Criminal Responsibility Defenses and Standards

Although the insanity defense is numerically insignificant, it remains profoundly important to the criminal justice system as the focal point of the ongoing debate on the relationship between legal responsibility, free will, mental illness, and punishment. The insanity defense has substantially survived in spite of persistent philosophical and political criticism. Its history reflects a balance and tension between changes in attitudes toward developments in psychiatry and psychology and changes in attitudes toward criminal justice, incapacitation, and the desire to punish. Probably no other area of criminal law and procedure reflects a jurisprudence that is so driven by myths as that of the insanity defense. Yet only a handful of American jurisdictions have legislatively abolished it.

Insanity defense issues have concerned the courts and legislative bodies for hundreds (perhaps thousands) of years. As the various tests have developed— M’Naghten, irresistible impulse; Durham, the test proposed in the American Law Institute’s Model Penal Code (ALI-MPC); the federal Insanity Defense Reform Act, diminished capacity—and as efforts are made to limit the scope and use of the defense, either by use of a “guilty but mentally ill” verdict or by outright abolition, it is clear that the symbolic values of the insanity defense must be considered carefully at all times. No area of our legal system has engendered a more intense level of debate than the role of the insanity defense in the criminal justice process. On the one hand, this difficult subject is seen as a reflection of the fundamental moral principles of the criminal law, resting on beliefs about human rationality, deterrability (i.e., whether the punishment of a person whose profound mental illness leads him to commit what would otherwise be a criminal act would serve as a deterrent to others), and free will, and as a bulwark of the law’s moorings of condemnation for moral failure. On the other hand, it is castigated by a former attorney general of the United States as the major stumbling block in the restoration of “the effectiveness of Federal law enforcement” and as tilting the “balance between the forces of law and the forces of lawlessness.” Yet the percentage of insanity defenses pled is small (at the most 1%), the percentage of those successful is smaller (1/4 of 1%), and the percentage of those successful in contested cases is minuscule (1/10 of 1/4 of 1%).

Notwithstanding the defense’s relative numerical insignificance, it touches—philosophically, culturally, and psychologically—on our ultimate social values and beliefs; it is rooted in moral principles of excuse that are accepted in both ordinary human interaction and criminal law; and it continues to serve as a surrogate for resolution of the most profound issues in criminal justice. Although the defense has been significantly narrowed in many jurisdictions in the past 25 years—a condition intensified by the verdict in the John Hinckley case (which involved the attempted assassination of President Ronald Reagan) as well as several other unpopular or “wrong” jury verdicts in cases involving sensationalized crimes or public figure victims—reports of its demise are, to a great extent, exaggerated and, in spite of public outrage, the doctrine has remained alive in most jurisdictions.

The insanity defense has been a major component of the Anglo-American common law for more than 700 years. Rooted in Talmudic, Greek, and Roman history, its forerunners actually can be traced back to more than 3,000 years. The sixth-century Code of Justinian explicitly recognized that the insane were not responsible for their acts and also articulated the early roots of the temporary insanity and diminished capacity doctrines. By the ninth century, the “Dooms of Alfred” (a code of laws compiled by Alfred the Great) acknowledged that an impaired individual— who could not acknowledge or confess his offenses— was absolved from personally making restitution. In pre-Norman England, the law similarly shifted reparations responsibility in the event that a “man fall out of his senses or wits,… and kill someone.”

The defense’s “modern” roots can be traced at least as far back as 1505, the first recorded jury verdict of insanity, but it is clear that even prior to that case, juries considered “acquittal to be the appropriate result” in certain insanity defense cases. Furthermore, William Lambard’s late-16th-century text on criminal responsibility (The Eirenarcha) suggested that the insanity defense was already well settled in England, and Sir Edward Coke’s 1628 treatise, Institutes of the Laws of England, gave the law the familiar maxim that the ” madman is only punished by his madness.”

