JO ANN CAB CORP.
United States Court of Appeals Second Circuit.
Jerome Hoffer, New York City (Philip Hoffer, New York City, of counsel), for defendant-appellee.
Before CLARK, FRANK and HINCKS, Circuit Judges.
FRANK, Circuit Judge.
1. Plaintiff complains of this only, that the judge erred in that part of his charge, quoted above, concerning the burden of proof, when he spoke repeatedly of the jury’s “conviction.”
2. Courts and commentators have said that, in the ordinary civil suit, usually a judge commits reversible error if he instructs the jury that the plaintiff cannot recover unless he “convinces” them (or the like). Such a charge, it is maintained, indicates something midway between a “preponderance” and “beyond a reasonable doubt,” and should therefore be reserved for certain exceptional 931*931 sorts of civil cases where more than a “preponderance” is required. It may be that such nice verbal shadings possess no significance for jurymen. Indeed, no one can say with assurance that most jurors have anything like a clear understanding of the phrase “preponderance of the evidence”; and if those words tell jurors little, it is questionable whether “convince” means more to them.
“Preponderance” is but a long latinism for the short English words “weigh more.” Today, it is merely a metaphor; it suggests that, when the jury considers the evidence, it is as if the evidence were weighed on a scale. The metaphor reflects ideas dominant in a long distant past. For, once upon a time, centuries ago, this metaphor received an almost completely literal interpretation: each item of testimony was then assigned a quantitative value or specific weight. (The antiquity of this idea appears from the way in which, outside the legal realm, words relating to weighing have been metaphorized into words relating to thought, e. g., “deliberate,” “ponder,” “putative,” “impute,” and “examine.”)
The quantitative method of assaying evidence has long been abandoned. But the old idea recurrently attracts those who are hot for certainty: In the nineteenth century, Jeremy Bentham proposed that the courts use a sort of “thermometer of persuasion”; and in the twentieth century, a similar fatuous notion has been solemnly put forward. Fatuous it is, as Wigmore has declared. “The truth is,” he wrote, “that no one has yet invented or discovered a mode of measurement for the intensity of human belief. Hence there can be yet no successful method of communicating intelligibly to a jury a sound method of self-analysis for one’s belief. If this truth be appreciated, courts will cease to trust any particular form of words as necessary or decisive in the law for that purpose; for the Law cannot expect to do what Logic and Psychology have not yet done.”
Yet we still strive to guide juries with the metaphorical term “preponderance.” Attempting to explain it to jurors, trial judges speak of “weight” and “scales,” thus tending to impress on the twelve “lay gents” something which approximates the metaphor’s literal meaning: “The application of the phrase `preponderance of the evidence’ is apt to lead the judicial discussion close to the danger line of the fallacious quantitative or numerical theory of testimony.”
932*932 Of course, no one can speak (and most men cannot think) minus metaphors, even in the physical sciences or mathematics. This fact goes largely unobserved because most of the metaphors in daily speech are of the “dead” or “dormant” or “fatigued” kind. However, danger lurks in the literal use of a metaphor as if it were a complete statement of actual fact rather than a sort of analogy or “fiction” (i. e., an “as if”). Such literalism has subtle consequences that are the more dangerous just because the “fictional” character of the metaphor is not adequately recognized. Sensing this danger lurking in 933*933 the phrase “preponderance of the evidence,” trial judges often wisely inform the jurors (as did the trial judge here) of the irrelevance of the mere “number of witnesses or the quantity of exhibits.” Yet, in all likelihood, the pull of the “preponderance” metaphor in the direction of literalness does frequently puzzle jurors.
Recognizing the imponderable effects of “preponderance” on jurors, one may (to repeat) wonder whether it is true that “conviction” or “convinced” conveys something greater to their minds. Perhaps the judges at one time had a basis for the idea that it does. But today we lack any solid foundation for that idea. For alterations in word-fashions notoriously cause words to undergo changes of meaning. Judges therefore seem rather naive when they confidently remark that they are “sufficiently in touch with affairs” to know the “meaning which the ordinary man on the street attributes to ordinary everyday English”, or what particular words “convey * * * to the common mind.” Tough language-barriers often exist between the man in the library or the judge and the man on the street (or on a bus or in the subway).
Judge Wanamaker, some 17 years ago, learned from a considerable number of former jurors that they had found “preponderance of the evidence” the most difficult legal term to understand. Perhaps so-called “opinion research” can furnish, via polls, more illumination of popular responses to “preponderance” and “convince.” The results of any such polls, however, should, as yet, be skeptically eyed; and it may be doubted whether any poll will ever adequately reflect the attitude of the dozen chosen fact-triers in any particular law suit, especially as word-meanings often vary from social group to social group in the community.
