The ethical test of a judge is not whether his judgments run parallel to the judgments of a moralist, but whether the judge administers his office true to its traditional limitations as well as to its aspirations. From the day he takes his seat he is aware that while he has more personal discretion than the books reveal, he is hemmed in by impersonal usages, canons, and legitimate expectations. While he has choice, he cannot exercise it even to his own satisfaction unless it is disciplined according to standards. The minima are supplied by reversals administered by appellate courts. Those, however, are necessarily negative in nature. What counts more is the establishment of affirmative norms of judicial behavior. One man knows the practices of his own and perhaps a few other courts; so, to evolve standards, he must become critical of his own shortcomings, attentive to the reactions of the bar, informed of the unrecorded practices of his colleagues, and, above all, reflective about subtle differences in the tasks assigned to him.
The trial judge's first problem is his relationship to the jury. Much of the debate about the jury system rests on political premises as old as the eighteenth century. Montesquieu, Blackstone, and their followers contended that lay tribunals with a plurality of members were the safeguard of liberty. Bentham and more modern reformers replied that when the rule of law itself is sound, its integrity requires that its application be entrusted to magistrates acting alone. In their view, responsibility is the secret of integrity, and a reasoned choice is the secret of responsibility.
Experience will not give a sovereign answer to these warring contentions. Yet the disagreement can be narrowed if the question of the jury's utility is subdivided with specific emphasis on separate types of suits.
The importance of this subdivision is concealed by Chief Justice Hughes's striking phrase that a federal judge is the "governor of the trial." Some regard this as an implied acceptant of the practice of English courts. They construe it as a broad invitation to exercise in all types of cases a right to comment upon the evidence, provided of course that the judge reminds the jury in his charge that they are not bound to follow the court's view of the facts or the credibility of the witnesses.
Such boldness is not the surest way to end disputes in defamation cases. In 1944 a discharged OPA official brought a libel suit against the radio commentator, Fulton Lewis, Jr. At one stage in the examination I suggested that Mr. Lewis's counsel was throwing pepper in the eyes of the jury; and at the final summation I indicated plainly that, although the jury was free to reject my opinion, I thought Mr. Lewis had been reckless in his calumnious charges against the ex-OPA official. It makes no difference whether what I said was true; I should not have said it, as the reaction of the bar and public reminded me. A political libel suit is the modern equivalent of ordeal by battle. It is the means which society has chosen to induce bitter partisans to wager money instead of exchanging bloody noses. And in such a contest the prudent and the second-thinking judge will stand severely aside, and will act merely as a referee applying the Marquis of Queensberry rules.
Later a libel suit was brought by James Michael Curley against the publisher of the Saturday Evening Post for having said that Mr. Curley was a Catholic of whom His Eminence Cardinal O'Connell would have no part. Who knew better than the Cardinal whether that charge was true ? Mr. Curley, the plaintiff, did not call the Cardinal to the stand. The defendant's distinguished counsel did not desire to find out what would be the effect upon a Greater Boston jury if a Protestant lawyer should call a Catholic prelate to the witness stand. Should the court have intervened and summoned the Cardinal on its own initiative? The Fulton Lewis case gave the answer. In a political libel suit the judge is not the commander but merely the umpire.
Those cases which involve sordid family disputes also are better left to the jury without too explicit instructions. Plato implied and Holmes explicitly stated that judges are apt to be naive men. If judges seem to comment on the morality of conduct or the extent of damages, they may discover that the jurors regard their own knowledge as superior to the judges'. At any rate when brother sues brother or when spouse sues paramour, the very anonymity of the jury's judgment often does more than the most clearly reasoned opinion or charge of an identified judge could have done to still the controversy.
What of the trial judge's role in accident cases? How far should he go in requiring available evidence to be produced, in commenting on the testimony, and in using special verdicts and like devices to seek to keep the jury within the precise bounds laid down by the appellate courts?
