Opinion
TOBRINER, J.
For over a century California has followed the M’Naghten1 test2 to define the defenses of insanity and idiocy. The deficiencies of that test have long been apparent, and judicial attempts to reinterpret or evade the limitations of M’Naghten have proven inadequate. We shall explain why we have concluded that we should discard the M’Naghten language, and update the California test of mental incapacity as a criminal defense by adopting the test proposed by the American Law Institute3 and followed by the federal judiciary and the courts of 15 states. _
[337]*337Understandably, in view of our past adherence to M’Naghten, neither the psychiatrists who examined defendant nor- the jury evaluated defendant’s capacity in terms of the ALI test. Since the evidentiary record indicates that defendant, a former mental patient with a history of irrational assaultive behavior, lacked the capacity to conform his conduct to legal requirements, we conclude that the court’s failure to instruct the jury under the ALI test was prejudicial, and therefore reverse the conviction.
Defendant Drew also contends that Evidence Code section 522, which requires a defendant to prove insanity by a preponderance of the evidence, is unconstitutional. Controlling precedent in the United States Supreme Court (see Patterson v. New York (1977) 432 U.S. 197 [53 L.Ed.2d 281, 97 S.Ct. 2319]), and in this court (see People v. Miller (1972) 7 Cal.3d 562 [102 Cal.Rptr. 841, 498 P.2d 1089]) mandates the rejection of this contention.
Finally, Drew argues that the record is insufficient to sustain the jury finding of sanity. We conclude, however, that a jury instructed under the M’Naghten rule could reasonably find that defendant failed to prove by a preponderance of the evidence that he was unaware of the wrongfulness of his conduct. Thus Drew is not entitled to an order directing the trial court to find him insane, but only to a new trial on the issue of sanity in which the jury is instructed under the ALI test.
1. Statement of Facts.
Defendant Drew, a 22-year-old man, was drinking in a bar in Brawley during the early morning of October 26, 1975. He left $5 on the bar to pay for drinks and went to the men’s room. When he returned, the money was missing. Drew accused one Truman Sylling, a customer at the bar, of taking the money. A heated argument ensued, and the bartender phoned for police assistance.
Officers Guerrero and Bonsell arrived at the bar. When Guerrero attempted to question Sylling, Drew interfered to continue the argument. Bonsell then asked Drew to step outside. Drew refused. Bonsell took Drew by the hand, and he and Officer Schulke, who had just arrived at
[338]*338the bar, attempted to escort Drew outside. Drew broke away from the officers and struck Bonsell in the face. Bonsell struck his head against the edge of the bar and fell to the floor. Drew fell on top of him and attempted to bite him, but was restrained by Guerrero and Schulke. Drew continued to resist violently until he was finally placed in a cell at the police station.
Charged with battery on a peace officer (Pen. Code, § 243), obstructing an officer (Pen. Code, § 148), and disturbing the peace (Pen. Code, § 415), Drew pled not guilty and not guilty by reason of insanity. At the guilt trial, Drew testified on his own behalf; he denied striking Bonsell and maintained that the officer’s injuries were accidental. Bonsell’s testimony, however, was corroborated by Guerrero and Sylling. The jury found Drew guilty as charged.
Two court-appointed psychiatrists testified at the sanity trial. Dr. Otto Gericke, former Medical Director at Patton State Hospital, testified that Drew was committed to that hospital for nine months in 1972 after Drew was found incompetent to stand trial on an unspecified charge. He examined him on that occasion; again on February 1, 1976, to determine Drew’s competency to stand trial on the instant charge; and a third time on June 6, 1976, on the question of Drew’s sanity.
Dr. Gericke described Drew’s condition as one of latent schizophrenia, characterized by repeated incidents of assaultive behavior and by conversing with inanimate objects and nonexistent persons; this condition could be controlled by medication but if left untreated would deteriorate to paranoid schizophrenia.
Relying upon his examinations and Drew’s medical history at Patton State Hospital, Dr. Gericke concluded that Drew was unable to appreciate the difference between right and wrong at the time he attacked Officer Bonsell.