“Lawyers,” said the literary critic John Mason Brown, recently, “are excused from the necessity of interesting their readers, and all too often — let’s face the evidence — they take advantage of this enviable exemption.” But a 934*934 judge, when charging a jury, has the duty not only of interesting his twelve listeners but of having them understand him as far as possible, particularly in that part of his charge dealing with the burden of proof. So that the judges ought to make every practicable effort to discover what their words on that subject signify to most non-lawyers, and then to revise jury-charges accordingly. If not, they will be taking a dogmatic position vis à vis communications from judges to laymen which sagacious judges and lawyers reject with respect to communications between laymen. Corbin recently has published a masterful discussion of that topic in which he reports the following: There exist no methods which will “infallibly lead to one correct understanding” of another’s words because in “reading each other’s words, men certainly see as through a glass, darkly”; the trouble with the belief that words have “one true meaning” is that the phrase “one true meaning” lacks “one true meaning”; to “elucidate the meaning of the word `mean’ requires fourteen long columns of fine print in the Oxford Dictionary”; it “is the universal custom of mankind to speak elliptically and to assume the existence and the understanding of things not expressed in words”; it “may be unfortunate, but it is true, that men often use written and spoken words without having any clear notion of what they want to say.” Corbin’s report accords with what sagacious laymen have said of the same subject: “Our speech is a compromise between the ultimate incommunicability of one person with another and the conventional communication values attached to certain symbols.” “Ambiguity, indefiniteness, vagueness and equivocation are ever with us. [We should] learn to curb the arrogance, the presumption, that we necessarily know what others are saying.” “The greatest enemy of communication * * * is the illusion of it.”
It is idle, then, to hope that all language — legal language included — can be made entirely ambiguity-proof, and doubtful whether such an achievement 935*935 would be desirable. But no matter what the difficulties may be of rendering many of the legal rules intelligible to the jury, surely the most important part of the judge’s charge relative to the facts — i. e., that of dealing with the burden of proof — ought to be so worded that the jurors can comprehend it. We ought not, therefore, forever base decisions on unverified assumptions about jurors’ comprehension of differences between “preponderance” and “conviction.”
Nevertheless, at least until such time as we have satisfactory returns from checks of popular reactions to the traditional formulas, we think we should adhere to the ruling that, in an ordinary civil suit, such as the instant case, words like “convince” and “conviction” should be shunned in a jury charge. But here, before the jury had retired, the judge had correctly charged that the burden was on the plaintiff to prove his case “by a preponderance of the evidence”; and, although the language above criticized was also included, no exception was taken to the charge at that stage. We think that nothing thereafter occurring could have reasonably given the jury to understand that the correct instruction theretofore given was modified. However that may be, reading the charge as a whole, we think the instructions were legally sufficient and leave no room for a just inference that the verdict was legally defective.
 See, e. g., 9 Wigmore, Evidence (3d ed.) § 2498, p. 325, note 1; West v. Boston & M. R. R., 81 N.H. 522, 129 A. 768, 773, 42 A.L.R. 176; People v. Miller, 171 Cal. 649, 154 P. 468; Aarons v. Levy Brothers & Adler Rochester, Inc., 44 Ohio App. 488, 185 N.E. 62; Livanovitch v. Livanovitch, 99 Vt. 327, 131 A. 799.
 See, e.g., 9 Wigmore, Evidence (3d ed.) § 2498, pp. 330-334.
 See Millar, in Engelmann, History of Continental Civil Procedure (1927) 41-49; 7 Wigmore, Evidence, 241 et seq.; Holdsworth, 1 History of English Law (3d ed. 1922) 302-304.
Consider, too, the ordeal of the scales.
 Bentham, Rationale of Judicial Evidence (J. S. Mill ed.) Bk. I, Ch. 6, s. 1.
By this device, the trial court would estimate the value of each witness’ testimony to a nice degree on a scale running from 1 to 10 or 1 to 1000. In this way, such testimony would be measured in terms of so many degrees of positive or negative persuasion. So, for instance, it would be noted that the “persuasion is at 9 above,” just “as in speaking of temperature, a man says, the mercury stands at 9 above.” “For want of an adequate mode of expression,” said Bentham, “the real force of testimony in a cause has hitherto been exposed to perpetual misinterpretation. * * * Old measures of every kind receive additional correctness: the electometer, the calorimeter, the photometer, the eudiometer. * * * Has not justice its use as well as gas?”
Best, Evidence (12th ed.) 62, rejecting Bentham’s “therometer,” says, “The substitution of arithmetic for observation and reasoning, when estimating the value of evidence, is not confined to past ages.”
