(Pertinent facts are included in tlie opinion.)
The plaintiff Carruthers, as administrator of Herbert L. Burroughs, and as administrator of Luther Burroughs, brought separate actions against the defendant Railway Company for recovery of damages for the injury and death of his intestates, respectively, alleged to have been brought about by the negligence of the defendant. The actions were consolidated and tried by consent in the lower court, and in this Court the appeals of plaintiff were heard together.
The evidence pertinent to our investigation may be succinctly stated. It tends to show that plaintiff’s intestates were approaching and attempting to cross defendant’s track at a highway crossing, and that a train operated by defendant collided with the motor vehicle in which plaintiff’s intestates were traveling, causing injuries which resulted in the death of both of them.
The occupants of the car were approaching the track from a westerly direction and the train was approaching the crossing from a southerly direction. The evidence tended to show that to the right of the occupants of the car and in the direction from which the train was approaching there was an embankment which partially obscured the approach of the train for some distance in that direction from anyone approaching at some distance from the track. The evidence as to these distances does not purport to be exact.
The plaintiff offered the evidence of Arlie Dunn that he was approaching the crossing from the north over a road parallel with the railroad *51and saw the train approaching the crossing from the south. That he was about 400 feet from the crossing, and that the whistle did not blow and that the bell did not ring; that there is a bank on the right-hand side of the highway along which the deceased were traveling and on the west side of the railroad; that there were woods on the right of the road and a dirt embankment 12 or 15 feet high.
Witnesses for the defendant testified that the whistle did blow.
Several witnesses testified that they did not see Arlie Dunn at the scene of the wreck, and there was evidence on the part of the defendant tending to show that he arrived later.
All of plaintiff’s assignments of error relate to instructions given the jury upon the trial. Mainly, they hinge about the oft-repeated instructions given by the judge characterizing the evidence of the plaintiff as to the blowing of the whistle and ringing of the bell of the approaching train as “negative” evidence and “weak.”
Plaintiff contends that in these instructions the trial judge not only mistakenly characterized his evidence as negative but pointed out that it was weak, drawing a distinction between positive and negative evidence, thereby directly attacking the evidence on which he sought to show the negligence of' the defendant in not sounding a timely warning at the crossing as being of a negative character and weak.
The matter objected to is embodied in a special instruction asked for by the defendant and given in the charge, which in part reads as follows :
“Negative evidence, meaning testimony that an alleged fact did not exist, although weak, is admissible, if the witness’ situation was such that he would have known of it if it had existed. While the aifirmative testimony of a credible witness is ordinarily more reliable than the negative testimony of an equally credible witness, still testimony that a person nearby who could have heard and did not hear the sounding of a whistle or the ringing of a bell is some evidence that no such signal was given.”
“The entire probative value of the negative fact lies in the circumstance at once to be stated. Such evidence is meaningless, however, if the non-seeing or the non-hearing are equally consistent with the occurrence of the events themselves. Nothing is shown of any value in evidence if at the time of the alleged occurrence of these events the witness was so situated that they well might have occurred and he neither have seen nor heard them. 3 Modern Law of Evidence, sec. 1758. The basic psychological, as well as probative weakness of negative evidence lies in this: The fact may have taken place in the sight or hearing of a person who may not have perceived it; or who perceived it falsely because of defective perceptive apparatus, unfavorable surrounding conditions, or the state of mind of the witness; or who, having originally perceived it correctly, has since forgotten it. Testimony of witnesses that they did *52not bear a locomotive signal at a given time and place is given probative effect according to tbe surrounding circumstances, and is as ‘forceful as tbe opportunities for observation, and tbe concentration and attention of tbe witness on wbat was going on at tbe time, indicate, wben considered witb all tbe circumstances wbicb bear on tbe credibility of witnesses generally.’ ”
“Tbe witness Dunn’s testimony tbat tbe whistle did not blow nor tbe bell ring is wbat is called negative evidence. But before you bave tbe right to give any weight whatever to this testimony, you must find by tbe greater weight of tbe evidence tbat be was in a position to bear and could bave beard tbe whistle if it bad blown or tbe bell if it bad rung. Unless you find by tbe greater weight of tbe evidence tbat be was within bearing distance of tbe whistle of tbe engine or tbe ringing of tbe bell as tbe engine drew near to tbe crossing, if tbe whistle bad blown or tbe bell bad rung, you cannot find tbat be was in a position to bear and you would not be entitled to give any weight whatever to bis statement tbat tbe Avbistle did not blow nor tbe bell ring and you will answer tbe first issue ‘No.’ ”
We do not feel tbat tbe case at bar justifies an extended discussion of tbe distinctions between positive and negative evidence, either here or before tbe jury. Discussions of tbat sort, often scholastic, or at least highly metaphysical, could give tbe jury little aid in arriving at tbe truth of tbe matter and are likely to lead to confusion. Scientifically trained jurors, whether desirable or not, cannot be produced at one sitting while in medias res. And, perhaps, tbe school in wbicb jurors learn to think straight is to be preferred, although it never .closes a session, dismisses a pupil, or gives a diploma. It will be found tbat most of tbe things tbe court needs to impress are well within tbe everyday experience and understanding of tbe jurors in simpler form, and in dealing witb tbe elements of credibility actually involved tbe court will usually find familiar instruments of guidance for tbe jury, without tbe necessity of making an excursion into tbe field of epistemology and psychology.
Perception of tbe existence of tbe fact or tbe nonexistence of tbe fact depends upon tbe exercise of tbe same sensory faculties — usually in legal investigations tbat of sight or bearing — and frequently it is a matter of common experience tbat an observer may be as positive of one as of tbe other. Taylor v. Security Co., 145 N. C., 383, 59 S. E., 139. Tbe reliability of tbe testimony in either case depends on tbe same factors: ability to perceive, opportunity, degree of attention, memory, and honesty of subsequent statement in evidence.