The second witness, Dr. Ethel Chapman, was a staff psychiatrist at Patton State Hospital. She also examined Drew under court appointment in February and June of 1976, and was acquainted with him from his stay at the hospital in 1972. She concurred with Dr. Gericke’s diagnosis of his condition, adding the observation that his symptoms would be aggravated by the ingestion of alcohol, and joined in Dr. Gericke’s conclusion that Drew did not understand that his assault upon Officer Bonsell was wrong.
[339]*339The prosecution presented no evidence at the sanity trial. Nevertheless the jury, instructed that the defendant has the burden of proving insanity under the M’Naghten test, found him sane. The court thereupon sentenced Drew to prison on the battery conviction. He appeals from the judgment of conviction.
2. This court should adopt the American Law Institute test, as stated in section 4.01, subpart (1) of the Model Penal Code, to define the defense of insanity.
The trial court instructed the jury that - “Legal insanity . . . means a diseased or deranged condition of the mind which makes a person incapable of knowing or understanding the nature and quality of his act, or makes a person incapable of knowing or understanding that his act was wrong.”4 We explain that this instruction, based on the M’Naghten test, was erroneous, and on the record before us constitutes prejudicial error requiring reversal of the judgment.
The purpose of a legal test for insanity is to identify those persons who, owing to mental incapacity, should not be held criminally responsible for their conduct. The criminal law rests on a postulate of free will—that all persons of sound mind are presumed capable of conforming their behavior to legal requirements5 and that when any such person freely chooses to violate the law, he may justly be held responsible. (Seb Goldstein, The Insanity Defense (1967) pp. 9-10.) From the earliest days of the common law, however, the courts have recognized that a few persons lack the mental capacity to conform to the strictures of the láw. Thus in 1582 William Lambart of Lincoln’s Inn wrote that “If. . . a mad man or a natural fool, or a lunatic in the time of his lunacy, or a child who apparently had no knowledge of good or evil, do kill a man, this is no felonious act . . . for they cannot be said to have any understanding will.” (Lambart, Eirenarcha (1582) Cat. 21.218. (Spelling modernized).) [340]*340The principle that mental incapacity constitutes a defense to crime is today accepted in all American jurisdictions. (See Weihofen, Mental Disorder as a Criminal Defense (1954) p. 51.)
The California Penal Code codifies the defense of mental incapacity. Section 20 states that “[i]n every crime . . . there must exist a union . . . of act and intent.” Section 21 provides as to persons of sound mind “[t]he intent... is manifested by the circumstances connected with the offense” and that “All persons are of sound mind who are neither idiots nor lunatics, nor affected with insanity.” Finally section 26 specifies that “All persons are capable of committing crimes except those belonging to the following classes” and includes among those classes “Idiots” and “Lunatics and insane persons.”6
Although the Legislature has thus provided that “insanity” is a defense to a criminal charge, it has never attempted to define that term. The task of describing the circumstances under which mental incapacity will relieve a defendant of criminal responsibility has become the duty of the judiciary.
Since People v. Coffman (1864) 24 Cal. 230, 235, the California courts have followed the M’Naghten rule to define the defense of insanity. The curious origin of the M’Naghten rule has been frequently recounted. (See, e.g., United States v. Freeman (2d Cir. 1966) 357 F.2d 606, 616-617; Weihofen, Mental Disorder as a Criminal Defense, supra, pp. 59-63.) In 1843 Daniel M’Naghten, afflicted with paranoia, attempted to assassinate the Prime Minister of England, and succeeded in killing the Prime [341]*341Minister’s secretary. M’Naghten’s acquittal on grounds of insanity so disturbed Queen Victoria that she summoned the House of Lords to obtain the opinion of the judges on the law of insanity. The 15 judges of the common law courts were called in an extraordinary session, “under a not subtle atmosphere of pressure” (United States v. Freeman, supra, 357 F.2d 606, 617), to answer five hypothetical questions on the law of criminal responsibility.
In response to two of the questions propounded the judges stated that “to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing the act, the parly accused was labouring 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 he was doing what was wrong.” (M’Naghten’s Case, supra, 10 Clark & Fin. 200, 210 [8 Eng. Rep. 718, 722].) Although an advisory opinion, and thus most questionable authority (see 2 Stephen, History of the Criminal Law of England (1883) p. 153), this language became the basis for the test of insanity in all American states except New Hampshire. (Weihofen, Mental Disorder as a Criminal Defense, supra, pp. 68-69.)
