The opinion of the Court was delivered by WEINTRAUB, C.J.

Defendant was convicted of murder in the second degree. The Appellate Division reversed, holding the trial court should have directed an acquittal on the ground of insanity at the time of the killing. The matter was remanded to determine whether insanity continues, defendant to be discharged if sane and to be committed if his insanity persists. 114 N.J. Super. 470 (1971). We granted the State's petition for certification. 58 N.J. 596 (1971). Although we find error in the trial court's charge requiring a new trial as to the defense of insanity, we agree with the trial court that that issue was for the jury.

I

Defendant and his victim, John Tomlinson, were students at Trenton State College. They were friends. The details of their relationship were not developed at the trial. We know that on October 18, 1969, a few hours before the homicide, they purchased hunting knives. One was used in the murder.

Defendant stabbed Tomlinson 66 times. Defendant apparently hurried from the scene when some young men happened along. He later appeared at a hospital where he asked the night supervisor whether a policeman was there, saying he wanted to talk to the police about a "murder." When the police arrived, defendant said he wanted to report a murder, that it was a stabbing, and that he did it. Defendant asked to see a psychiatrist. The witnesses as to defendant's behavior at the hospital described him as calm but tense, a police officer adding that "I didn't figure he was crazy, but I didn't think he was all there." Defendant wrote upon two sheets of paper while waiting for the police. Upon one he said he wanted to see a psychiatrist; the writing upon the other was meaningless. Defendant later gave a statement to the State Police, which the defense offered on its case as a predicate for a psychiatric opinion.

There was no dispute that defendant killed Tomlinson. The only real issue, as phrased by the defense, was "temporary insanity." It was claimed that defendant killed in a psychotic episode during which he was legally insane. The psychiatrists agreed that this was so but their opinions depended upon defendant's extra-judicial statements as to why he killed the deceased and as to delusions he said attended the event.

Defendant, however, did not testify. If he had, the jury would not have had to believe his account, and especially in the light of the facts recited above from which it could be inferred that defendant knew he had committed "murder" and wanted to report that fact to the police. That the psychiatrists relied upon defendant's extra-judicial statements of course did not establish their truth. On the contrary, if the psychiatrists depended upon those statements, their opinions were vulnerable on that account. See State v. DiPaolo, 34 N.J. 279 , 293 (1961); State v. Trantino, 44 N.J. 358, 370-371 (1965), cert. denied, 382 U.S. 993, 86 S.Ct. 573, 15 L.Ed.2d 479 (1965). Defendant alone was in a position to supply the proof and he did not.

In State v. Lucas, 30 N.J. 37 (1959), we held a psychiatrist may testify as to what the defendant told him if the psychiatric opinion depended upon the fact that the defendant said what he said or the manner in which it was said. But we stressed that if the psychiatric opinion rested upon the factual content of the defendant's statement, the value of the expert opinion would depend upon independent proof of the truth of that content. In the words of Lucas, 30 N.J. at 79-80:

* * * If it further appears that the psychiatrist's opinion hinges upon the truth of the matter asserted, rather than the fact that it was said, then the jury should be instructed that the probative value of the psychiatrist's opinion will depend upon whether there is, from all the evidence in the case, independent proof of the statement made by the accused.

See 2 Wigmore, Evidence (3d ed. 1940), § 288, p. 14; McCormick, Evidence (1954), § 228, pp. 467-68. See also State v. Whitlow, 45 N.J. 3 , 19 (1965).

We recently suggested that "a more definitive exploration of the insanity defense by both defendant and the State would be helpful in resolving the claim." State v. Risden, 56 N.J. 27 , 39 (1970). It is the jury that must ultimately decide the case. The role of the psychiatric witness is to aid the jury in the search for the truth to the extent that he, as an expert, is able to help. He should be asked to identify the facts he takes as true so that the jury may know whether to accept or to reject his opinion in accordance with the jury's findings as to the truth of those facts. And so, here, the experts should have been queried as to the role played by defendant's statement to them or to others of the reason for the killing and of the delusional ideas interlaced in that account.

