Before SWAN, Chief Judge, LEARNED HAND and FRANK, Circuit Judges.
The Board's decision and order, reported in 96 N.L.R.B. No. 215, state the facts sufficiently.
FRANK, C.J.: 1. We think that, on the record as a whole, the evidence supports the findings of fact which in turn justify the Board's legal conclusions and order. Respondent offered no proof concerning the percentage of its employees who were union members on July 21, 1950. Absent such proof, the fact that, in the circumstances, more than 90% of those discharged on that date were members of the union suffices to make not unreasonable the Board's inference that respondent discriminated against union members, and that the discharges on that particular date were not caused by respondent's fear (engendered by President Truman's July 19 speech) that respondent's production of certain civilian goods would have to be drastically curtailed in favor of defense production - especially as, within a week of the election held on August 15 and at which the union succeeded, respondent began to hire back some of the discharged union employees and soon rehired almost all of them.
2. However, one employee, Tennent, although an active union member, was not discharged until July 28, a week later than the others. Holland, respondent's vice-president, testified that the sole reason for Tennent's discharge was his signal inefficiency or carelessness. According to Holland, when he learned in Baltimore on July 28th that a customer of respondent had received from it more than 6,000 defective transformers, he telephoned from Baltimore to respondent's plant directing the discharge; this he did, he testified, because the defects resulted from Tennent's failure to "set up" the machines on which the transformers were wound.Were this testimony believed, Tennent's discharge would not have violated the Act.
But there was also this testimony by Tennent: Ayers (whom the Board, on sufficient evidence, found to be Holland's "right hand man") on July 28 "just a couple of minutes before quitting time" handed Tennent a discharge slip, and told Tennent he did not know the reason for the discharge. The discharge slip, identical in wording with those which had been handed to the employees dicharged on July 21, read as follows:
"Due to changes in our production requirements, it is necessary that we reduce our work force. Therefore, you are hereby notified that your employment with the Company is terminated."
On December 5, 1950, Tennent, at respondent's request, returned to his former job with respondent. On February 6, 1951, he received an increase in pay. Moreover, Holland also testified that, if Tennent followed erroneous specifications, he would not have been responsible for the defective transformers; and that these transformers, after leaving Tennent's hands, had been subjected to "probably in the vicinity of between four and six inspections" for the purpose of discovering "just the thing * * * that failed in this job."
Respondent argues that there is nothing in this testimony which cannot reasonably be reconciled with Holland's testimony about the reasons for this discharge.*fn1 Whether, on that basis, we would over-turn the finding adverse to respondent we need not consider. For all the testimony was given orally before the Trial Examiner who stated in his report:
"On the entire record, including his observation of the witnesses, the undersigned is not persuaded that Tennent was discharged by the Respondent for the reasons advanced by it. The undersigned does not credit Holland's testimony to the effect that he ordered Tennent's discharges because defective material had been made in and shipped from the respondent's plant."
The Board adopted this finding. If it stands, then Holland's testimony must be ignored. On that basis, we cannot say that the Examiner and the Board did not have ample evidence to support their conclusion that Tennent's union activities were the reason for his discharge: Although he was fired a week later than the other union members, the firing occurred before the election; when the company notified him of his discharge on July 28, his inefficiency or negligence was not assigned as a reason;*fn1a despite this alleged inefficiency, he was later rehired and subsequently his pay was increased. These facts constitute a sufficient foundation for a rational inference that Tennent's union activity induced the discharge.
If, in similar circumstances, a trial judge made such a finding, we would be obliged to accept it. For the pivotal factor here is the Examiner's disbelief in Holland's testimony, a disbelief that rested on an evaluation of Holland's credibility, which in turn the Examiner founded upon "his observation of the witnesses." Repeatedly, the courts have said that, since observation of such "demeanor evidence" is open to a trier of the facts when witnesses testify orally in his presence, and since such observation is not open to a reviewing tribunal, that fact-trier's findings, to the extent that they comprise direct or "testimonial" inferences,*fn2 are ordinarily unreviewable. True, demeanor evidence may sometimes mislead; but our courts regard it nevertheless as an excellent clue to the trustworthiness of testimony. The federal civil procedural Rules reflect this view.*fn3
It has had a long history.In the earlier period of Roman legal development, according to Millar, the witnesses testified orally before the judex, and the practice of having oral testimony heard by the judge prevailed originally in the Romanocanonical procedure.*fn3a Ullman tells us that the 14th century Postglossators - who, as judges or advocates, "had their eyes fixed upon the practical administration of the law" - maintained that the "indispensable requisite for the judge to form his opinion on the trustworthiness of witnesses was that they appeared before him personally * * * The personal impressions made upon the judge by the witnesses, their way of answering questions, their reaction and behavior in court, were the only means of ascertaining whether the statements were trustworthy or not * * * It was thought necessary, therefore, that the judge * * * should put on the record in the files any specific reactions, e.g., that the witness stammered, hesitated in replying to a specific question, or showed fear during the interrogation * * *"*fn3b Subsequently, however written testimony became in general the norm in canon and lay continental courts until the 19th century.*fn3c In English Chancery it came about that the "canon law influence prevented the oral examination of witnesses save as an extraordinary measure", while at English common law the testimony was oral.*fn3d For the most part, America inherited this difference between chancery and common law procedures. In the federal courts, except for a short period from 1789 to 1802, oral testimony in open court was not required in equity litigation; indeed, for many years it was virtually banned.*fn3e But Rule 46 of the Equity Rules of 1912 reverted to the 1789-1802 practice of reliance on oral testimony as the normal method in equity suits. The present Civil Rules continue that valuable reform.