Despite its widespread acceptance, the deficiencies of M’Naghten have long been apparent. Principal among these is the test’s exclusive focus upon the cognitive capacity of the defendant, an outgrowth of the then current psychological theory under which the mind was divided intd separate independent compartments, one of which could be diseased without affecting the others. (See United States v. Currens (3d Cir. 1961) 290 F.2d 751, 766.) As explained by Judge Ely of the Ninth Circuit: “The M’Naghten rules fruitlessly attempt to relieve from punishment only those mentally diseased persons who have no cognitive capacity. . . . This formulation does not comport with modem medical knowledge that an individual is a mentally complex being with varying degrees of awareness. It also fails to attack the problem presented in a case wherein an accused may have understood his actions but was incapable of controlling his behavior. Such a person has been allowed to remain a danger to himself and to society whenever, under M’Naghten, he is imprisoned without being afforded such treatment as may produce rehabilitation and is later, potentially recidivistic, released.” (Wade v. United States (9th Cir. 1970) 426 F.2d 64, 66-67.) (Fns. omitted.)7
[342]*342M’Naghten’s exclusive emphasis on cognition would be of little consequence if all serious mental illness impaired the capacity of the affected person • to know the nature and wrongfulness of his action. Indeed, the early decision of People v. Hoin (1882) 62 Cal. 120, 123, in rejecting the defense of “irresistible impulse,” rested on this gratuitous but doubtful assumption. Current psychiatric opinion, however, holds that mental illness often leaves the individual’s intellectual understanding relatively unimpaired, but so affects his emotions or reason that he is unable to prevent himself from committing the act. (See United States v. Smith (6th Cir. 1968) 404 F.2d 720, 725.) “[I]nsanity does not only, or primarily, affect the cognitive or intellectual faculties, but affects the whole personality of the patient, including both the will and the emotions. An insane person may therefore often know the nature and quality of his act and that it is wrong and forbidden by law, and yet commit it as a result of the mental disease.” (Rep. Royal Com. on Capital Punishment, 1949-1953, p. 80.)
The annals of this court are filled with illustrations of the above statement; the deluded defendant in People v. Gorshen, supra, 51 Cal.2d 716, who believed he would be possessed by devilish visions unless he killed his foreman; the schizophrenic boy in People v. Wolff, supra, 61 Cal.2d 795, who knew that killing his mother was murder but was unable emotionally to control his conduct despite that knowledge; the defendant in People v. Robles (1970) 2 Cal.3d 205 [85 Cal.Rptr. 166, 466 P.2d 710], suffering from organic brain damage, who mutilated himself and killed others in sudden rages. To ask whether such a person knows or understands that his act is “wrong” is to ask a question irrelevant to the nature of his mental illness or to the degree of his criminal responsibility.
Secondly, “M’Naghten’s single track emphasis on the cognitive aspect of the personality recognizes no degrees of incapacity. Either the defendant knows right from wrong or he does not. . . . But such a test is grossly unrealistic. ... As the commentary to the American Law Institute’s Model Penal Code observes, ‘The law must recognize that when there is no black and white it must content itself with different shades of gray.’ ” (United States v. Freeman, supra, 357 F.2d 606, 618-619, quoting ALI, Model Pen. Code (Tent. Drafts, Nos. 1, 2, 3, and 4, p. 158.)
In short, M’Naghten purports to channel psychiatric testimony into the narrow issue of cognitive capacity, an issue often unrelated to the [343]*343defendant’s illness or crime. The psychiatrist called as a witness faces a dilemma: either he can restrict his testimony to the confines of M’Naghten, depriving the trier of fact of a full presentation of the defendant’s mental state (see United States v. Freeman, supra, 357 F.2d 606, 620; Durham v. United States, supra, 214 F.2d 862, 874), or he can testify that the defendant cannot tell “right” from “wrong” when that is not really his opinion because by so testifying he acquires the opportunity to put before the trier of fact the reality of defendant’s mental condition. (See Brakel & Rock, The Mentally Disabled and the Law (2d ed. 1971) p. 387; Diamond, Criminal Responsibility of the Mentally Ill (1961) 14 Stan.L. Rev. 59, 60-61.) As Justice Frankfurter stated before the Royal Commission on Capital Punishment, “I think to have rules which cannot rationally be justified except by a process of interpretation which distorts and often practically nullifies them ... is not a desirable system. . . . [T]he M’Naghten Rules are in large measure shams. That is a strong word, but I think the M’Naghten Rules are very difficult for conscientious people and not difficult enough for people who say ‘We’ll just juggle them.’ ” (Royal Com. on Capital Punishment, op.cit. supra, p. 102.)