There are some who say that the trial judge has not fulfilled his moral obligation if he merely states clearly the law regarding negligence, causation, contributory fault, and types of recoverable damage. In their opinion it is his duty to analyze the evidence and demonstrate where the evidence seems strong or thin and where it appears reliable or untrustworthy.
Most federal judges do not make such analyses. They are not deterred through laziness, a sentimental regard for the afflatus of the Seventh Amendment, or even a fear of reversal. They are mindful that the community no longer accepts as completely valid legal principles basing liability upon fault. They perceive a general recognition of the inevitability of numerous accidents in modern life, which has made insurance widely available and widely used. Workmen's compensation acts and other social and economic legislation have revealed a trend that did not exist when the common-law doctrines of tort were formulated. And the judges sense a new climate of public opinion which rates security as one of the chief goals of men.
Trial judges cannot, without violating their oaths, bow directly to this altered policy. In instructions of law they must repeat the doctrines which judges of superior courts formulated and which only they or the legislatures can change. But trial judges are not giving "rein to the passional element of our nature" or forswearing themselves by following Lord Coke's maxim that "the jurors are chancellors." Traditionally juries are the device by which the rigor of the law is modified pending the enactment of new statutes.
Some will say that this abdication is not merely cowardly but ignores the "French saying about small reforms being the worst enemies of great reforms." To them the proper course would be to apply the ancient rules with full rigidity. They anticipate that adverse reactions would then lead to a complete resurvey of accident law; to a scrutiny of the costs, delays, and burdens of present litigation; to a comparative study of what injured persons actually get in cash as a result of lawsuits, settlements out of court, administrative compensation proceedings, and other types of insurance plans; and ultimately to a new codification. To this, one answer is that in Anglo-American legal history reform has rarely come as a result of prompt, comprehensive investigation and legislation. The usual course has been by resort to fictions, to compromises with logic, to juries. Only at the last stages are outright changes in the formal rules announced by the legislators or the appellate judges. This is consistent with Burke's principle that "reform is impracticable in the sense of an abrupt reconstruction of society, and can only be understood as the gradual modification of a complex structure."
I am not at all certain that it would be a desirable reform in those personal injury controversies, known as tort cases, to substitute trial by judges for trial by juries. Just such a substitution has been made in the Federal Tort Claims Act. And experience under that statute does not prove that in this type of case a single professional is so satisfactory a tribunal as a group of laymen of mixed backgrounds. In estimating how a reasonable and prudent man would act, judges' court experience counts for no more than juries' out-of-court experience. In determining the credibility of the type of witness that appears in accident cases an expert tribunal is somewhat too ready to see a familiar pattern. Shrewdness founded on skepticism and sophistication has its place in scrutinizing the stories of witnesses. But there is a danger that the professional trier of fact will expect people of varied callings and cultures to reach levels of observation and narration which would not be expected by men of the witness's own background. Moreover, when it comes to a calculation of damages under the flexible rules of tort law the estimate of what loss the plaintiff suffered can best be made by men who know different standards of working and living in our society. Indeed, I have heard federal judges confess that in a Federal Tort Claims case they try to make their judgments correspond with what they believe a jury would do in a private case. And many judges would prefer to have such cases tried by juries.
In commercial eases and those arising under regulatory statutes there is reason to hold a jury by a much tighter rein than in tort cases. This is not because the rules of law are more consonant with prevailing notions of justice. In these controversies judges have a specialized knowledge. Parties have usually acted with specific reference to their legal rights. Departures from the declared standard would undermine the legislative declaration and would be less likely to produce reform than confusion and further litigation.
An extreme example will serve as an illustration. In a tax case tried before a jury at the suit of one holder of International Match Company preference stock, the issue was whether for tax purposes those certificates had become worthless in the year 1936. In another taxpayer's case the Second Circuit Court of Appeals had Affirmed a ruling of the Board of Tax Appeals that similar stock had become valueless in the year 1932. Technically this adjudication did not bind the jury, though the evidence before it was substantially the same as that in the earlier case. To preserve uniformity on a factual tax problem of general application I had no hesitation in strongly intimating to the jury that they should reach the same result as the Second Circuit.
