to enter into the decision whether to publish the appellate review. While it is assumed that decisions of district judges reviewed in the regional circuits have usually not been published, most of the tribunals appealable to the Federal Circuit, that is, the Claims Court, Court of International Trade (CIT), Patent and Trademark Office (PTO), Merit Systems Protection Board (MSPB), and Contract Appeal Boards, have procedures in place for routine publication of their own decisions. [FN45] They all have special constituencies of attorneys who look to publication of their trial level tribunals as 'The Law' in all cases, most of them, not reversed, etc. on appeal. That the trial level decision, untouched on appeal, is 'The Law,' usually is a good reason for not publishing appellate affirmances, but the appellate court's selective publication policy must never be allowed to operate in a manner that leaves the trial bar in the dark as to what has happened to published decisions of their *927 tribunal recently appealed. Enough attention was not paid to this in the early days of the Federal Circuit and the result was bitter protest by the CIT bar, [FN46] and, I think, like protests by other bars would have been equally warranted.

One may hope, in cases of affirmances, that quarterly listing of unpublished decisions with disposition noted by the word 'affirmed' may suffice. Since, however, an appellate court may affirm if it finds any reason for doing so in the record, its grounds may be quite unlike those of the trial court. Even if the one word 'affirmed' is sufficiently informative, delay to the next quarterly listing may be unwarranted. According to the report by Stienstra, several circuits routinely publish if the decision below is published, whether they affirm or reverse. [FN47] Seemingly, we do not need this for affirmances. Except with the CIT, the practice of unpublished affirmances of published decisions has gone on with little or no protest, which I suppose means that those concerned have found a way to live with the practice. In cases of reversals, the mess can be much worse. The bloat of the Federal Reporter can be controlled even with published decisions by resort to summary per curiam affirmances or reversal when the full treatment is unwarranted. The selective publication policy does not require relegation to the quarterly listing of all cases not worthy of the elaborate adornments of the full-scale opinion. While the Federal Circuit does at any rate make full use of the quarterly listing, this is not true of all the circuits, [FN48] which I do not understand.

CONCLUSION

The Federal Circuit's decision to selectively publish its opinions was the only possible one in the circumstances. Factors to consider, besides the position of the Judicial Conference of the United States, were the practicalities of publishing court opinions, the limitations on the capacity of the human mind to absorb and manage data, the poverty of much material put before courts of appeal as subjects for legal discussion, the preferences of the judges as authors, and the absurdity of destroying forests to distribute masses of prolix and repetitious material. Objections on the ground of courts maintaining libraries or files of secret data, or of being tempted into careless writing, are not sustainable.

*928 The Federal Circuit selects proportionately more cases for nonpublication than do other federal appellate courts, but this is due to the peculiar composition of its jurisdiction. The standards for selection are more comprehensive than most, and allow, but do not require, a proper choice between publication and nonpublication in virtually every instance. They should require such a choice.

Special care is given in selecting for nonpublication split decisions, reversals of trial courts, and appellate reviews of decisions themselves published. The current prohibition against citing unpublished material should be maintained, and those materials should continue to be nonprecedential.

FNa Senior Circuit Judge, United States Court of Appeals for the Federal Circuit.

FN1. Pub. L. No. 97-164, 96 Stat. 25 (1982).

FN2. FED. CIR. R. 18. Rule 18 was amended as of Oct. 15, 1986. The amendment is consistent with the analysis contained in this article.

FN3. For an account of this development, see Reynolds & Richman, The Non- Precedential Precedent, 78 COLUM. L. REV. 1167, 1168-72 (1978) [hereinafter cited as Reynolds & Richman, Non-Precedential Precedent]. See also JUDICIAL CONFERENCE OF THE UNITED STATES REP. 33 (1972) (approving circulation to all circuit judges of detailed recommendation of Board of Federal Judicial Center concerning publication of opinions of courts of appeals). For the text of these recommendations, see BOARD OF FEDERAL JUDICIAL CENTER, RECOMMENDATION AND REPORT TO THE APRIL 1972 SESSION OF THE JUDICIAL CONFERENCE OF THE UNITED STATES ON THE PUBLICATION OF COURTS OF APPEALS OPINIONS 4-7 (1972).