We should not be thought to bar whatever expert help the psychiatrist may be able to give as to whether a defendant's narration of the event and his related thoughts is medically credible. If there is an expert basis for thus assisting the jury to decide whether to accept a defendant's assertions, the psychiatrist's testimony along that line would also be welcome. But the ultimate fact finder remains the jury.

For the reasons we have given, the trial court could not have directed a verdict of not guilty because of insanity. The jury was not obliged to accept the expert opinions. This is so because there was no evidence in the record to support critical facts upon which the psychiatric opinions apparently rested, and because, if there were, the credibility of the opinions would nonetheless remain an issue for the trier of the facts. Indeed, it would be a rare case in which the insanity issue could be determined by a court on motion to direct a verdict. A trial court, of course, has a different role if a jury's finding is later assailed as against the weight of the evidence, but it is something else to take the issue from the jury.

II

As we have already said, we think the charge to the jury was erroneous. To show that this is so, the evidence must be sketched.

Defendant's relevant history is poor. His mother had been institutionalized six times, upon a diagnosis, we are told, of schizophrenia of the paranoid type. A brother, too, had had a psychiatric problem. The family life may be summed up as unusually unfortunate.

Defendant was described as shy and withdrawn. At some point he became interested in psychology and pursued studies in that area. In August 1969, about two months before the homicide, his personality began to change. The testimony showed he took LSD on two occasions in August, with no apparent effect according to statements attributed to him. He began to smoke hashish once a week and did so up to the time of the killing. Defendant became outgoing and talkative. When his father learned of the use of drugs, he took defendant to a psychiatrist. Defendant was also referred to the clinical psychologist at the college, and this because a female member of the faculty was disturbed by defendant's undue attention to her. That psychologist thought defendant was psychotic when he last saw him, two days before the homicide.

In jail, a day or two after the killing, defendant became violent. He was committed to the New Jersey State Hospital for observation. The history in the hospital record attributed to him a purpose to kill himself because of what he had done. He seemingly tried to gouge his eyes out "because of his guilt and sin of murder," although apparently no serious injury resulted. He remained at the hospital until April 11, 1970 when, after medication and a course of shock treatments, he was found to be "in remission," and able to stand trial.

All the psychiatrists who saw defendant after the homicide ultimately agreed he suffered from schizophrenia, paranoid type. The admission staff note in the hospital record on October 21 had read:

He is obviously delusional and psychotic. Whether this is due to the lasting effects of the LSD or a true schizophrenic break, it is difficult to say at this time.

But we gather that it was later concluded defendant was not under the direct influence of LSD, but rather had experienced a "true schizophrenic break." The testimony reveals that the etiology of schizophrenia is unknown. In lay terms, defendant had an underlying condition, of long duration, which led to a break from reality when he was subjected to some stress he could not handle. We gather that the term schizophrenia was used by the witnesses to describe that psychotic episode rather than the underlying condition. We understand too that the psychiatrists agreed that after the psychotic episode abated, defendant, though "in remission," remained susceptible to further psychotic breaks if he should again be subjected to a stress his personality could not cope with. In short, the underlying condition or illness continued even though the overt indications of the schizophrenic break had disappeared.

In expressing the opinion defendant was legally insane at the time of the killing, the witnesses apparently accepted one or another of defendant's accounts of the event and of his attendant thoughts. Those accounts were not precisely the same, but for present purposes may be summarized this way: that he and the victim were engaged in tracking down dope pushers; that they agreed that would rather die at each other's hands if the peddlers they pursued should turn upon them; that they purchased the knives and also handcuffs to further their detectional efforts; that defendant suddenly believed Tomlinson wanted to die, and hence he stabbed him repeatedly despite his effort to escape the blows. One of the accounts was specially laden with religious allusions. Although defendant said he thought he and Tomlinson would quarrel over some man while they were together in defendant's automobile on the fatal evening, defendant said the quarrel did not occur and that he killed Tomlinson to gratify the victim's supposed wish to die and did so out of love for him.