Even if the psychiatrist is able to place before the trier of fact a complete picture of the defendant’s mental incapacity, that testimony reaches the trier of fact weakened by cross-examination designed to show that defendant knew right from wrong (see, for example, the cross-examination described in People v. Wolff, supra, 61 Cal.2d 795, 812-814) and limited by the M’Naghten instruction. As a result, conscientious juries have often returned verdicts of sanity despite plain evidence of serious mental illness and unanimous expert testimony that the defendant was insane. (See People v. Wolff, supra, 61 Cal.2d 795, 812 and cases there cited.)
Conscious of the inadequacies of the M’Naghten test, California decisions have modified that test in two significant respects. First, in People v. Wolff, supra, 61 Cal.2d 795, we held that the mere capacity to verbalize socially acceptable answers to questions did not prove sanity; the defendant must not only know but also “appreciate” or “understand” the nature and wrongfulness of his act. (Pp. 800-801.) Second, in a series of decisions dating from People v. Wells (1949) 33 Cal.2d 330 [202 P.2d 53] and People v. Gorshen, supra, 51 Cal.2d 716, we developed the concept of diminished capacity, under which a defendant can introduce evidence of mental incapacity to negate specific intent, malice, or other subjective elements of the charged crime. Recently in People v. Cantrell (1973) 8 Cal.3d 672 [105 Cal.Rptr. 792, 504 P.2d 1256], we expressly held that [344]*344“irresistible impulse”—a concept evolved to supply the volitional element lacking in the M’Naghten test—can be utilized to prove diminished capacity. (Pp. 685-686.)
But these innovative modifications to the M’Naghten rule fail to cure its basic defects. Wolff ameliorates only one of the rigid categories of M’Naghten; as Professor Sherry explains: “It still . . . falls short of acknowledging the teaching of psychiatry that mental aberration may not only impair knowledge of wrongfulness but may very well destroy an individual’s capacity to control or to restrain himself.” (Sherry, Penal Code Revision Project—Progress Report (1968) 43 State Bar J. 900, 916.) The doctrine of diminished capacity, once hailed as a possible replacement for the defense of insanity (see Diamond, Criminal Responsibility of the Mentally Ill (1961) 14 Stan. L.Rev. 59), can now be seen to create its own problems.
The availability of a defense of diminished capacity turns largely on the nature of the crime charged. If the defendant is charged with a general intent crime, he cannot raise a defense of diminished capacity regardless of his impaired mental state. (People v. Noah (1971) 5 Cal.3d 469, 477 [96 Cal.Rptr. 441, 487 P.2d 1409]; People v. Nance (1972) 25 Cal.App.3d 925, 929 [102 Cal.Rptr. 266].) If charged with a specific intent crime, he may be able to reduce the offense to a lesser included general intent crime. If evidence of diminished capacity is used to negate criminal intent in a crime which contains no lesser offense, however, the defendant may secure his outright acquittal and release. (See Goldstein, The Insanity Defense, supra, pp. 201-202.) The effectiveness of the defense, and the disposition of the defendant, thus turn less on the nature and ' seriousness of the defendant’s mental disability than on the technical structure of the criminal law.
A defendant whose criminal activity arises from mental illness or defect usually requires confinement and special treatment. Penal Code sections 1026 and 1026a provide such confinement and treatment for persons acquitted on grounds. of insanity. A successful diminished capacity defense, on the other hand, results either in the release of the defendant or his confinement as an ordinary criminal for a lesser term. Because the diminished capacity defense thus fails to identify the mentally disturbed defendant, it may result in the defendant’s not receiving the care appropriate to his condition. Such a defendant, who may still suffer from his mental disturbance, may serve his term, be. released and thus permitted to become a danger to the public. (See [345]*345Goldstein, The Insanity Defense, supra, p. 202; Brakel & Rock, The Mentally Disabled and the Law, supra, p. 395.)