FN4. Besides Reynolds & Richman, Non-Precedential Precedent, supra note 3, Reynolds and Richman also wrote under sponsorship of the Federal Judicial Center. See Reynolds & Richman, An Evaluation of Limited Publications in the United States Courts of Appeals: The Price of Reform, 48 U. CHI. L. REV. 573, 573-81 & nn.3-5 (1981) (analyzing costs and benefits of new publication policy) [hereinafter cited as Reynolds & Richman, The Price of Reform]. Other related writings are cited in their nn.3-5 and in D. STIENSTRA, UNPUBLISHED DISPOSITIONS: PROBLEMS OF ACCESS AND USE IN THE COURTS OF APPEALS n.3 (1985).

A few pages of comment appear in the recent and very able work of Seventh Circuit Judge Posner, The Federal Courts (1985) at 120. He regards selective publication as an unfavorable consequence of the litigation explosion and appears to have no hope of restoring nonselective publication until the explosion is brought under control. His assumption that nonpublication has a deleterious effect on opinion quality must be weighed in reference to his devastating analysis of the many faults in contemporary published opinions, largely due to the fact of publication itself and the posturing thereby induced.

FN5. See 1 Hearings Before the Commission on Revision of the Federal Court Appellate System, Second Phase 519-39 (1974-75) (statement of Judge Robert A. Sprecher, United States Court of Appeals for the Seventh Circuit) [hereinafter cited as Hearings on the Revision of Federal Courts]. The nature of this Commission is explained in D. STIENSTRA, supra note 4, at n.21.

FN6. By designation under 28 U.S.C. § 291 (1982).

FN7. See supra notes 3 and 4 and accompanying text (noting academic commentary discussing publication policies).

FN8. No published critique has been found expressly stating this, but no writer postulates any limits.

FN9. For example, the Bureau of National Affairs prints brief descriptions of unpublished opinions in their Weekly Advance Sheets Highlights and, occasionally, prints or excerpts from an otherwise unpublished opinion in their Patent Trademark & Copyright Journal or their Federal Contracts Report. See, e.g., United States v. Manistique Tool & Co., 45 Fed. Cont. Rep. (BNA) 744 (Fed. Cir. per Nichols, J. April 14, 1986); J. W. Bateson Co. v. United States, 45 Fed. Cont. Rep. (BNA) 434 (Fed. Cir. Mar. 10, 1986); DiMarco Corp. v. United States, 45 Fed. Cont. Rep. (BNA) 556 (Fed. Cir. Mar. 31, 1986). In addition, the electronic services, Westlaw and Lexis, publish opinions. Westlaw will publish everything the courts supply and will track the various looseleaf BNA services to locate decisions not given them by the courts. Lexis also publishes all the courts supply but only to the degree the court allows. For example, at the request of the Fourth Circuit, Lexis does not publish the text of designated unpublished opinions.

FN10. I M. BENNETT, THE UNITED STATES COURT OF CLAIMS HISTORY 40 (1976).

FN11. The final volume, 231 Ct. Cl. (1982), which includes decisions from June 5 to September 30, 1982, did not appear in print until November 15, 1984. It includes 345 pages of decisions in 'Order' form never previously in print. For a discussion of Orders, see infra, pages 13-14.

FN12. Beginning with the April 1910 to April 1911 volume, the official reports were published in the Court of Customs Appeals Reports. In 1929, that court became the Court of Customs and Patent Appeals and published two official reports. Each series was titled Court of Customs and Patent Appeals Reporter, but one dealt exclusively with patents and the other with customs. The last issue of the patent reporter was volume 59, for the years October 1971 to October 1972; the last issue of the customs reporter was volume 69, October 1981 to October 1982.