The psychiatrists were unable to point to the specific factor which, operating upon the underlying illness, triggered the psychotic episode they found. The possibilities were (1) the effect of the drugs; (2) a romantic failure with a young lady whom defendant thereafter tried to "reform." The witnesses apparently assumed the psychotic episode was not precipitated by the fact of the homicide. At least that possibility was not discussed. Rather the onset of the psychotic episode was deemed to have preceded the killing, and this apparently upon defendant's recitals of his delusional thinking with respect to his victim's wish to die, related above.

In this setting the trial court charged the jury that if the psychosis was triggered by the voluntary use of LSD or hashish, the defense of insanity could not stand; that upon that premise defendant would be guilty at least of murder in the second degree. The jury returned for further instruction in this regard and the charge was repeated. The question is whether it was error thus to say that a psychosis could not lead to an acquittal if it was precipitated by the voluntary use of drugs.

A

It is well to begin with the law's concept of criminal responsibility. The common law spoke of the bad and of the sick. The premise was that a hostile act may in one case spring from wickedness and in another from some infirmity of the mind which the individual did not author. It was the moral judgment of the common law that a forbidden act should not be punished criminally unless done with mens rea, a sense of wrongdoing. The M'Naghten test of legal insanity followed hard upon that conception of crime. If at the time of committing the act, the accused was laboring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing, or if he did know it, that he did not know what he was doing was wrong, he was legally insane. State v. Wolak, 26 N.J. 464, 476-477 (1958), cert. denied, 365 U.S. 822, 81 S.Ct. 710, 5 L.Ed.2d 701 (1961); State v. White, 27 N.J. 158, 164 (1958); Aponte v. State, 30 N.J. 441 , 450 (1959); State v. Trantino, 44 N.J. 358 , 367-368 (1965), cert. denied, 382 U.S. 993, 86 S.Ct. 573, 15 L.Ed.2d 479 (1965); State v. Coleman, 46 N.J. 16 , 39 (1965). Thus afflicted, the accused did not have the evil mind required by the common law's concept of criminal accountability. The act was therefore chargeable to illness.

We need not discuss the competing concepts of legal insanity which others have projected since M'Naghten became the rule in this State. See State v. Lucas, supra, 30 N.J. 37 ; State v. Sikora, 44 N.J. 453 (1965). For present purposes it is enough to say that all the doctrines which would excuse an offender from criminal accountability because of insanity have the common characteristic of attempting to distinguish between the sick and the bad. And the distinction is made even though no scientific evidence separates the sick from the bad in terms of personal blameworthiness. Indeed, to a psychiatrist the sick and the bad are equally unfortunate. Blame is something he leaves to the moral judgment of philosophers, and they draw upon their unverifiable view of man and his endowments.

The point to be stressed is that in drawing a line between the sick and the bad, there is no purpose to subject others to harm at the hands of the mentally ill. On the contrary, the aim of the law is to protect the innocent from injury by the sick as well as the bad. The distinction bears only upon whether the stigma of criminal shall be imposed and upon the measures to be employed to guard against further transgressions. Thus our statute, N.J.S.A. 2A:163-3 directs that a defendant acquitted because of insanity shall be held in custody until "restored to reason." Indeed, an offender may fare better as bad than as sick, for, if merely bad, his maximum term may be limited even though his disposition for badness may continue, whereas, if he is adjudged sick, he may be confined for his remaining years.

It is in this context that we must decide whether the voluntary use of drugs should support a defense of insanity in the circumstances of this case, and if so, what change in the offender's basic personality makeup should be found before he is released from the grip needed to protect others from a repetition of the behavior thus attributed to sickness rather than badness.

B

It is generally agreed that a defendant will not be relieved of criminal responsibility because he was under the influence of intoxicants or drugs voluntarily taken. This principle rests upon public policy, demanding that he who seeks the influence of liquor or narcotics should not be insulated from criminal liability because that influence impaired his judgment or his control. The required element of badness can be found in the intentional use of the stimulant or depressant. Moreover, to say that one who offended while under such influence was sick would suggest that his sickness disappeared when he sobered up and hence he should be released. Such a concept would hardly protect others from the prospect of repeated injury.