In our opinion the continuing inadequacy of M’Naghten as a test of criminal responsibility cannot be cured by further attempts to interpret language dating from a different era of psychological thought, nor by the creation of additional concepts designed to evade the limitations of M’Naghten. It is time to recast M’Naghten in modem language, taking account of advances in psychological knowledge and changes in legal thought.
The definition of mental incapacity appearing in section 4.01 of the American Law Institute’s Model Penal Code represents the distillation of nine years of research, exploration, and debate by the leading legal and medical minds of the country. (United States v. Freeman, supra, 357 F.2d 606, 622.) It specifies that “A person is not responsible for criminal conduct if at the time of such conduct as a result of mental disease or defect he lacks substantial capacity either to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the requirements of law.”8
Adhering to the fundamental concepts of free will and criminal responsibility, the American Law Institute test restates M’Naghten in language consonant with current legal and psychological thought. (Goldstein, The Insanity Defense, supra, p. 87.) It has won widespread acceptance, having been adopted by every federal circuit except for the first circuit9 and by 15 states.10
[346]*346“In the opinion of most thoughtful observers this proposed test [the ALI test] is a significant improvement over M’Naughton.” (People v. Kelly (1973) 10 Cal.3d 565, 581-582 [111 Cal.Rptr. 171, 516 P.2d 875] (Mosk, J. conc.).) The advantages may be briefly summarized. First, the ALI test adds a volitional element, the ability to conform to legal requirements, which is missing from the M’Naghten test. Second, it avoids the all-or-nothing language of M’Naghten and permits a verdict based on lack of substantial capacity. (See Blake v. United States, supra, 407 F.2d 908, 914; United States v. Shapiro, supra, 383 F.2d 680, 685.) Third, the ALI test is broad enough to permit a psychiatrist to set before the trier of fact a full picture of the defendant’s mental impairments and flexible enough to adapt to future changes in psychiatric theoiy and diagnosis. Fourth, by referring to the defendant’s capacity to “appreciate” the wrongfulness of his conduct the test confirms our holding in People v. Wolff, supra, 61 Cal.2d 795, that mere verbal knowledge of right and wrong does not prove sanity. Finally, by establishing a broad test of nonresponsibility, including elements of volition as well as cognition, the test provides the foundation on which we can order and rationalize the convoluted and occasionally inconsistent law of diminished capacity. (See discussion in Comment, Keeping Wolff From the Door: California’s Diminished Capacity Concept (1972) 60 Cal.L.Rev. 1641, 1649; Comment, Diminished Capacity: Its Potential Effect in California (1970) 3 Loyola L.A. L.Rev. 153, 154.)
In light of the manifest superiority of the ALI test, the only barrier to the adoption of that test we perceive lies in the repeated judicial declarations that any change in the M’Naghten rule requires legislative action. (See People v. Wolff, supra, 61 Cal.2d 795, 803; People v. Darling (1962) 58 Cal.2d 15, 23 [22 Cal.Rptr. 484, 372 P.2d 316]; People v. Rittger (1960) 54 Cal.2d 720, 732 [7 Cal.Rptr. 901, 355 P.2d 645]; People v. Nash, supra, 52 Cal.2d 36, 48-49; People v. Berry (1955) 44 Cal.2d 426, 433 [282 P.2d 861]; People v. Daugherty (1953) 40 Cal.2d 876, 894 [256 P.2d 911].) This pronouncement rests on two bases: the lengthy history of the M’Naghten rule in California (see People v. Daugherty, supra) and the failure of the 1927 Legislature, when it revised the procedures for pleading and trying the defense of insanity, to overturn the M’Naghten test (see People v. Nash, supra).