 Loevinger, An Introduction To Legal Logic, 27 Ind.L.Rev. (1952) 472 at 494-495; see also Loevinger, Jurimetrics, 33 Minn.L.Rev. (1949) 455.
Cowan, 96 U. of Pa.L.Rev. (1948) 484, 490, writes of a need for “credibility scales.”
 9 Wigmore, Evidence (3d ed.) § 2498, p. 325.
 9 Wigmore, Evidence (3d ed.) § 2498, p. 334.
 But see Hadamard, The Psychology of Invention In The Mathematical Field (1945) 66-97, as to thinking without verbal images.
 See, e. g., Richards, The Philosophy of Rhetoric (1936) 92-93: “The metaphors we are avoiding steer our thought as much as those we accept. * * * I hold with Bradley that our pretence to do without metaphor is never more than a bluff waiting to be called”; Stebbing, Thinking To Some Purpose (1939) Ch. 9.
 Consider, e.g., “lines of force,” the “affinity” of chemicals for one another. See, e.g., Cohen, A Preface to Logic (1944) 84; Stebbing, Thinking To Some Purpose (1939) 109-110.
 See, e.g., Fowler, Modern English Usage (1926) 348; Fuller, Legal Fictions, 25 Ill.L.Rev. (1930) 362, 374-375.
Sleeping metaphors can be awakened. Richards (loc. cit.) 102, notes that the “distinction between dead and living metaphors” is “itself a twofold metaphor.”
Murry, Countries of The Mind (1931) says that, since metaphor is as “ultimate as speech itself, and speech as ultimate as thought,” to investigate metaphor is like investigating “any of the primary data of consciousness: it cannot be pursued very far without our being led to the borderline of sanity. * * * The earth trembles and yawns beneath the explorer’s feet. * * * Therefore we instinctively seek to circumscribe our own inquiries by leaving out of account, as far as may be, the countless lost or dormant metaphors of which the most part of language is composed. * * *”
Even common words, such as “quickly” or “lively” contain “ly” which originally meant “like.”
 For the relation between analogies, metaphors and “fictions,” see, e. g., Vaihinger, The Philosophy of “As If” (transl. 1925) passim; Fuller, Legal Fictions, 25 Ill.L.Rev. (1930-31) 370, 380, 383, 387-388, 878ff, 892ff. Cf. Frank, Law and The Modern Mind (1930) 166-167, 312-322; Tourtoulon, Philosophy In The Development of Law (transl. 1922) 385 et seq., 644-653.
One can sympathize with the commonsense reaction of the 18th century physiologist, William Hunter: “Some physiologists will have it that the stomach is a mill, others that it is a fermenting vat, others that it is a stew-pan; but, in my view of the matter, it is neither a mill, a fermenting vat nor a stewpan but a stomach, gentlemen, a stomach.” Quoted in Muller, Science and Criticism (1943) 106. (Yet a stomach abstracted from a body is a fiction.)
 See United Shipyards, Inc., v. Hoey, 2 Cir., 131 F.2d 525, 526-527. “But elliptical discussions of cases partly alike, as if there were complete identity, is merely for convenience. There is present, although it may be unexpressed, an `as if,’ a `let’s pretend’ — a simile or metaphor. Such `as-if’ or metaphorical thinking is invaluable in all provinces of thought (not excepting that of science). However, some of the greatest errors in thinking have arisen from the mechanical, unreflective, application of old formulations — forgetful of a tacit `as if’ — to new situations which are sufficiently discrepant from the old so that the emphasis on the likeness is misleading and neglect of the differences leads to unfortunate or foolish consequences.”
But see Collingwood: Language “never is its own meaning and is therefore always symbolic or metaphorical; but, when this fact is as yet undiscovered by the user of language we say that he is using it `metaphorically,’ and when he realizes that words are mere symbols and distinguishes what they are from what they mean, then by facing and accepting the metaphorical character of all language he has overcome it and is henceforth using it `literally.'” Collingwood, Speculum Mentis (1924) 157.
Richards (loc. cit. 132) remarks that “there is no whole to any analogy; we use as much of it as we need; and, if we tactlessly take any analogy too far, we break it down.”
Brooks and Warren, Modern Rhetoric (1949) 426, say that an effective metaphor must be neither too “far-fetched” nor too “nearly fetched.” See also Embler, Metaphor and Social Belief, 8 Etc. 9 (1951) 83.
As to the paradoxical (or pun-like) character of metaphors, see, e.g., Frank, Some Tame Reflections on Some Wild Facts, in the volume Vision and Action (1953).