There are qualifications upon that view. One is that the unexpected or bizarre result of drugs taken for medication should excuse. In such circumstances no public interest is served by punishing the individual, and if the effect of the drug has spent itself, there is no likelihood of repetition warranting deprivation of liberty on the thesis of a continuing threat. Although the subject was not discussed at any length in State v. Lynch, 130 N.J.L. 253 (E. A. 1943), that case may be read in that light. There the defense of "a total amnesia" was found to be tenable upon a claim that defendant "had suffered from chronic bromide poisoning, and that this condition became acute prior to the shooting as a result of the taking of a large quantity of bromides" ( p. 255). Presumably bromides were there intended as medication, and the so-called "temporary" insanity was an unexpected result of the medication.

Cases may also be found which say that voluntary intoxication may be a defense if it in fact prevented the formation of a "specific intent" required in the definition of a particular offense. See Annotation, 8 A.L.R.3d 1236, 1246 (1966). The distinction thus made between a "specific intent" and a "general intent" is quite elusive, and although the proposition stated in the preceding sentence is echoed in some opinions in our State, see State v. White, supra, 27 N.J. at 165-167; cf. State v. Letter, 4 N.J. Misc. 395, 133 A. 46 (Sup.Ct. 1926), it is not clear that any of our cases in fact turned upon it. A similar proposition has played a role in homicide cases. There, the voluntary use of liquor or drugs has been held to be relevant in determining whether the defendant in fact performed the mental operations necessary to raise a murder from second degree to first degree. But the influence of liquor or drugs thus voluntarily taken, no matter how pervasive that influence may be, will not lead to an acquittal. It cannot reduce the crime below murder in the second degree, and this because of the demands of public security. See State v. Wolak, supra, 26 N.J. at 477-478; State v. White, supra, 27 N.J. at 165; State v. DiPaolo, supra, 34 N.J. at 295-296; State v. King, 37 N.J. 285, 297-298 (1962); State v. Hudson, 38 N.J. 364 , 368-369 (1962); State v. Trantino, supra, 44 N.J. at 369. This is equally true as to a felony homicide. Thus a defendant who in fact participated in the felony in which the homicide occurred, can seek nothing more favorable than a conviction of murder in the second degree by proof that he could not, on that account, form the intent to commit the felony. Again, the limitation rests upon the demands for public security. State v. Sinclair, 49 N.J. 525 , 544 (1967).

Here defendant seeks to rely upon still another exception to the overall proposition that the voluntary use of such substances will not excuse. That exception is that if the use of liquor or drugs, though voluntary, results in a fixed state of insanity after the immediate influence of the intoxicant or drug has spent itself, insanity so caused will be a defense if it otherwise satisfies the M'Naghten test. State v. White, supra, 27 N.J. at 165; State v. Hudson, supra, 38 N.J. at 368-369; Annotation, 8 A.L.R.3d, supra, at 1265-1268. The premise is that the specific cause of the medical condition is too remote a consideration, and that the public interest with respect to the threat of further harm can be adequately served by securing the offender so long as that state of insanity persists.

In the case at hand, the medical thesis was somewhat different, for the underlying illness from which the psychotic episode emerged was not caused by the use of drugs. Rather the thesis was that the drugs, acting upon that underlying illness, triggered or precipitated a psychotic state which continued after the direct or immediate influence of the drug had dissipated, and that it was the psychosis, rather than the drug, which rendered defendant unable to know right from wrong at the time of the killing. In other words, defendant urges that when a psychosis emerges from a fixed illness, we should not inquire into the identity of the precipitating event or action. Indeed, it may be said to be unlikely that the inquiry would be useful, for when, as here, the acute psychosis could equally be triggered by some other stress, known or unknown, which the defendant could not handle, a medical opinion as to what did in fact precipitate the psychosis is not apt to rise above a speculation among mere possibilities.