In sales cases something close to a scientific appraisal of the facts is possible; there are strong mercantile interests favoring certainty; and future litigation can be reduced by strict adherence to carefully prescribed statutory standards. These considerations sometimes warrant giving juries written instructions or summaries and often justify the use of special verdicts. Either method makes jurors focus precisely on the formalities of the contract, the warranties claimed to have been broken, the types of damage alleged to have been sustained, and the allowable formulae for calculating those damages.
Indeed, except for tort cases, I find myself in agreement with Judge Frank that the trial judge ought to use special verdicts to a much larger extent, though it is more difficult than may at first be realized to frame questions to the satisfaction of counsel and to the comprehension of juries. Once when I used what I thought simple questions, a fellow judge, half in jest, accused me of trying to promote a disagreement of the jury and thus to force a settlement.
The arguments supporting special verdicts in commercial or statutory cases also support a trial judge in giving in such cases a more detailed charge and more specific guidance in estimating the testimony. In complicated cases or those in fields where the experience of the average juror is much less than that of the average judge, there is a substantial risk of a miscarriage of justice unless the judge points rather plainly to the "knots" in the evidence and suggests how they can be unraveled. The only time I have ever entered judgment notwithstanding a verdict was in a private antitrust suit. The jury had awarded damages of over one million dollars as a result, I believe, of the generality of my instructions.
So far I have said nothing of federal criminal cases. About 90 per cent of all defendants in the federal court plead guilty. In those federal cases which come to trial the crime charged frequently concerns economic facts; and generally, though not invariably, the preliminary investigation by the FBI and other agencies of detection has reduced to a small compass the area of doubt. Often the only remaining substantive issue of significance is whether the defendant acted "knowingly." Indeed the usual federal criminal trial is as apt to turn on whether the prosecution has procured its evidence in accordance with law and is presenting it fairly, as on whether the defendant is guilty as charged. All these factors combine to concentrate the judge's attention upon the avoidance of prejudicial inquiries, confusion of proof, and inflammatory arguments. Counsel can aid the judge to maintain the proper atmosphere by stipulation, by refraining from putting doubtful questions until the judge has ruled at the bench, and by other cooperative efforts. But if cooperation is not forthcoming the judge should hesitate to fill the gap by becoming himself a participant in the interrogation and should not indicate his view of the evidence. For the criminal trial is as much a ceremony as an investigation. Dignity and forbearance are almost the chief desiderata.
I turn now to the freedom of the trial judge when he sits without a jury. In nonjury as in jury cases, a substantial part of the bar prefers to have the judge sit patiently while the evidence comes in and then at the end of the trial summarize the testimony. This seems the sounder practice in the great bulk of trials. But in cases of public significance, Edmund Burke admonished us: "It is the duty of the Judge to receive every offer of evidence, apparently material, suggested to him, though the parties themselves through negligence, ignorance, or corrupt collusion, should not bring it forward. A Judge is not placed in that high situation merely as a passive instrument of parties. He has a duty of his own, independent of them, and that duty is to investigate the truth...." Let me give some examples of when I believe the judge has a duty to elicit facts in addition to those that are offered by the parties. The plaintiff, an owner of a multiple dwelling, brought suit for a declaratory judgment seeking to have the premises declared a "hotel" and thus exempt from the rent regulations of the OPA. Only one of the numerous tenants was named as defendant. In the trial the plaintiff offered evidence that showed the building was a hotel and not an apartment. Because of lack of funds or of forensic skill, the tenant's counsel failed to shake the stories of the plaintiff's witnesses or to offer adequate testimony to the contrary. Yet if the trial judge had called specialists and others familiar with the community and the property, the evidence would have demonstrated that in truth the building was a mere apartment house. I took no step myself to call witnesses or to interrogate those who did testify but, relying exclusively on what the parties offered, entered a judgment declaring the premises a "hotel" and thus exempt. Since this declaration of status became in effect a general rule binding on scores of persons not represented in the proceedings, would it not have been sounder if the court had taken a larger initiative in seeing that the record corresponded with reality?
