JENKS, J.
The plaintiff, a young married woman, has recovered a verdict for an assault and battery. The plaintiff’s evidence would *993have warranted a finding that the purpose of the defendant was a rape, which was fraustrated by resistance, by outcries, and by the coming up of the husband. The defendant testified that the plaintiff assaulted him because he had chided her for profanity. . On his cross-examination he testified that he did not know whether his companion on the occasion of the assault was a married man, and he was immediately asked: “Do you live with your wife? (Objected to.) The Court: That is, did he at the time, do you mean, or now? Plaintiff’s Counsel: Yes. (Objection overruled. Exception.)” The witness was thereupon asked: “Did you live with your wife then? A. No, sir. Q. How long haven’t you been living with your wife? A. About six years.” The plaintiff had not testified that he was married, and therefore the testimony objected to did not relate to evidence elicited on the direct examination. In Dore v. Babcock (Conn.) 50 Atl. 1016, in an action for negligence, a witness was asked whether his wife lived with him, and whether she had not obtained from him a divorce on the ground of desertion. The witness answered under objection and exception. The supreme court of errors of Connecticut held that the evidence was irrelevant, and therefore inadmissible to show bad moral character, and that it did not tend to prove any such general bad moral character as would or ought to affect the character of the witness for veracity, and that it should have been excluded, saying that “a man may be in many ways and in very different degrees a bad citizen, an unkind husband and father, and an unfaithful friend, and yet be a truthful witness.” I think, then, in the case at bar, the question was not admissible as affecting credibility. It could not be regarded as in the nature of a mere introductory question, in order to determine the status of the witness, akin to the questions so commonly asked, as the full name, the residence, or the age of a witness. The defendant had just been asked whether his companion was married. I infer, rather, that the questions were admitted as relevant and material to establish the commission of the act. But I think that the fact that a man has lived for some years past in nonaccess to his wife, or at the time in question is so living, is not evidence to establish, or tending to establish, the commission of an assault and battery preliminary to rape. Though such facts may afford a subject for the disquisitions of sociologists or physiologists upon the questions of celibacy, continence, morality, or lawlessness, they -are too remote, too subtle, too speculative for the consideration of a jury. We are morally certain that in countless instances celibacy means chastity, and it is common knowledge that the criminal courts frequently show outrages committed upon women by married men. It cannot be contended either that bachelorhood in itself, or the nonaccess of a married man in itself, is any evidence that the man would attempt an assault and battery upon a woman, either with or without the purpose of rape. It may be asked, if the fact is so remote, how could the defendant have been prejudiced? My answer is that, though such is my view of the force of such testimony, I am not clear how much or how little the testimony might affect the jury. Where testimony is admitted by the court over objection, *994the jury properly take such testimony as evidence in such a case as this, and it is impossible to say that such testimony, when stamped by the court as evidence, might not have affected the jury. Certainly there are cases where the court can safely conclude that improper testimony did not affect the result, but this, in my opinion, is not one of them. Anderson v. Railroad Co., 54 N. Y. 334 Neudecker v. Kohlberg, 81 N. Y. 296, 304.
I do not deem it necessary to discuss the further points raised by the learned counsel for the appellant, save his contention that the learned court erred in charging the jury that exemplary damages might be awarded. I think that the learned court was right, for the reason that the jury might well have found upon the evidence that the acts of the defendant were wanton and malicious. Conners v. Walsh, 131 N. Y. 590, 592, 30 N. E. 59. In the only authority cited' contra by the learned counsel for the appellant the plaintiff was a trespasser, and the defendant committed an assault in removing him from his premises under circumstances which, the court said, palliated, if they did not wholly justify, the offense. And in that very case the court further say:
“Whether the doctrine which permits vindictive, or, as they are termed here, punitive or exemplary, damages, can, in any action of this character, be justified upon principle, it is not necessary to inquire; for, assuming that it is so firmly established as to make any save legislative investigation useless (Hunt v. Bennett, 19 N. Y. 173; Hamilton v. Railroad Co., 53 N. Y. 25), we think that it has no application to the case before us, and that the learned trial judge erred in submitting it to the jury as one which, in any aspect, could be responded to by such allowance.” ICiff v. Youmans, 86 N. Y. 324, 40-Am. Rep. 543.
Plainly, if any case of assault admits of punitive damages, it is such a one as this, when, if the plaintiff’s evidence is to be credited, a violent assault and battery, with the purpose of rape, was made upon a young married woman, and continued after she had resisted, had declared herself, and had cried for help, the violence being accompanied by profane and lewd language which would be resented by the commonest woman of the town. But as I am of opinion that we cannot conclude that the error in the admission of the testimony ,was not prejudicial to the defendant, I advise that we order a new trial.
The judgment should be reversed, and a new trial granted, costs to abide the event. All concur, except HIRSCHBERG, J., who dissents.