Early Developments

In the early 18th century, English judges began the process of attempting to define for juries that condition of the mind which would excuse, as a matter of law, otherwise criminal behavior. In Rex v. Arnold (1724), the first of the historically significant insanity defense trials, Judge Tracy charged the jury in the following manner:

That is the question, whether this man hath the use of his reason and sense? If he … could not distinguish between good and evil, and did not know what he did…he could not be guilty of any offence against any law whatsoever….On the other side…it is not every kind of frantic humour or something unaccountable in a man’s actions, that points him out to be such a madman as is to be exempted from punishment: it must be a man that is totally deprived of his understanding and memory, and doth not know what he is doing, no more than an infant, than a brute, or a wild beast, such a one is never the object of punishment.

The law of criminal responsibility evolved further in 1800, in the case of James Hadfield, which envisioned insanity in the following manner:

That a man could know right from wrong, could understand the nature of the act he was about to commit, could manifest a clear design and foresight and cunning in planning and executing it, but if his mental condition produced or was the cause of a criminal act he should not be held legally responsible for it.

This trend toward a more liberal defense continued in the case of Regina v. Oxford (1840), which concerned the attempted assassination of Queen Victoria, in which the jury charge combined portions of what would later be known as the “irresistible impulse” test and the “product” test.

M’Naghten Case

The most significant case in the history of the insanity defense in England (and perhaps in all common-law jurisdictions) arose out of the shooting by Daniel M’Naghten of Edward Drummond, the secretary of the man he mistook for his intended victim, Prime Minister Robert Peel (as with all the other cases already discussed, the victim was a major political figure). Enraged by the jury’s insanity verdict, Queen Victoria questioned why the law was of no avail, since “everybody is morally convinced that [the] malefactor…[was] perfectly conscious and aware of what he did,” and demanded that the legislature “lay down the rule” so as to protect the public “from the wrath of madmen who they feared could now kill with impunity.”

In response to the Queen’s demand, the House of Lords asked the Supreme Court of Judicature to answer five questions regarding the insanity law; the judges’ answers to two of these five became the M’Naghten test (1843):

The jurors ought to be told in all cases that every man is presumed to be sane, and to possess a sufficient degree of reason to be responsible for his crimes, until the contrary be proved to their satisfaction; and that to establish a defence on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party 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 he was doing what was wrong.

There are three main features of this formulation: First, it is predicated on proof that the defendant was suffering from a “defect of reason, from disease of the mind.” From the time of M’Naghten until today some finding of “mental disease or defect” has been a necessary predicate for the insanity defense. Second, once such a “disease” is shown, the inquiry focuses on what the defendant was able to “know.” That is, the interest of the law under this test is in the ability of the defendant to “know” certain things. It is for this reason that the inquiry is sometimes referred to as a “cognitive” formula. Third, the M’Naghten test focuses on two things the defendant must be able to “know” to be guilty of a crime. One is “the nature and quality” of the act that was committed. The other is that the act “was wrong.” In both instances, the question is whether the defendant was “capable” of knowing these things, that is, whether the mental illness had robbed the defendant of the capacity to know what “normal” people are able to know about their behavior. The idea, in sum, is that people who are unable to know the nature of their conduct or who are unable to know that their conduct is wrong are not proper subjects for criminal punishment. In commonsense terms, such persons should not be regarded as morally responsible for their behavior.

This test has been severely criticized as rigid and inflexible, based on outmoded views of the human psyche, of little relation to the truths of mental life, reflecting antiquated and outworn medical and ethical concepts. Furthermore, the use of language such as “know” and “wrong” was criticized as “ambiguous, obscure, unintelligible, and too narrow.” Donald Hermann and a colleague have argued, by way of example, that the cognitive aspect of one’s personality cannot be seen as the sole determinant of one’s subsequent behavior (and the basis of one’s ultimate criminal guilt) because the psyche is an integrated entity.

Critics also maintain that the narrow scope of the expert testimony required by the M’Naghten test deprives the jury of a complete picture of the psychological profile of the defendant as the test ignores issues of affect and control.

Nevertheless, American courts readily adopted the M’Naghten formulation and codified it as the standard test, “with little modification,” in virtually all jurisdictions until the middle of the 20th century.

