Pennsylvania R. Co. v. Chamberlain, 288 U.S. 333 (1933)

Primary Holding

A court can order a directed verdict if the evidence is so insubstantial on one side that the other party would be entitled to a new trial if an adverse verdict were reached.


U.S. Supreme Court

Pennsylvania R. Co. v. Chamberlain, 288 U.S. 333 (1933)

Pennsylvania Railroad Co. v. Chamberlain

No. 379

Argued January 19, 1933

Decided February 13, 1933

288 U.S. 333



U.S. Supreme Court

Pennsylvania R. Co. v. Chamberlain, 288 U.S. 333 (1933) Pennsylvania Railroad Co. v. Chamberlain

No. 379

Argued January 19, 1933

Decided February 13, 1933

288 U.S. 333




1. A plaintiff in an action for wrongful death cannot recover by proof of facts from which it may be inferred that the injury resulted from an event attributable to the defendant's negligence if the same facts give equal support to an inference that it resulted from another and different event not so attributable. P. 288 U. S. 339.

2. When the plaintiff has failed to sustain his burden of proof, because, as to the existence of a vital fact, two equally justifiable inferences may be drawn from the facts proven, one for and the other against him, the mere conclusion of a witness as to which

Page 288 U. S. 334

inference should be drawn cannot resolve the doubt in his favor. P. 288 U. S. 340.

3. Where the plaintiff's right of recovery depends upon the existence of a particular fact being inferred from proven facts, such inference is not permissible in the face of the positive and otherwise uncontradicted testimony of unimpeached witnesses, consistent with the facts actually proved, showing affirmatively that the fact sought to be inferred did not exist. P. 288 U. S. 340.

4. A brakeman who was riding a cut of cars moving on a distributing track of a large freight yard, in a switching operation, fell and was killed, and the case against the company depended on whether the accident resulted from a collision by a second cut from behind. Three employees who were on the second cut testified positively that there was no collision or contact between it and the first one, and in this they were corroborated by every other employee who was in a position to see. One witness alone, who was standing fifty feet from the track and nine hundred feet from the spot where the body was found, testified that he heard a crash -- a thing not unusual in the yard, only "extra loud" -- which was not sufficient to attract his attention, and did not cause him to turn at once, but that, shortly thereafter, he did turn and saw the two strings of cars moving together, with the deceased, who had been visible before, no longer in sight. Held that the witness' further statement that the crash was caused by a collision of the two strings in question was not testimony to a fact, but merely an inference or conclusion of his own, and that the manifest impossibility of telling by sight, from the place where he was, whether the cars were moving in contact, made his testimony on that point incredible. P. 288 U. S. 342.

5. In a personal injury case, verdict should be directed for the defendant if the evidence for the plaintiff is so insufficient that a verdict in his favor would be improper and must be set aside on motion for a new trial. P. 288 U. S. 343.

6. The scintilla rule has been definitely and repeatedly rejected so far as the federal courts are concerned; the verdict cannot rest on mere speculation and conjecture. P. 288 U. S. 343.

59 F.2d 986 reversed.

District Court affirmed.

Certiorari, 287 U.S. 589, to review the reversal of judgment on a verdict directed in favor of the railroad

Page 288 U. S. 335

company in an action for wrongful death, under the Employers' Liability Act.

MR. JUSTICE SUTHERLAND delivered the opinion of the Court.

This is an action brought by respondent against petitioner to recover for the death of a brakeman alleged to have been caused by petitioner's negligence. The complaint alleges that the deceased, at the time of the accident resulting in his death, was assisting in the yard work of breaking up and making up trains and in the classifying and assorting of cars operating in interstate commerce; that, in pursuance of such work, while riding a cut of cars, other cars ridden by fellow employees were negligently caused to be brought into violent contact with those upon which deceased was riding, with the result that he was thrown therefrom to the railroad track and run over by a car or cars, inflicting injuries from which he died.

At the conclusion of the evidence, the trial court directed the jury to find a verdict in favor of petitioner. Judgment upon a verdict so found was reversed by the Court of Appeals, Judge Swan dissenting. 59 F.2d 986.

That part of the yard in which the accident occurred contained a lead track and a large number of switching tracks branching therefrom. The lead track crossed a "hump," and the work of car distribution consisted of pushing a train of cars by means of a locomotive to the top of the "hump," and then allowing the cars, in separate strings, to descend by gravity, under the control of

Page 288 U. S. 336

hand brakes, to their respective destinations in the various branch tracks. Deceased had charge of a string of two gondola cars, which he was piloting to track 14. Immediately ahead of him was a string of seven cars, and behind him a string of nine cars, both also destined for track 14. Soon after the cars ridden by deceased had passed to track 14, his body was found on that track some distance beyond the switch. He had evidently fallen onto the track and been run over by a car or cars.