A later controversy of even greater public importance posed a similar problem. In a case still undecided, the United States sued the United Shoe Machinery Corporation for violation of the antitrust laws. Among the issues that were presented was the effect of the corporation's acts upon its customers and its competitors. The Government in its case relied exclusively on the corporation's documents and officers. The corporation planned to call some customers, though the method by which they were drawn was not disclosed to the court. This seemed an inadequate survey. So the court asked the parties to take depositions from forty-five customers, selected from a standard directory by taking the first fifteen names under the first, eleventh, and twenty-first letters of the alphabet; and the court itself called to the stand the officers of the principal competitor. In the summons the court listed topics appropriate for questioning the officers. The actual examination was conducted in turn by the competitor's counsel, the Government's counsel, and the defendant's counsel. Both these types of testimony give a much clearer understanding of the total picture of the industries that will be affected by any decision.
Another problem in the United Shoe case has been to determine what have been the usual methods followed by the defendant in setting prices, in supplying services, and in suing competitors. An adequately grounded conclusion can hardly be based entirely on the plaintiff's selection of a few dramatic incidents and on the defendant's testimony of the general attitude of its officers. The critical point in determining liability and the form of relief may turn on what has been the typical pattern of the defendant's conduct and the typical effect of that conduct on outsiders. Here the judge can perform a useful function if he, through pre-trial conferences or at a later stage of the litigation when he is more aware of its dimensions, provides for appropriate samplings of the conduct and the effects. If the judge is fortunate, the parties may agree on the sampling. But where they do not, it seems to me to be the judge's responsibility first to elicit from witnesses on the stand the criteria necessary to determine what are fair samples and then to direct the parties to prepare such samples for examination and cross-examination. Sampling will make the record not merely more informative but shorter.
The question as to what has been the custom of the market and what would be the consequence of a judicial decree altering those practices arises not only in antitrust cases but also when the judge is faced with the problem of determining either the appropriate standard of fair competition in trademarks or the appropriate standard for fiduciaries. Usually, to be sure, diligent counsel offer in evidence enough relevant material. But where this has not been done, there have been times when a judge has tended to reach his result partly on the basis of general information and partly on the basis of his studies in a library.
This tendency of a court to inform itself has increased in recent years following the lead of the Supreme Court of the United States. Not merely in constitutional controversies and in statutory interpretation but also in formulation of judge-made rules of law, the justices have resorted, in footnotes and elsewhere, to references drawn from legislative hearings, studies by executive departments, and scholarly monographs. Such resort is sometimes defended as an extension of Mr. Brandeis's technique in Muller v. Oregon. In Muller's case, however, Mr. Brandeis's object was to demonstrate that there was a body of informed public opinion which supported the reasonableness of the legislative rule of law. But in the cases of which I am speaking these extra judicial studies are drawn upon to determine what would be a reasonable judicial rule of law. Thus the focus of the inquiry becomes not what judgment is permissible, but what judgment is sound. And here it seems to me that the judge, before deriving conclusions from any such extrajudicial document or information should lay it before the parties for their criticism.
How this criticism should be offered is itself a problem not free from difficulty. In some situations the better course may be to submit the material for examination, cross-examination, and rebuttal evidence. In others, where expert criticism has primarily an argumentative character, it can be received better from the counsel table and from briefs than from the witness box. The important point is that, before a judge acts upon a consideration of any kind, he ought to give the parties a chance to meet it. This opportunity is owed as a matter of fairness and also to prevent egregious error. As Professor Lon Fuller observed, the "moral force of a judgment is at maximum if a judge decides solely on the basis of arguments presented to him. Because if he goes beyond these he will lack guidance and may not understand interests that are affected by a decision outside the frame work."