FN13. The first year I joined the stable of judicial authors for the Federal Reporter with Chernick v. United States, 372 F.2d 492 (Ct. Cl. 1967), West published 16 volumes of that series, 1,000 pages each roughly, say 16,000 pages. My last year on the Court of Claims, 1982, I finished my cases for that court with Estate of Berg v. United States, 687 F.2d 377 (Ct. Cl. 1982). That year West published 29 volumes of that series, 1,400 pages average, not counting index, say 40,600 pages. This was when the Judicial Conference selective publication plans had been several years in effect. Without them, one can only guess what the figure might have been.

FN14. 25 U.S.C. § 70 (repealed 1978).

FN15. Hearings on the Revision of Federal Courts, supra note 5, at 536 (statement of Judge Robert A. Sprecher, United States Court of Appeals for the Seventh Circuit); see Reynolds & Richman, Non-Precedential Precedent, supra note 3, at 1197 (discussing Judge Sprecher's testimony before Hruska Commission).

FN16. D. STIENSTRA, supra note 4, at 11; Reynolds & Richman, Non-Precedential Precedent, supra note 3, at 1197.

FN17. Some circuit courts send prospective visiting judges volumes of briefing materials.

FN18. In Finkbohner v. United States, 788 F.2d 723 (11th Cir. 1986), the government cited to an unpublished Fifth Circuit case, a proper practice in the Eleventh Circuit which allows citation of unpublished opinions as precedent.

FN19. See, e.g., Reynolds & Richman, Non-Precedential Precedent, supra note 3, at 1200 (expressing concern that judges will not expend sufficient energy on unpublished cases). Posner, supra note 4, voices the same concern.

FN20. Id. (suggesting nonpublication rules may increase inconsistency in unpublished decisions).

FN21. The argument of Reynolds and Richman, id. at 1201, seems to be that an incompetent judge should be compelled to expose his incompetence by writing for a learned readership.

FN22. Posner, supra note 4, at 301, estimates that about three percent of published federal circuit court decisions receive notice and comment in the academic law reviews. Some others are discussed elsewhere, e.g., in publications of professional societies.

FN23. Reynolds & Richman, The Price of Reform, supra note 4, at 581.

FN24. D. STIENSTRA, supra note 4, at 23.

FN25. United States v. Cagnina, 697 F.2d 915, 923 (11th Cir.), cert. denied, 464 U.S. 856 (1983) (citing United States v. Ellis, 547 F.2d 863, 868 (5th Cir. 1977)).

FN26. See Reynolds & Richman, Non-Precedential Precedent, supra note 3, at 1181 (describing Fourth Circuit's procedures). See Posner, supra note 4, at 122.

FN27. See id. (describing arrangements previously existing in Tenth Circuit).

FN28. See supra text pp. 5-7 (discussing history of West's refusal to publish many decisions submitted for publication).

FN29. See supra text accompanying note 2 (text of Rule 18) and infra text pp. 22-23 (text of ¶¶5 & 6 of Standard Operating Procedure No. 7).

FN30. See Reynolds & Richman, Non-Precedential Precedent, supra, at 1195-96, 1199 (discussing problem of unequal access to unpublished opinions).

FN31. Rule 47.6 of the Fifth Circuit Court of Appeals reads as follows:

Affirmance Without Opinion

When the Court determines that any one or more of the following circumstances exists and is dispositive of a matter submitted to the Court for decision: (1) that a judgment of the District Court is based on findings of fact which are not clearly erroneous, (2) that the evidence in support of a jury verdict is not insufficient, (3) that the order of an administrative agency is supported by substantial evidence on the record as a whole; and the Court also determines that no error of law appears and an opinion would have no precedential value, the judgment or order may be affirmed or enforced without opinion.

In such case, the Court may in its discretion enter either of the following orders:

'AFFIRMED. See Loc.R. 47.6.' or 'ENFORCED. See Loc.R 47.6.'