The case for respondent rests wholly upon the claim that the fall of deceased was caused by a violent collision of the string of nine cars with the string ridden by deceased. Three employees, riding the nine-car string, testified positively that no such collision occurred. They were corroborated by every other employee in a position to see, all testifying that there was no contact between the nine-car string and that of the deceased. The testimony of these witnesses, if believed, establishes beyond doubt that there was no collision between these two strings of cars, and that the nine-car string contributed in no way to the accident. The only witness who testified for the respondent was one Bainbridge, and it is upon his testimony alone that respondent's right to recover is sought to be upheld. His testimony is concisely stated, in its most favorable light for respondent, in the prevailing opinion below by Judge Learned Hand, as follows (p. 986):

"The plaintiff's only witness to the event, one Bainbridge, then employed by the road, stood close to the yardmaster's office, near the 'hump.' He professed to have paid little attention to what went on, but he did see the deceased riding at the rear of his cars, whose speed when they passed him he took to be about eight or ten miles. Shortly thereafter, a second string passed which was shunted into another track, and this was followed by the nine, which, according to the plaintiff's theory, collided

Page 288 U. S. 337

with the deceased's. After the nine cars had passed at a somewhat greater speed than the deceased's, Bainbridge paid no more attention to either string for a while, but looked again when the deceased, who was still standing in his place, had passed the switch and onto the assorting track where he was bound. At that time, his speed had been checked to about three miles, but the speed of the following nine cars had increased. They were just passing the switch, about four or five cars behind the deceased. Bainbridge looked away again, and soon heard what he described as a 'loud crash,' not, however, an unusual event in a switching yard. Apparently this did not cause him at once to turn, but he did so shortly thereafter, and saw the two strings together, still moving, and the deceased no longer in sight. Later still, his attention was attracted by shouts, and he went to the spot and saw the deceased between the rails. Until he left to go to the accident, he had stood fifty feet to the north of the track where the accident happened, and about nine hundred feet from where the body was found."

The court, although regarding Bainbridge's testimony as not only "somewhat suspicious in itself, but its contradiction . . . so manifold as to leave little doubt," held, nevertheless, that the question was one of fact depending upon the credibility of the witnesses, and that it was for the jury to determine, as between the one witness and the many, where the truth lay. The dissenting opinion of Judge Swan proceeds upon the theory that Bainbridge did not testify that, in fact, a collision had taken place, but inferred it because he heard a crash and because thereafter the two strings of cars appeared to him to be moving together. It is correctly pointed out in that opinion, however, that the crash might have come from elsewhere in the busy yard, and that Bainbridge was in no position to see whether the two strings of cars were actually together; that Bainbridge repeatedly said he was

Page 288 U. S. 338

paying no particular attention, and that his position was such, being 900 feet from the place where the body was found and less than 50 feet from the side of the track in question, that he necessarily saw the strings of cars at such an acute angle that it would be physically impossible even for an attentive observer to tell whether the forward end of the nine-car cut was actually in contact with the rear end of the two-car cut. The dissenting opinion further points out that all the witnesses who were in a position to see testified that there was no collision; that respondent's evidence was wholly circumstantial, and the inferences which might otherwise be drawn from it were shown to be utterly erroneous unless all of petitioner's witnesses were willful perjurers. "This is not a case," the opinion proceeds,

"where direct testimony to an essential fact is contradicted by direct testimony of other witnesses, though even there it is conceded a directed verdict might be proper in some circumstances. Here, when all the testimony was in, the circumstantial evidence in support of negligence was thought by the trial judge to be so insubstantial and insufficient that it did not justify submission to the jury."

We thus summarize and quote from the prevailing and dissenting opinions because they present the divergent views to be considered in reaching a correct determination of the question involved. It, of course, is true generally that, where there is a direct conflict of testimony upon a matter of fact, the question must be left to the jury to determine, without regard to the number of witnesses upon either side. But here, there really is no conflict in the testimony as to the facts. The witnesses for petitioner flatly testified that there was no collision between the nine-car and the two-car strings. Bainbridge did not say there was such a collision. What he said was that he heard a "loud crash" which did not cause him at once to turn,

Page 288 U. S. 339

but that, shortly thereafter, he did turn, and saw the two strings of cars moving together with the deceased no longer in sight; that there was nothing unusual about the crash of cars -- it happened every day; that there was nothing about this crash to attract his attention except that it was extra loud; that he paid no attention to it; that it was not sufficient to attract his attention. The record shows that there was a continuous movement of cars over and down the "hump," which were distributed among a large number of branch tracks within the yard, and that any two strings of these cars moving upon the same track might have come together and caused the crash which Bainbridge heard. There is no direct evidence that, in fact, the crash was occasioned by a collision of the two strings in question, and it is perfectly clear that no such fact was brought to Bainbridge's attention as a perception of the physical sense of sight or of hearing. At most there was an inference to that effect drawn from observed facts which gave equal support to the opposite inference that the crash was occasioned by the coming together of other strings of cars entirely away from the scene of the accident, or of the two-car string ridden by deceased and the seven-car string immediately ahead of it.