[347]*347The concept that an extended line of judicial decisions, accompanied by legislative inaction, can freeze the evolution of judicial principles, divesting the courts of authority to overturn their prior decisions, is not in good repute. In Li v. Yellow Cab Co. (1975) 13 Cal.3d 804 [119 Cal.Rptr. 858, 532 P.2d 1226, 78 A.L.R.3d 393], when we adopted a rule of comparative negligence, we rejected the argument that any change in the law of contributory negligence was a matter for the Legislature, and overturned more than a century of precedent. In Rodriguez v. Bethlehem Steel Corp. (1974) 12 Cal.3d 382 [115 Cal.Rptr. 765, 525 P.2d 669], we expressly repudiated the assertion that recognition of a spousal action for loss of consortium required legislative action (see pp. 393-395) and, overruling prior decisions, endorsed that cause of action. In fact the decision of Cole v. Rush (1955) 45 Cal.2d 345 [289 P.2d 450, 54 A.L.R.2d 1137], on which People v. Nash, supra, 52 Cal.2d 36, 43 relied for its theory that legislative failure to overturn M’Naghten immunized that doctrine from judicial revision, was itself overruled in Vesely v. Sager (1971) 5 Cal.3d 153, 167 [95 Cal.Rptr. 623, 486 P.2d 151].
The principle underlying the cited cases is that the judiciary has the responsibility for legal doctrine which it has created.11 (See Rodriguez v. Bethlehem Steel Corp., supra, 12 Cal.3d 382, 393; People v. Pierce (1964) 61 Cal.2d 879, 882 [40 Cal.Rptr. 845, 395 P.2d 893].) The power of the court to reshape judicial doctrine does not authorize us to overturn constitutionally valid, statutes. But as Justice Mosk explained in his concurring opinion in People v. Kelly, supra, 10 Cal.3d 565, 580, the M’Naghten rule is not an integral part of the statutory structure of California criminal law. The Legislature has never enacted the M’Naghten rule as a test of insanity, and its provisions relating to criminal responsibility do not incorporate the M’Naghten formula. (See Pen. Code, §§ 26, 1016, 1367.) Thus replacement of the M’Naghten rule with the ALI test will not contradict or nullify any legislative enactment.
The concept expressed by some California courts, that any change must emanate from the Legislature, ignores the fact that the courts have already departed substantially from the M’Naghten formula. As we explained earlier, we rejected in People v. Wolff, supra, 61 Cal.2d 795, the [348]*348portion of M’Naghten holding that a defendant who “knows” his act is wrong is sane, and held that he must “understand” or “appreciate” the wrongfulness of his act. We also created the doctrine of diminished capacity specifically to “ameliorate the law governing criminal responsibility prescribed by the M’Naghten rule” (People v. Henderson (1963) 60 Cal.2d 482, 490 [35 Cal.Rptr. 77, 386 P.2d 677]), and included in that doctrine concepts of irresistible impulse which the M’Naghten test rejects. Thus adoption of the ALI rule does not represent a radical break from precedent, but simply an additional step in the constant revision and updating of the test of criminal responsibility to correspond to current legal and psychological thought.
Five years ago Justice Mosk, in his concurring opinion in People v. Kelly, supra, 10 Cal.3d 565, 578, called upon this court to depart from the M’Naghten formulation and adopt the test proposed by the American Law Institute. For the foregoing reasons, derived in large part from the analysis in Justice Mosk’s concurrence, we now conclude that the California courts should employ the ALI test to define the defense of insanity.12 This decision will apply retroactively only to those cases not yet final in which the defendant has pled not guilty by reason of insanity and to cases that have not yet come to trial as of the date of the finality of this opinion.
3. The defendant retains the burden of proof on the issue of insanity.
Evidence Code section 522 provides explicitly that “The party claiming that any person, including himself, is or was insane has the burden of proof on that issue.” The trial judge in the present case accordingly charged the jury that “the defendant has the burden of proving his legal insanity by a preponderance of the evidence.”
Drew contends that the court’s instruction denied him due process of law under the Fourteenth Amendment. He relies on Mullaney v. Wilbur (1975) 421 U.S. 684 [44 L.Ed.2d 508, 95 S.Ct. 1881], in which the United States Supreme Court struck down a Maine statute which required a homicide defendant to prove that he acted in the heat of passion to reduce the offense to manslaughter; the court’s language suggested broadly that the state must prove beyond a reasonable doubt every fact [349]*349critical to the guilt of the offender. More recent decisions of the Supreme Court have confirmed, however, that notwithstanding the broad dictum of Mullaney, “it remained constitutional to burden the defendant with proving his insanity.” (Patterson v. New York, supra, 432 U.S. 197, 205 [53 L.Ed.2d 281, 289, 97 S.Ct. 2319, 2324]; see Rivera v. State (Del. 1976) 351 A.2d 561, app. dism. sub nom. Rivera v. Delaware (1976) 429 U.S. 877 [50 L.Ed.2d 160, 97 S.Ct. 226].)
