The only substantial question for the jury to decide was whether the defendant was sane or insane when he took his sister's life. Strong support for a reasonable doubt upon this subject was found in the absence of apparent motive; the treatment of the defendant for insanity at an insane asylum about a year before the homicide; the opinion of the superintendent of that institution that he was insane then and insane when the killing took place; the rejection of the defendant, six months before the homicide when an applicant for life insurance, upon the ground that he was insane; the opinion of the medical examiner, who was also an examiner in lunacy, that he was insane then as well as at the time of the alleged murder; the opinion of another expert to the same effect and the strange habits, conduct and conversation of the accused when there was no reason to believe he was feigning.
Upon the controlling question of insanity the district attorney was allowed to show, not by way of proper cross-examination but as a part of his own case, the opinion of a jail prisoner that certain important acts of the defendant, not observed by the witness but stated to him in a hypothetical question, were rational under the circumstances. This witness was even allowed to state in answer to one question, not that the acts assumed and recited therein were rational, but that the man who did the acts was rational. All this evidence was objected to upon the proper ground and exceptions were duly taken to the rulings which virtually permitted an ignorant lay witness to give his opinion as to the mental condition of the defendant.
Neither argument nor authority is needed to show that this was wrong. No one disputes it. The prevailing opinion demonstrates it. It stands as a conceded fact that the defendant was convicted upon evidence, a part of which was clearly erroneous. The conviction cannot be sustained unless we are able to adjudge that this evidence did no harm to the defendant. How can we so decide, as the evidence, though incompetent, was material, for it bore upon the vital point of the case *Page 244 and it was the duty of the jury to consider it. The error was not technical, for it allowed a non-expert to give expert evidence as to the defendant's sanity. While we may overlook erroneous evidence received without objection, and even objections and exceptions which do not affect a substantial right, we cannot disregard an error founded upon an exception duly taken, unless we can say judicially that it "could not in reason have changed the result." (People v. Bonier, 179 N.Y. 315, 324.) As we have recently said, "a presumption of injury conclusively arises whenever it is apparent that the erroneous ruling may have affected the verdict." (Id.) Could this evidence, in view of the strong probability that the defendant was insane but a short time before the homicide, have aided in removing a reasonable doubt? Can we say that it did not? Can we say it had no power to persuade, or to influence, or to turn the scale against the defendant? Can we say that a witness, called by the defendant and presumed to be friendly to him, in characterizing his actions as rational and in giving an opinion that he was rational, could not have offset to some extent the opinions of the defendant's experts, whose testimony was subject to the popular prejudice against professional witnesses? Can we look into the minds of the jury and say that they gave no heed to the opinion of the defendant's friend, who saw him daily when a fellow-prisoner in jail?
I cannot answer these questions, and, hence, I cannot concur in the conclusion that, although the evidence was erroneous and the objections thereto should have been sustained, still the judgment is right, because it is reasonably certain that no harm came to the defendant. I vote to reverse.
GRAY, HAIGHT and WERNER, JJ., concur with CULLEN, Ch. J.; O'BRIEN and BARTLETT, JJ., concur with VANN, J.
Judgment of conviction affirmed. *Page 245