Irresistible Impulse

In a partial response to criticisms of the M’Naghten test, several courts developed an alternative test that later became known as the “irresistible impulse” test, adapted from a test first formulated in 1883 by Lord Stephen:

If it is not, it ought to be the law of England that no act is a crime if the person who does it is at the time…prevented either by defective mental power or by any disease affecting his mind from controlling his own conduct, unless the absence of the power of control has been produced by his own default.

This rule allowed for the acquittal of a defendant if his mental disorder caused him to experience an “irresistible and uncontrollable impulse to commit the offense, even if he remained able to understand the nature of the offense and its wrongfulness.” It was based, in the words of Abraham Goldstein, one of the leading legal scholars on the history of the insanity defense, on four assumptions:

First, that there are mental diseases which impair volition or self-control, even while cognition remains relatively unimpaired; second, that the use of M’Naghten alone results in findings that persons suffering from such diseases are not insane; third, that the law should make the insanity defense available to persons who are unable to control their actions, just as it does to those who fit M’Naghten; fourth, no matter how broadly M’Naghten is construed, there will remain areas of serious disorder which it will not reach. (p. 67)

At its high-water mark, this test had been adopted in 18 jurisdictions, but today, far fewer states follow its teachings.

The Product Test

Charles Doe, a mid-19th-century New Hampshire State Supreme Court judge, first crafted what became known as the product test: “If the [crime] was the offspring or product of mental disease in the defendant, he was not guilty by reason of insanity” (State v. Pike, 1870). This test first entered the legal public’s consciousness in 1954, when it was adopted by the District of Columbia in Durham v. United States, rejecting both the M’Naghten and the irresistible impulse tests as based on “an entirely obsolete and misleading conception of the nature of insanity,” one that ignored the reality that “the science of psychiatry now recognizes that a man is an integrated personality and that reason, which is only one element in that personality, is not the sole determinant of his conduct” (p. 871) and that a far broader test would be appropriate.

Durham held that an accused would not be criminally responsible if his “unlawful act was the product of mental disease or mental defect” (pp. 874—875). This test would provide for the broadest range of psychiatric expert testimony, “unbound by narrow or psychologically inapposite legal questions” (Weiner, 1985, p. 710). The case was the first modern, major break from the M’Naghten approach and created a feeling of intellectual and legal ferment. It was adopted, however, in fewer than a handful of jurisdictions and became the topic of fairly rigorous criticism, that it allegedly failed to provide helpful guidelines to the jury and that it was—at its core—a “nonrule,” providing the jury with no standard by which to judge the evidence; that it misidentified the moral issue of responsibility with the scientific issues of diagnosis and causation; and that it was too heavily dependent on expertise, leading to the usurpation of jury decision making by psychiatrists. Within a few years after the Durham decision, the court began to modify and— ultimately—dismantle it, culminating in its decision in United States v. Brawner (1973), the most important of the many federal cases that had rejected M’Naghten and adopted instead the ALI-MPC test.

American Law Institute’s Model Penal Code Test

In an effort to avoid the major criticisms of M’Naghten, the irresistible impulse test, and Durham, the ALI couched the substantive insanity defense standard of its MPC in language that focused on volitional issues as well as cognitive ones. According to the ALI-MPC standard, a defendant is not responsible for his criminal conduct if, as a result of mental disease or defect, he “lacks substantial capacity either to appreciate the criminality of his conduct or to conform his conduct to the requirements of law” (§ 4.01(1)). Under this formulation, the term mental disease or defect specifically excluded “an abnormality manifested only by repeated criminal or otherwise antisocial conduct”(§ 4.01(2)).

Although the ALI-MPC test was rooted in the M’Naghten standard, there were several significant differences. First, its use of the word substantial was meant to respond to case law developments that had required a showing of total impairment for exculpation from criminal responsibility. Second, the substitution of the word appreciate for the word know showed that a sane offender must be emotionally as well as intellectually aware of the significance of his or her conduct and that mere intellectual awareness that the conduct is wrongful when divorced from an appreciation or understanding of the moral or legal import of behavior can have little significance. Third, by using a broader language of mental impairment than had M’Naghten, the test captured both the cognitive and affective aspects of impaired mental understanding. Fourth, its substitution in the final proposed official draft of the word wrongfulness for criminality reflected the position that the insanity defense dealt with an impaired moral sense rather than an impaired sense of legal wrong.