Drew further contends that requiring him to bear the burden of proving insanity violates the due process clause of the California Constitution. (Cal. Const., art. I, § 7.) California courts, however, have consistently upheld the constitutionality of our rule placing the burden of proof on the defendant (see, e.g., People v. Hickman (1928) 204 Cal. 470, 477-478 [268 P. 909, 270 P. 1117]). Recently in People v. Miller, supra, 7 Cal.3d 562, 574, we unanimously rejected a defendant’s contention that the rule conflicted with due process requirements. The validity of this settled line of authority was called into question only because of the broad language of the United States Supreme Court opinion in Mullaney v. Wilbur, supra, 421 U.S. 684; following that court’s narrow interpretation of Mullaney in Patterson v. New York, supra, and its confirmation that a state may constitutionally require a defendant to prove insanity, doubts respecting the constitutionality of the California rule have been laid to rest.13
4. The record supports the jury’s finding that defendant was sane under the M’Naghtén test of insanity.
Defendant Drew argues that even under the M’Naghten test the jury’s finding of sanity is not supported by substantial evidence. If Drew should prevail in this contention, he would be entitled to an order directing the trial court to find him insane, thus avoiding a retrial of the case under the ÁLI test.14 Thus although we have today rejected the M’Naghten rule, we must nevertheless determine whether the jury’s verdict based on that rule [350]*350is supported by the record. We therefore explain our conclusion that on the present record a jury instructed under the M’Naghten rule could reasonably reject the opinions of psychiatric witnesses; finding that Drew had thus failed to prove his lack of understanding of the nature or wrongfulness of his act, the jury accordingly could return a verdict of sanity.
Drew relies on the fact that both court-appointed psychiatrists testified that he was unaware of the wrongfulness of his assault. The jurors, however, are not automatically required to render a verdict which conforms to the expert opinion. We explained in People v. Wolff, supra, 61 Cal.2d 795, that: “However impressive this seeming unanimity of expert opinion may at first appear . . . our inquiry on this just as on other factual issues is necessarily limited at the appellate level to a determination whether there is substantial evidence in the record to support the jury’s verdict of sanity . . . under the law of this state. [Citations.] It is only in the rare case when ‘the evidence is uncontradicted and entirely to the effect that the accused is insane’ (In re Dennis (1959) 51 Cal.2d 666, 674 [335 P.2d 657]) that a unanimity of expert testimony could authorize upsetting a jury finding to the contrary.” (61 Cal.2d at p. 804.) Indeed we have frequently upheld on appeal verdicts which find a defendant to be sane in the face of contrary unanimous expert opinion. (See People v. MacPherson (1970) 2 Cal.3d 109, 114 [84 Cal.Rptr. 129, 465 P.2d 17]; People v. Coogler (1969) 71 Cal.2d 153, 166 [77 Cal.Rptr. 790, 454 P.2d 686]; People v. Ford (1966) 65 Cal.2d 41, 55 [52 Cal.Rptr 228, 416 P.2d 132]; People v. Wolff, supra, 61 Cal.2d 795, 803-804, and cases there cited; People v. Ruiz (1970) 11 Cal.App.3d 852, 858 [82 Cal.Rptr. 408].)
In People v. Coogler, supra, 71 Cal.2d 153 [77 Cal.Rptr. 790, 454 P.2d 686], we pointed out that “The chief value of an expert’s testimony in this field, as in all other fields, rests upon the material from which his opinion is fashioned and the reasoning by which he progresses from his material to his conclusion.” (71 Cal.2d at p. 166, quoting People v. Bassett (1968) 69 Cal.2d 122, 141 [70 Cal.Rptr. 193, 443 P.2d 777].) (Italics in original.) In the present case the jurors might well note that both experts were unfamiliar with Drew’s conduct during the four years following his release from Patton State Hospital, and that their subsequent examinations óf him were relatively brief. More significantly, the jurors could note that although both psychiatrists stated an opinion that Drew did not appreciate the wrongfulness of his act, nothing in their testimony explained the reasoning which led to this opinion. Although the psychiatric testimony described Drew’s repeated aggressive acts, and [351]*351diagnosed his condition as one of latent schizophrenia, neither psychiatrist explained why that behavior and diagnosis would lead to the conclusion that Drew was unable to appreciate the wrongfulness of his aggressive acts.