Tbe omission' of a legal duty for wbicb actionable negligence may be imputed is always a negative fact. In tbe majority of negligence cases plaintiff must establish tbat tbe defendant did not perform some act or *53did not do some thing wbieb it was his duty to do or perform in the observance of due care. “In following out this distinction, courts have sometimes overlooked the fundamental fact that in such a case the plaintiff is necessarily confined to negative evidence. If such evidence is unworthy of belief simply because it is negative, then the plaintiff must nearly always fail.” Chicago, R. I. & P. R. Co. v. Stepp, 164 F., 785, 22 L. R. A. (N. S.), 350. The testimony of a witness is to be considered in the light of attendant circumstances.
Textwriters usually make a distinction between the form of the statement — “It did not blow” and “I did not hear it blow”- — referring to a required warning or signal, since the former expression implies that the witness was giving attention to the matter, while the latter suggests that he may not have been giving such attention. Chamberlayne, Modern Law of Evidence, section 1758; 22 C. J., p. 66, section 9; Cox v. Schuylkill Valley Track Co., 214 Pa., 223, 228, 63 A., 599. The comment from Chamberlayne, forming the basis of the special instruction, was upon evidence of the latter kind. See McConnell v. State, 67 Ga., 633.
Rut whatever the form of the statement, its value as evidence depends on the factors we have mentioned, and it is, therefore, obvious that there may be circumstances under which such evidence loses its purely negative character and becomes of such substance that, when contradicted, only a question as to credibility between the witnesses is raised. Philadelphia, B. & W. R. Co. v. Gatta (Del.), 85 A., 721, 47 L. R. A. (N. S.), 933; Coughlin v. People, 18 Ill., 266; Cotton v. Wilmar & S. F. Railroad Co., 99 Minn., 366, 109 N. W., 835; Georgia P. R. Co. v. Bowers, 86 Ga., 22, 12 S. E., 182.
In such cases it will be found that supporting circumstances give to the evidence a quality of objectivity and substance that would render the rule .applied in this case inapplicable. Such evidence, when properly supported by circumstances, may run through “all degrees of credibility.” Chicago, R. I. & P. Co. v. Stepp, supra; Mobile & O. R. Co. v. Johnson, 157 Miss., 266, 126 So., 827. In Johnson v. R. R., 214 N. C., 484, 199 S. E., 704, there were no supporting circumstances — in fact, the circumstances were of a contrary nature such as to render the evidence purely negative and sterile of any probative value. Hofford v. Illinois C. R. Co., 138 Iowa, 543, 110 N. W., 446; Baltimore & Ohio R. Co. v. State, 96 Md., 67, 53 A., 672.
Speaking of the application of the distinction between the two classes ■of evidence made by rule of thumb, Mr. Wigmore says: “The rule is a ■discredit to the science of law, and should be discarded. The vain lucubrations to which it leads have no relation to the real probative value •of specific testimony.” 1 Wigmore Ev., 2d Ed., section 664. In most states where, as in ours, the trial judge is prohibited from expressing an *54opinion on the evidence, such comparisons are held for error as necessarily involving an expression of opinion. The Chamberlayne Trial Evidence, Tompkins (1936), p. 107, section 136; Atchison, etc., R. Co. v. Feehan, 149 Ill., 202, 36 N. E., 1036; Erhman v. Nassau Flee. R. Co., 48 N. Y. S., 379; Lonis v. Lake Shore, etc., R. Co., 111 Mich., 458, 69 N. W., 642.
The attitude of our own Court on the subject is clearly set forth in Rosser v. Bynum, 168 N. C., 340, 344, 84 S. E., 393, and cases cited.
In Smith v. McIlwaine, 70 N. C., p. 287, the Court said: “Such rule is subject to so many exceptions as not to be of practical use; and if carelessly administered may work much mischief.” And in Reeves v. Poindexter, 53 N. C., 308, Judge Manly, speaking for the Court, said: “With respect to the rule, it is clear that its applicability to any state of facts must depend upon whether the negative testimony can be attributed to inattention, error, or defect of memory.” In the case at bar it amounts to no more than a contradiction between witnesses, and the question is “to which side, under all circumstances, is credit due?” Reeves v. Poindexter, supra. In Rosser v. Bynum, supra, these authorities are cited in support of the principle involved in our present holding. There the inquiry was directed toward the presence or absence of a certain notation on a check at the time it was given, there being evidence both ways, and the Court said: “In the case before us there is a direct contradiction between the witnesses on a material fact to which their attention was directed, and the issue should have been submitted to the jury without comment as to the existence and application of the rule referred to.”
We think the case at bar comes within the principle approved in Rosser v. Bynum, supra, and cases cited therein, and that the special instructions asked for should have been declined.
It may be noted here that much of the defendant’s evidence challenging the truth of Arlie Dunn’s statement that the signal was not given is in the negative form of statement. Witnesses said they did not see him at the scene of the wreck, yet this was thought of such substantial import that attention was drawn to this phase of the ease repeatedly throughout the charge.
In Johnson v. R. R., supra, the Court was dealing with the question whether there was a scintilla of evidence and the negative character of the evidence was a proper subject of discussion. This Court, of course, is not bound by the rule forbidding an expression of opinion, and its discussions may not always be embodied in instructions to the jury in ipsissimis verbis without danger of infringing the rule.
Rut whatever attitude is taken with regard to the propriety of instituting such a comparison, we come to the independent statement in the *55challenged instruction that plaintiffs evidence is not only negative, but that it is weak. This is an inadvertent but nevertheless direct expression ■of opinion on the weight of the evidence which entitles the plaintiff to a new trial. It is so ordered.
New trial.