We therefore have a case belonging to that class of cases where proven facts give equal support to each of two inconsistent inferences, in which event, neither of them being established, judgment, as a matter of law, must go against the party upon whom rests the necessity of sustaining one of these inferences as against the other before he is entitled to recover. United States F. & G. Co. v. Des Moines Nat. Bank, 145 F. 273, 279-280, and cases cited; Ewing v. Goode, 78 F. 442, 444; Louisville & N. R. Co. v. East Tennessee, V. & G. Ry. Co., 60 F. 993, 999; Tucker Stevedoring Co. v. W. H. Gahagan, 6 F.2d 407, 410; Blid v. Chicago & N.W. R. Co., 89 Neb. 689, 691, et

Page 288 U. S. 340

seq., 131 N.W. 1027; White v. Lehigh Valley R. Co., 220 N.Y. 131, 135, 115 N.E. 439; McGrath v. St. Louis Transit Co., 197 Mo. 97, 104, 94 S.W. 872; Quisenberry v. Met. Street Ry. Co., 142 Mo.App. 275, 281, 126 S.W. 182; Glancy v. McKees Rocks Borough, 243 Pa. 216, 219, 89 A. 972. Compare New York Central R. Co. v. Ambrose, 280 U. S. 486, 280 U. S. 489-490; Stevens v. The White City, 285 U. S. 195, 285 U. S. 203-204; Southern Ry. Co. v. Walters, 284 U. S. 190, 284 U. S. 194.

The rule is succinctly stated in Smith v. First National Bank in Westfield, 99 Mass. 605, 611-612, quoted in the Des Moines National Bank case, supra:

"There being several inferences deducible from the facts which appear, and equally consistent with all those facts, the plaintiff has not maintained the proposition upon which alone he would be entitled to recover. There is strictly no evidence to warrant a jury in finding that the loss was occasioned by negligence, and not by theft. When the evidence tends equally to sustain either of two inconsistent propositions, neither of them can be said to have been established by legitimate proof. A verdict in favor of the party bound to maintain one of those propositions against the other is necessarily wrong."

That Bainbridge concluded from what he himself observed that the crash was due to a collision between the two strings of cars in question is sufficiently indicated by his statements. But this, of course, proves nothing, since it is not allowable for a witness to resolve the doubt as to which of two equally justifiable inferences shall be adopted by drawing a conclusion which, if accepted, will result in a purely gratuitous award in favor of the party who has failed to sustain the burden of proof cast upon him by the law.

And the desired inference is precluded for the further reason that respondent's right of recovery depends upon the existence of a particular fact which must be inferred

Page 288 U. S. 341

from proven facts, and this is not permissible in the face of the positive and otherwise uncontradicted testimony of unimpeached witnesses consistent with the facts actually proved, from which testimony it affirmatively appears that the fact sought to be inferred did not exist. This conclusion results from a consideration of many decisions, of which the following are examples: Wabash R. Co. v. De Tar, 141 F. 932, 935; Ragsdale v. Southern Ry. Co., 121 F. 924, 926; Cunard S.S. Co. v. Kelley, 126 F. 610, 617; Frazier v. Georgia Railroad & Banking Co., 108 Ga. 807, 33 S.E. 996; Bowsher v. Grand Rapids & I. Ry. Co., 174 Mich. 339, 344, 140 N.W. 524; Rashall v. Railroad, 249 Mo. 509, 522, 155 S.W. 426; George v. Mo. Pac. R. Co., 213 Mo.App. 668, 674, 251 S.W. 729, 732; Stines v. Dillman, 4 S.W.2d 477, 478; Akerson v. Great Northern Ry. Co., 158 Minn. 369, 374, 197 N.W. 842; Butterfield v. Trittipo, 67 Ind. 338, 343, 344; Blid v. Chicago & N.W. R. Co., supra, p. 691-693. A rebuttable inference of fact, as said by the court in the Wabash Railroad case, "must necessarily yield to credible evidence of the actual occurrence." And, as stated by the court in George v. Mo. Pac. R. Co., supra,

"It is well settled that, where plaintiff's case is based upon an inference or inferences, the case must fail upon proof of undisputed facts inconsistent with such inferences."