The prosecution presented no evidence at the sanity trial. Defendant, however, has the burden of proof on the issue of insanity; if neither party presents credible evidence on that issue the jury must find him sane. Thus the question on appeal is not so much the substantiality of the evidence favoring the jury’s finding as whether the evidence contrary to that finding is of such weight and character that the jury could not reasonably reject it. Because the jury could reasonably reject the psychiatric opinion that Drew was insane under the M’Naghten test on the ground that the psychiatrists did not present sufficient material and reasoning to justify that opinion, we conclude that the jury’s verdict cannot be overturned as lacking support in the trial record.
Our conclusion that the jury could reasonably find Drew sane under the M’Naghten rule, however, is itself a commentary on the inadequacy of that rule. In order to meet his burden of proof under M’Naghten, defendant had to prove his lack of cognitive capacity at the moment of the assault. This condition, however, is not a matter which psychiatrists can detect by testing or interview; neither is such cognitive incapacity a characteristic symptom of defendant’s illness. The psychiatric account of the history and characteristics of Drew’s illness does not bear directly on his ability to understand the difference between right and wrong because that alleged inability is not a relevant psychiatric fact. In sum, inadequacy of the expert evidence to prove insanity under the M’Naghten rule is essentially the result of the fact that M’Naghten requires proof of a subjective cognitive state which is largely unrelated to the reality of mental illness.
5. Disposition of this appeal.
It is not surprising that in view of the fact that we had not then endorsed the ALI test of mental incapacity neither witnesses nor counsel structured their presentation at trial in terms of the ALI test, and the court did not instruct the jury on that standard. The record on appeal, nevertheless, adduces substantial evidence of incapacity under the ALI criteria. His pattern of repetitive irrational assaults suggests the likelihood that he is unable to control his behavior to conform to legal requirements. His incapacity to control his behavior, moreover, is psychologically [352]*352relevant to the diagnosis and treatment of his condition; thus the defense psychiatrists, who observed Drew’s behavior and succeeded temporarily in controlling it through medication, might well have been able to state and support an opinion that he was insane under the ALI test. In view of the absence of prosecution evidence on the insanity issue, we conclude that if the case had been tried under the ALI standard and the jury instructed accordingly, it probably would have returned a verdict finding Drew insane. The trial court’s failure to employ the ALI test therefore constitutes prejudicial error.15
Finally, we recognize that in setting out a legal test to decide whether or not a person is insane we deal in a matter so delicate and obscure that it cannot be captured in a perfect definition. Yet because of the grave, and often life and death consequences that follow from a decision as to the sanity of an offender, we are surely enjoined to spare no effort to frame the best standard that is currently extant. We cannot justify the retention of a test based upon the single factor of recognition that has been abandoned by almost all the experts in the field and that has been rejected by all except one of the federal circuits and by an impressive number of state courts.
Nor can we escape our obligation by relegating the problem to the Legislature in face of the fact that the M’Naghten test is a court-made test that this court itself has adopted and at times reiterated and at other times amended and variously construed. We cannot continue to cast human beings in an ancient and discarded psychological mould. We must at least to the best of our limited ability accept the reality of the human psyche, as expert opinion depicts it, and bring the law as close as possible to an appraisal of the human being. We must recognize in certain cases his substantial incapacity either to appreciate the criminality of his conduct or to conform his conduct to the requirements of law. In judgment upon the sanity of the fragile and often inadequate human being we cannot be frozen into a stereotyped, rejected formula of the past.
The judgment is reversed and the cause remanded for a new trial on the issue raised by defendant’s plea of not guilty by reason of insanity.
Bird, C. J., Mosk, J., and Newman, J., concurred.