 See Weiler v. United States, 323 U.S. 606, 608, 65 S.Ct. 548, 549, 89 L.Ed. 495: “Our system of justice rests on the general assumption that the truth is not to be determined merely by the number of witnesses on each side of a controversy. In gauging the truth of conflicting evidence, a jury has no simple formulation of weights and measures upon which to rely. The touchstone is always credibility; the ultimate measure of testimonial worth is quality and not quantity.”
 Hall, Leave Your Language Alone! (1950) Ch. 10.
He quotes Dante: “Since man is a most unstable and variable being, language cannot be long-lasting nor stable; but like other human things such as customs and dress, it has to vary in space and time.”
Hall also quotes Horace: “Words which are now in honor fall into disuse, if usage so wills it.”
 Wanamaker, Trial By Jury, 11 U. of Cin.L.Rev. (1937) 11; see also McBaine, Burden of Proof, Degrees of Belief, 32 Cal.L.Rev. (1944) 242, 262, note 37.
McBaine (255) remarks that “beyond a reasonable doubt” is far less puzzling to laymen, and approves (257) of the “view frequently expressed that no explanation of the term is helpful to the jury.” See also 9 Wigmore, § 2497, pp. 318 et seq.
 Cf. Sorenson and Sorenson, The Admissibility and Use of Opinion Research Evidence, 28 N.Y.U.L.Rev. (1953) 1213, 1258, 1259.
 Cf. Roth v. Goldman, 2 Cir., 172 F.2d 788, 796, concurring opinion; cf. also Sumner and Keller, Science of Society, II (1927) 1305: “One of the ways * * * for disconcerting even an intelligent adversary is to overwhelm him with mathematical formulas, graphs, and other esoteric devices which he cannot withstand because he cannot understand them. They are similar to the secret, fetishistic jargon of the medicine man.”
 Hall, loc. cit., passim.
“Once, when counsel was citing case after case to establish the meaning of a will made by an old woman, Lord Mansfield is reported to have interrupted: `Sir, do you think that this old lady ever read those cases or would have understood if she did?'” Holliday, History of English Law, 127, quoted in 3 Corbin, Contracts (1951) 23, note 23.
 As to the importance of studying the social effects of legal rules and doctrines, see, e.g., the citations in Cairns, Legal Science (1941) 6-7.
 McBaine, loc. cit. at 246, suggests that the charge on burden of proof be revised so as to tell the jurors the proper degree of belief, varying it according to the type of case: (a) In the ordinary civil suit, the standard would be the jury’s belief of “what probably has happened”; (b) in civil cases calling for a higher standard, it would be a belief of “what highly probably has happened”; (c) in a criminal case, a belief as to “what almost certainly has happened.” The suggestion may sound promising; but apparently McBaine has not attempted to try out these formulas on laymen.
 Corbin, 3 Contracts, Part Three.
 Schlauch, The Gift of Tongues (1948) 113.
 Lee, The Language of Wisdom and Folly (1949) 135. See also Johnson, People In Quandry (1947) 471ff.
 Whyte, Is Anybody Listening? (1952) 38.
 Bentham seemed to have entertained such a hope of “disambiguation” (akin to his notion of a persuasion-thermometer). See Bentham, Theory of Fictions (1932), Introduction (by Ogden) xvii as to Bentham’s eager search for “orthological clarity.”
Whitehead warned that, from too much respect for the possibilities of language, there issues a demand for impossible clarity: “Insistence on clarity at all costs is based on sheer supersitition as to the mode in which the intellect functions.” He also spoke of the “dangers of a logic which presupposes linguistic adequacy,” and said that “the success of language in conveying information is vastly over-rated, especially in learned circles.” Whitehead, Adventures of Ideas (Pelican ed. 1942) 89, 165.
 Richards has written of the “resources of ambiguity.” Richards, How to Read A Page (1942) 22; Richards, The Philosophy of Rhetoric (1936) 40, 72-73; cf. Kenneth Burke, A Grammar of Motives (1945), Introduction and p. 5.
J. Stone declares that the ability of courts wisely to adapt old legal rules to changing conditions is “due to the fact that the same norm is used but with a changing meaning, the same verbal formulas with an apparent identity of substance which is verbal only”; Stone, Fallacies of the Logical Form, in the volume Modern Interpretations of Legal Philosophy (1947) 696, 721. See also Max Radin, The Trail of the Calf, 32 Cornell L.Q. (1946) 137; Frank, Law and The Modern Mind (1930) 26-31, 60-61; Frank, If Men Were Angels (1942) 313; Levi, An Introduction to Legal Reasoning (1949) 73-74. For a somewhat cynical discussion, see Seagle, Law, The Science of Inefficiency (1952) 23-30.
 Defendant argues that because of the paucity of plaintiff’s proof, the defendant was entitled to a directed verdict, and therefore the error was harmless. We have not considered that argument.