The duty of the judge to act only on the basis of material debated in pubic in no sense implies that the judge's findings should be in the precise terms offered by counsel. Nor does Rule 52(a) of the Federal Rules of Civil Procedure require the judge always to recite all relevant evidence and to rely for persuasive effect exclusively upon mass and orderly arrangement. Yet in corporate cases or other litigation where the issues turn on documentary analysis and precise analysis of business details, and where appeal is almost certain to be taken, the trial judge may perform the greatest service by acting almost as a master summarizing evidence for a higher tribunal.
On the other hand, if a judge sitting alone hears a simple tort or contract case falling within a familiar framework and analogous to jury litigation, it is perhaps the best practice for him to state his findings of fact from the bench in those pungent colloquial terms with which the traditional English judge addresses the average man of common sense. When credibility of witnesses is the essence of the controversy, the parties and the lawyers like to have judges act as promptly as juries and on the basis of fresh impressions.
Where the search for truth is more subtle, the trial court faces the same stylistic challenge as the appellate court. Fortunate are those who, like Judge Learned Hand, have the gift of many tongues. His admiralty opinions breathe salt air, his commercial cases echo the accents of the market place, and his patent rulings reflect an industrial society developed by Yankee ingenuity. Even those whose narrower experience makes them stutter, occasionally strike a subject where they have both the sensitivity and the self-confidence to put the story simply and selectively. But in most cases we average judges can only try, without much hope, to make our summations of facts pithy, sympathetic, and illuminating.
While, in summarizing the facts, we trial judges may seek to imitate our superiors on the higher courts, when we wrestle with the substantive law we should not regard ourselves as the appellate judges writ small. Our freedom is inevitably more narrowly exercised. Most of the time we do not see the points of difficulty too clearly. With us the pace is quicker, the troublesome issues have not been sorted from those which go by rote, the briefs of counsel have not reached their ultimate perfection. Yet even when we have the clearest perception of the legal issues, certain inhibitions are peculiarly appropriate to restrain a judge who sits alone and subject to review by judges higher in commission.
If the trial judge is presented with the claim that a legislative act is unconstitutional, he ought to remind himself that every possible presumption is in favor of the validity of the legislation and that in certain constitutional controversies a district judge has no jurisdiction to act unless he is sitting with two other judges. Though in a constitutional case or any other case he must not surrender his deliberate judgment and automatically accept the views of others, he can ordinarily best fulfill his duty in a constitutional case by explicitly stating for the benefit of an appellate court any doubts he has, without going so far as to enter a decree against a statute which has commanded the assent of a majority of the legislature and, generally, of the executive.
If there is no constitutional question and the trial judge is presented with a judicial precedent or precedents contrary to his own view of what would be the sound rule of law, the problem is more subtle. First, take the situation where the hostile precedents are in the tribunals that sit on review of his own decisions. If the precedents have been so severely impaired by recent cases that it is reasonably clear they no longer represent the present doctrine of the appellate court, the trial judge is generally thought to be free to minimize their directive force, though there is strong opinion to the contrary. Where the precedent has not been impaired, the balance is in favor of the trial judge following it in his decree and respectfully stating in his accompanying opinion such reservations as he has. The entry of the decree preserves the "priority and place" which Shakespeare reminded us were indispensable to justice. Moreover, the reservation in the opinion promotes the growth of the law in the court where it most counts, for if the criticism of the precedent be just, the appellate court will set matters straight, and any trial judge worthy of his salt will feel complimented in being reversed on a ground he himself suggested. No trial judge of any sense supposes his quality is measured by a naked tabulation of affirmances and reversals.
Where the hostile precedents come from a judge of equal rank or a court not in the direct line of superior authority, I doubt whether there should be absolute rules of deference. If the precedent is from a sitting judge in one's own court and represents his mature reflection, the argument in favor of following it rests not only on the appropriate amenities, but also on profounder considerations of equality in the treatment of litigants. But the situation is different where the precedent comes from an inferior court sitting in another geographical area. In the federal system conflict of judgments between the inferior courts is one of the ways that the Supreme Court is led to grant review of legal questions. And the most effective method of getting a significant issue over the Washington threshold is to challenge overtly a court in another circuit.