We think it compatible with the philosophical basis of M'Naghten to accept the fact of a schizophrenic episode without inquiry into its etiology. If protection against further harm can reasonably be assured by measures appropriate for the sickness involved, it would comport with M'Naghten to deal with the threat in those terms. We turn then to the question whether, if defendant were found not guilty because of the insanity here asserted, there would be authority so to deal with him as to meet any continuing threat harbored by his illness.

C

N.J.S.A. 2A:163-3 provides that if on the trial of any indictment a defendant is acquitted because he was insane at the time of the commission of the offense, the jury shall so declare and shall find "specially by their verdict also whether or not such insanity continues," and

if the jury shall find by their verdict that such insanity does continue, the court shall order such person into safe custody and commit him to the New Jersey state hospital at Trenton until such time as he may be restored to reason. (Emphasis added.)

The critical phrase "restored to reason" is not defined, but its meaning emerges from the common sense of the subject. As we have said, the law's distinction between the sick and the bad was not designed to loose upon others a continuing threat of harm merely because the threat resided in illness. Hence when the Legislature spoke of restoration to "reason," it must have had in view M'Naghten's concept that legal insanity resides in a "defect of reason," and the legislative intent must have been that a defendant found not guilty because of such an impairment of reason shall be confined until there is assurance that the threat of that defect of reason has been eliminated.

Here the defendant asserted "temporary" insanity, presumably to limit the defense to the acute psychotic eruption to which the homicide was attributed by the psychiatrists, in the hope that he would be set free because the acute psychotic episode of schizophrenia had abated. The closing entry in the hospital record reads: "Diagnosis: Schizophrenia, Paranoid, in remission," and the psychiatrists who testified apparently subscribed to the finding of "remission." But none said defendant no longer suffered from the underlying condition which erupted into a psychotic state in response to a stress defendant could not handle. In short, there was no medical assurance that this latent personality disorder would not be triggered again into violent expression by reason of some stress defendant could reasonably be expected to experience. On the contrary, the tenor of the testimony would suggest there is no medical basis for such assurance as a probability.

It would depart from the justification for the recognition of insanity as a defense to view the psychotic explosion in isolation from the underlying illness. To do so would fail to protect the citizens from further acute episodes. The protection must be equal to the risk of further violence. An offender is not "restored to reason" unless he is so freed of the underlying illness that his "reason" can be expected to prevail. Hence the underlying or latent personality disorder, and not merely the psychotic episode which emerged from it, is the relevant illness, and the statutory requirement for restoration to reason as a precondition for release from custody is not met so long as that underlying illness continues.

The statute which here controls may be contrasted with N.J.S.A. 30:4-82 which deals with persons who, while held in confinement, shall become mentally ill. N.J.S.A. 30:4-82 provides that the court "shall direct that such person be removed from imprisonment" and taken to an institution for care and treatment, and that he shall be returned to the place of his original confinement when he "is in a condition to be discharged * * * as being in a state of remission and free of symptoms of the mental disease which required his original transfer." It of course is thus appropriate upon such "remission" to return a prisoner to the prior place of custody, there to be dealt with in accordance with the original order of commitment. But, as we have said, it would be inappropriate, upon that showing, to free a man who, found to have committed a criminal act, was acquitted from criminal accountability because of legal insanity. It is significant that the Legislature did not speak in terms of remission or freedom from symptoms in dealing with the release of a man so situated. As to him, the Legislature required in N.J.S.A. 2A:163-3 that he be held until "restored to reason."

Hence, while a psychotic episode, though temporary in the sense that a defendant may be relieved of its grip and thereupon be in "remission," will be accepted as a state of insanity which may excuse under M'Naghten, insanity continues notwithstanding remission so long as the underlying latent condition remains, and the defendant will not be "restored to reason" within the meaning of the statute unless that condition is removed or effectively neutralized if it can be.

And it is for the court, rather than for the hospital, to decide whether a release should be ordered. We refer to the subject of the situs of the responsibility for release because the statute already cited, N.J.S.A. 2A:163-3, is silent with respect to who shall order the release if the defendant is restored to reason. The legislative history makes it plain that the responsibility is with the judiciary.