Compare 41 U. S. Gilson, 16 Pet. 327, 41 U. S. 330-331. In Southern Ry. Co. v. Walters, supra, the negligence charged was failure to stop a train and flag a crossing before proceeding over it. The Court concluded that the only support for the charge was an inference sought to be drawn from certain facts proved. In rejecting the inference, this Court said (p. 284 U. S. 194):

"It is argued that it may be inferred from the speed of the train when some of the witnesses observed it crossing other streets as well as Bond Avenue, and from a guess of the engineer as to the time required to get up such speed

Page 288 U. S. 342

after a full stop, that none could have been made at Bond Avenue. But the argument amounts to mere speculation in view of the limited scope of the witnesses' observation, the downgrade of the railway tracks at the point, and the time element involved. (Compare Chicago, M. & St.P. R. Co. v. Coogan, 271 U. S. 472.) Five witnesses for defendant [employees] testified that a full stop was made and the crossing flagged, and that no one was hit by the rear of the tender, which was the front of the train."

"An examination of the record requires the conclusion that the evidence on the issue whether the train was stopped before crossing Bond Avenue was so insubstantial and insufficient that it did not justify a submission of that issue to the jury."

Not only is Bainbridge's testimony, considered as a whole, suspicious, insubstantial, and insufficient, but his statement that, when he turned shortly after hearing the crash, the two strings were moving together, is simply incredible if he meant thereby to be understood as saying that he saw the two in contact, and, if he meant by the words "moving together" simply that they were moving at the same time in the same direction, but not in contact, the statement becomes immaterial. As we have already seen, he was paying slight and only occasional attention to what was going on. The cars were eight or nine hundred feet from where he stood, and moving almost directly away from him, his angle of vision being only 3� 33' from a straight line. At that sharp angle and from that distance, near dusk of a misty evening (as the proof shows), the practical impossibility of the witness being able to see whether the front of the nine-car string was in contact with the back of the two-car string is apparent. And certainly, in the light of these conditions, no verdict based upon a statement so unbelievable reasonably could be sustained as against the positive testimony to the contrary of unimpeached witnesses, all in a position to see,

Page 288 U. S. 343

as this witness was not, the precise relation of the cars to one another. The fact that these witnesses were employees of the petitioner, under the circumstances here disclosed, does not impair this conclusion. Chesapeake & Ohio Ry. v. Martin, 283 U. S. 209, 283 U. S. 216-220.

We think, therefore, that the trial court was right in withdrawing the case from the jury. It repeatedly has been held by this Court that, before evidence may be left to the jury,

"there is a preliminary question for the judge not whether there is literally no evidence, but whether there is any upon which a jury can properly proceed to find a verdict for the party producing it, upon whom the onus of proof is imposed."

Pleasants v. Fant, 22 Wall. 116, 89 U. S. And, where the evidence is "so overwhelmingly on one side as to leave no room to doubt what the fact is, the court should give a peremptory instruction to the jury." Gunning v. Cooley, 281 U. S. 90, 281 U. S. 94; Patton v. Texas & Pacific Ry. Co., 179 U. S. 658, 179 U. S. 660. The rule is settled for the federal courts, and for many of the state courts, that whenever, in the trial of a civil case, the evidence is clearly such that, if a verdict were rendered for one of the parties, the other would be entitled to a new trial, it is the duty of the judge to direct the jury to find according to the views of the court. Such a practice, this Court has said, not only saves time and expense, but "gives scientific certainty to the law in its application to the facts and promotes the ends of justice." Bowditch v. Boston, 101 U. S. 16, 101 U. S. 18; Barrett v. Virginian Ry. Co., 250 U. S. 473, 250 U. S. 476, and cases cited; Herbert v. Butler, 97 U. S. 319, 97 U. S. 320. The scintilla rule has been definitely and repeatedly rejected so far as the federal courts are concerned. Improvement Co. v. Munson, 14 Wall. 442, 81 U. S. 448; Commissioners v. Clark, 94 U. S. 278, 94 U. S. 284; Small Co. v. Lamborn & Co., 267 U. S. 248, 267 U. S. 254; Gunning v. Cooley, supra; Ewing v. Goode, supra, at pp. 443-444.

Page 288 U. S. 344

Leaving out of consideration, then, the inference relied upon, the case for respondent is left without any substantial support in the evidence, and a verdict in her favor would have rested upon mere speculation and conjecture. This, of course, is inadmissible. Chicago, M. & St. P. Ry. v. Coogan, 271 U. S. 472, 271 U. S. 478; Gulf, M. & N. R. Co. v. Wells, 275 U. S. 455, 275 U. S. 459; New York C. R. Co. v. Ambrose, supra; Stevens v. The White City, supra.

The judgment of the Circuit Court of Appeals is reversed, and that of the District Court is affirmed.

MR. JUSTICE STONE and MR. JUSTICE CARDOZO concur in the result.