We federal judges are told that in diversity jurisdiction cases our duty is to follow the state law. Most of the time that is readily discoverable. But what are we to do when no state law has been declared, or the state law has not been the subject of reconsideration for a generation or more? Take unfair competition cases, at least before the Lanham Act. Until the end of the rule of Swift v. Tyson the state law lay relatively dormant. Most of the important controversies in this field had always been adjudicated in the federal courts according to a general jurisprudence. What happens when these federal cases are not binding authorities? Shall we seek to evolve the state rules exclusively from state precedents, some of which are quite old, and ignore the federal precedents?
Shall we be equally conservative in corporation cases? A short time ago a policyholder brought a derivative suit in the United States District Court for the District of Massachusetts against an insurance company without first seeking to enlist the aid of his fellow policyholders. The reported Massachusetts cases involved stockholders' suits. None of them was precisely in point. Some of the rulings were not addressed to considerations recently stressed by other courts and by legislatures and administrative agencies. Should the federal court follow closely what the state has already said, or should it keep one eye on the national trend? Or look at the case of a stockholder seeking to procure an equity receivership for the purpose of liquidating a corporation. The only Massachusetts decisions are old and negative. The modern trend is favorable. Shall the federal court assume that the Massachusetts state court will follow its predecessors or its contemporaries?
The impression that I gather from the cases is that a federal judge sitting in a diversity jurisdiction case is less willing to depart from obsolete doctrines than when he sits in a purely federal case. Every time judges are called upon to apply the law of a foreign jurisdiction, are they not inclined to give undue weight to the recorded landmarks and to underestimate the mobile qualities and the thrusts of principle we discern in our domestic law?
And now, before I conclude, may I address myself to a doubt which should perhaps have been tackled at the outset. Are the usages followed by trial judges more than patterns of behavior; are they law in any sense; and even if they are law, are they too disparate and detailed ever to have an honored place in the study of jurisprudence?
Concede that the normative practices which we have been reviewing fall far short of the Austinian command of the sovereign. For a judge who chooses to depart from these particular standards does not lay himself open to reversal by courts of superior authority. And yet that which is generally approved as being good and being within the reach of average men does in time become law in the strictest sense. This, we all know, is how the law of fiduciaries and the law merchant have grown. And the principle applies in equal measure to the law governing trial judges. What is the whole law of procedure but the crystallization of judicial custom? The trial judges made the law of evidence by their usages; and perhaps now they are unmaking it by their usages. The revocation is hidden by appellate courts which treat departures from the proclaimed evidentiary rules not as though they represented new doctrine, but as though they were insignificant nonreversible errors.
What are the rules governing measure of proof? Today we say there exist in the federal courts only two standards: the criminal standard of proof beyond a reasonable doubt and the civil standard of the preponderance of the evidence. And yet already in some special classes of cases where fraud is the central issue, we seem to see the emergence of an intermediate rule, the requirement that the evidence shall be clear and convincing. This intermediate requirement reflects the unspoken practice of trial courts to move with extreme caution in fastening a finding of immoral conduct upon a party litigant.
What shall we say of remedies which trial judges have newly evolved in equitable suits founded on statutes? Novel remedies begin as permissible exercises of discretion by the court of first instance. They win approval and imitation by other similarly circumstanced courts. And in the end what was discretionary has become mandatory. Here is the common law at work—a progressive contribution by the judges, trial as well as appellate; less important perhaps today than formerly, and always less important than the additions made by legislative bodies; but more clearly ethical in its nature because the consent on which it rests has undergone a longer, more intimate, more pragmatic test.
Let us not suppose that because our jurisdiction is limited, because so much of our work goes unreported, because we are immersed in the detail of fact, we trial judges are clothed with small responsibility in relating law to justice. It is we who make the law become a living teacher as we transmit it from the legislature and the appellate court to the citizen who stands before us. It is we who watch the impact of the formal rule, explain its purpose to laymen, and seek to make its application conform to the durable and reasonable expectations of our communities. It is we who determine whether the processes of common-law growth shall decay or flower with a new vigor.