For present purposes, we may start with L. 1922, c. 101. N.J.S.A. 2A:163-3 is derived from section 3 of that act. Section 3 provided expressly that a person who escaped indictment or was acquitted upon the ground of insanity and who was committed because the insanity continued "shall not be released from confinement except upon the order of the trial justice or judge having jurisdiction to try such person who has escaped indictment or has been acquitted of such criminal charge as aforesaid." That section became R.S. 2:190-17 in the 1937 revision, the closing portion being changed to read that "no person so confined shall be released from such confinement except upon the order of the court by which he was committed."

R.S. 2:190-17 was supplemented and amended by L. 1943, c. 41. The purpose was to have the jury, rather than the judge, decide whether insanity continued. R.S. 2:190-17 continued to provide that there shall be no release from custody "except upon the order of the court by which he was committed."

The statute was further amended in the Revision of 1951. As we pointed out in Aponte v. State, supra, 30 N.J. at 454-455, the purpose of that amendment was to permit a trial of the issue of insanity even though a defendant is unable to stand trial, to the end that the charge may be dismissed if it is found that he was insane at the time of the deed. The defendant is given the benefit of a finding that he was insane at the time of the deed, but is not bound by an adverse finding if he should later be able to stand trial upon the indictment. The provisions thus dealing with a defendant unable to stand trial on an indictment became N.J.S.A. 2A:163-2 in the Revision of 1951, and that section provides that if the defendant is thus found insane at the time of the offense, he shall be held until restored to reason and shall not be released "except upon the order of the court by which he was committed." So much of the earlier statute as dealt with the trial of the indictment became N.J.S.A. 2A:163-3 in the Revision of 1951. N.J.S.A. 2A:163-3 continues to direct that if the jury finds a defendant insane at the time of the offense and that the insanity continues, the court shall order him into safe custody and commit him to the state hospital until such time as he may be restored to reason. Somehow, this section did not repeat the earlier provision designating the court as the authority empowered to order release, the section being wholly silent with respect to the release authority. The question is thus suggested whether the Legislature thereby intended to place the release power elsewhere.

There is no evidence of a purpose to transfer that responsibility to another agency, and nothing in the subject matter suggests a transfer to another authority would be appropriate. We therefore conclude there was no intent to change the prior law. We note that the statute adopting the revision, L. 1951, c. 344, states in section 6 that "The provisions of said Title 2A not inconsistent with those of prior laws shall be construed as a continuation of such laws." Accordingly we conclude that one committed under N.J.S.A. 2A:163-3 cannot be released from confinement except upon the order of the court by which he was committed.

Finally, we note that we do not deal with the question whether the court may order a release conditioned upon a defendant's return to custody if signs of an oncoming acute illness should appear. At the moment, there is no reason to assume the issue will arise in this case. If adequate medical assurance could be given that supervision is reasonably feasible, that course would be humane. At this time we can do no more than note the question.

III

In summary then, the judgment of the Appellate Division is modified (1) by deleting the direction for a judgment of acquittal and (2) by ordering a new trial because of the error in the charge with respect to insanity. Except for the issue of insanity, the finding of murder in the second degree is unassailable. Indeed, defendant could not have hoped for a more favorable result. This being so, the issue of guilt, apart from the defense of insanity, is concluded by the jury's verdict in this case. The sole issues to be retried are whether defendant was legally insane at the time of the killing, and if he was, whether insanity, as we have defined it in this opinion, continues. Of course the State, and the defense, may offer proof relating to the homicide since the whole story seems relevant upon the issue of insanity, but the jury should be instructed that defendant's guilt has been found to be of murder in the second degree, subject only to the issues as to insanity.

The judgment of the Appellate Division is amended accordingly and the matter remanded to the trial court for proceedings not inconsistent herewith.

For modification and remandment — Chief Justice WEINTRAUB and Justices JACOBS, FRANCIS, PROCTOR, HALL, SCHETTINO and MOUNTAIN — 7.

Opposed — None.