The result of the stress on demeanor is to confer immense discretion*fn4 on those who, in finding facts, rely on oral testimony.*fn5 But few doubt that the risk involved is, on the whole, well worthwhile. This is true despite the fact that methods of evaluating the credibility of oral testimony do not lend themselves to formulations in terms of rules and are thus, inescapably, "un-ruly." In his brilliant discussion of evidence, Sir James Stephen illuminated the difficult task of a trial judge who, observing a witness in the brief period when the witness appears in court, tries to ascertain how far the witness' "powers of observation and memory * * * enable him to tell the truth" and "how far the innumerable motives, by any one of which he may be activated, dispose him" to do so. "No rules of evidence * * * can perceptible affect this difficulty," Stephen remarked. "Judges must deal with it as well as they can by the use of their natural faculties and acquired experience, and the miscarriages of justice in which they will be involved by reason of it must be set down to the imperfection of our means of arriving at truth. The natural and acquired shrewdness and experience by which an observant man forms an opinion as to whether a witness is or is not lying, is by far the most important of all a judge's qualifications, infinitely more important than any acquaintance with law or with rules of evidence. No trial ever occurs in which the exercise of this faculty is not required; but it is only in exceptional cases that questions arise which present any legal difficulty, or in which it is necessary to exercise any particular ingenuity in putting together the different facts which the evidence tends to establish. This pre-eminently important power for a judge is not to be learnt out of books. Insofar as it can be acquired at all, it is to be acquired only by experience, for the acquisition of which the position of a judge is by no means peculiarly favourable. People come before him with their cases ready prepared, and give the evidence which they have determined to give. Unless he knows them in their unrestrained and familiar moments, he will have great difficulty in finding any good reason for believing one man rather than another * * * Upon the whole, it must be admitted that little that is really serviceable can be said upon the inference from an assertion to the truth of the matter asserted. The observations of which the matter admits are either generalities too vague to be of much practical use, or they are so narrow and special that they can be learnt only by personal observation and practical experience. Such observations are seldom, if ever, thrown by those who make them into the form of express propositions. Indeed, for obvious reasons, it would be impossible to do so. The most acute observer would never be able to catalogue the tones of voice, the passing shades of expression or the unconscious gestures which he had learnt to associate with falsehood; and if he did, his observations would probably be of little use to others. Every man must learn matters of this sort of himself, and though no sort of knowledge is so important to a judge, no rules can be laid down for its acquisition * * * No process is gone through, the correctness of which can be independently tested. The judge has nothing to trust but his own nature and acquired sagacity."*fn5a Sir Henry Maine agreed with Stephen. He said that there are no "rules to guide" a "Judge of the Fact" in "drawing inferences from the assertion of a witness to the existence of the facts asserted by him." "It is," he wrote, "in the passage from the statements of a witness to the inference that those statements are true, that judicial inquiries generally break down. The English procedure of examination is doubtless entitled to high praise; but, on the whole, it is the rarest and highest personal accomplishment of a judge to make allowance for the ignorance and timidity of witnesses, and to see through the confident and plausible liar. Nor can any general rules be laid down for the acquisition of this power which has methods of operation peculiar to itself, and almost indefinable."*fn5b This lack of rules ("un-ruliness"), with its concomitant wide discretion in the fact trier, yields inherent difficulties not surmountable by a reviewing court,*fn5c regardless of whether the fact-trier be a judge, a jury, or a trial examiner.
In line with the foregoing remarks, we said in N.L.R.B. v. Universal Camera Corp., 190 F.2d 429 (C.A.2), that, in the light of the Supreme Court's decision in that case,*fn6 the Board must give considerable weight to findings of an Examiner based directly upon the "bearing delivery" of witnesses who orally testified before him. So that, although the Board vis a vis the Examiner as to findings is not in exactly the same position as an upper court with respect to a trial judge,*fn7 we surely may not upset the Board when it accepts a finding of an Examiner which is grounded upon (a) his disbelief in an orally testifying winess' testimony because of the witness' demeanor or (b) the Examiner's evaluation of oral testimony as reliable, unless on its face it is hopelessly incredible - cf. Gindorff v. Prince, 189 F.2d 897 (C.A.2) - or flatly contradicts either a so-called "law of nature" or undisputed documentary testimony - cf. Orvis v. Higgins, 180 F.2d 537 (C.A.2). As we said in N.L.R.B. v. Smith Victory Corp., 190 F.2d at 57, "Upon the evidence before us we should unhesitatingly have held that a judge's finding to the same effect was not 'clearly erroneous,' and Universal Camera Corp. v. N.L.R.B., supra, did not make the findings of ...