See N.L.R.B. v. Amalgamated Clothing Workers of Am., 430 F.2d 966, 971-72 (5th Cir. 1970) (discussing criteria for publication and importance of this determination).

FN32. See, e.g., Reynolds & Richman, The Price of Reform, supra note 4, at 603-24 (criticizing summary affirmance policy as decision-making without justification). See also Commission on Revision of the Federal Court Appellate System, Structure and Internal Procedures: Recommendations for Change 50-52 (1975) (recommendation regarding opinion writing and publication policy), reprinted in 67 F.R.D. 195, 256-59 (1975).

FN33. See Hearings on the Revision of Federal Courts, supra note 5, at 826 (statement of Judge William E. Doyle, United States Court of Appeals for the Tenth Circuit); see also id. at 1107 (statement of Judge Byron G. Skelton, United States Court of Claims).

FN34. Id. at 1361-62.

FN35. D. STIENSTRA, supra note 4, at 40 (Table 2).

FN36. Id. at 37.

FN37. United States Court of Appeals for the Federal Circuit, Standard Operating Procedures (SOP) No. 7, Publication Policy, Opinions and Orders ¶¶5- 6 (July 23, 1984).

FN38. See D. STIENSTRA, supra note 4, at 28-39 (analyzing publication rules and practices). For example, the Seventh Circuit did not publish decisions on remand by the Supreme Court. Comment, A Snake in the Path of the Law, 39 U. PITT. L. REV. 309, 316 (1977). The original model rule put forward in 1978 had only four items. See Reynolds & Richman, Non-Precedential Precedent, supra note 3, at 1171 (quoting text of model rule).

FN39. See D. STIENSTRA, supra note 4, at 32 (noting that Second, Fifth, and Federal Circuits require unanimity for nonpublication).

FN40. See generally Comment, supra note 38 (offering sweeping attack on sampling of identified Seventh Circuit unpublished opinions (orders) on ground they should have been published).

FN41. D. STIENSTRA, supra note 4, at 37; Reynolds & Richman, The Price of Reform, supra note 4, at 614.

FN42. Reynolds & Richman, The Price of Reform, supra note 4, at 610, criticize nonpublication in some instances where defective working of the legal system has occurred, for example, an eight-year delay in an injured longshoreman getting compensation. Publication of the stark details, however, would probably get the usual public nonnotice, not reform. Yet persons apparently responsible, but not parties to the case, might be pilloried with no chance to defend themselves. In such an instance, nonpublication might permit details to be explored with less unfairness to persons whose acts might appear to their disadvantage. For example, in a U.S. government employee's 'adverse action' case, the employee frequently seeks to portray his supervisor as an unmitigated scoundrel, and sometimes the opinion adopts and reflects this view. See, e.g., Sullivan v. United States, 720 F.2d 1266 (Fed. Cir. 1983). Yet the supervisor is not a party and is not represented by counsel.

FN43. See D. STIENSTRA, supra note 4, at 34 (discussing rules in some circuits requiring publication of all reversals).

FN44. Id. at 41-43.

FN45. Claims Court-United States Claims Court Reporter published by West Publishing Co.; CIT-United States Court of International Trade Reports official reports published by Government Printing Office and advance sheets of Customs Bulletin published by Customs Service; PTO-United States Patent Quarterly published by Bureau of National Affairs, Inc.; Contract Appeal Boards-Board of Contract Appeals Decisions published by Commerce Clearing House, Inc.; MSPB- Merit Systems Protection Board Reporter published by West Publishing Co.

FN46. Second Annual Federal Circuit Judicial Conference, 104 F.R.D. 207, 222-30 (1984).

FN47. D. STIENSTRA, supra note 4, at 33.

FN48. See id. at 21 (noting that six circuits only submit lists of cases decided on merits, while three circuits include all unpublished decisions).

END OF DOCUMENT