Although there were some immediate criticisms of the ALI-MPC test, principally due to the attempt to bar “psychopaths” or “sociopaths” from successfully using the defense, the test was generally applauded as encouraging adjudication based on reality and the practical experience of psychiatrists by recognizing that both the volitional and the cognitive processes of an individual may be impaired. The test was subsequently adopted by more than half of the states and, in some form, by all but one of the federal circuits. Perhaps most significant, the District of Columbia Court of Appeals, in overruling its “product” test of Durham v. United States in United States v. Brawner, adopted the ALI-MPC test.

Insanity Defense Reform Act

Slightly more than a decade after Brawner, in the wake of John Hinckley’s failed attempt to assassinate U.S. President Ronald Reagan, Congress enacted the federal Insanity Defense Reform Act. This law had the effect of returning the insanity defense in federal jurisdictions to status quo ante 1843: the year of M’Naghten. The bill changed the federal law in several material ways:

  1. It shifted the burden of proof to defendants, by a quantum of clear and convincing evidence.
  2. It articulated, for the first time, a substantive insanity test, adopting a more restrictive version of M’Naghten, thus discarding the ALI-MPC test previously in place in all federal circuits.
  3. It established strict procedures for the hospitalization and release of defendants found not guilty by reason of insanity.
  4. It severely limited the scope of expert testimony in insanity cases.

Diminished Capacity

One of the most difficult concepts in substantive insanity defense formulation is that of diminished capacity, a doctrine that holds that evidence of reduced mental capacity tending to show the absence of any mental state essential to the alleged crime should be accepted by the trial court, whether or not an insanity plea was entered (People v. Wells, 1949, pp. 63-70). There seems little question that the diminished capacity doctrine was developed to be used in murder cases to mitigate the harshness of a potential death penalty by raising the question of whether the defendant could sufficiently appreciate the nature of his or her act so as to sustain a first-degree murder conviction. In addition, use of the doctrine has been justified as a means to ameliorate defects in a jurisdiction’s substantive insanity defense test criteria and as a means of permitting juries to make more accurate individualized culpability judgments.

This doctrine, however, has been routinely criticized for its difficulty and arbitrariness in application, leading to uneven and inequitable outcomes, and while it had been endorsed in some form in nearly 25 jurisdictions, it has failed to capture major support and has even lost the support it previously enjoyed.

“Guilty but Mentally Ill”

Perhaps the most significant development in substantive insanity defense formulations in the past 20 years has been the adoption in more than a dozen jurisdictions of the hybrid “guilty but mentally ill” (GBMI) verdict. It received its initial impetus in Michigan, as a reflection of legislative dissatisfaction with and public outcry over a state Supreme Court decision that had prohibited automatic commitment of insanity acquittees. There, legislation was enacted that provided for a GBMI verdict—as an alternative to the not guilty by reason of insanity (NGRI) verdict—if the following were found by the trier of fact beyond a reasonable doubt:

  1. That the defendant is guilty of an offense
  2. That the defendant was mentally ill at the time of the commission of the offense
  3. That the defendant was not legally insane at the time of the commission of the offense

The rationale for the passage of the GBMI legislation was that the implementation of such a verdict would decrease the number of persons acquitted by reason of insanity and ensure treatment of those who were GBMI within a correctional setting. It was conceived that once a defendant were to be found GBMI, he or she would be evaluated on entry to the correctional system and provided appropriate mental health services either on an inpatient basis as part of a definite prison term or, in specific cases, as a parolee or as an element of probation. This model was followed—in large part—in most of the other states that have adopted the GBMI test.

Most academic analyses have been far more critical, rejecting it as conceptually flawed and procedurally problematic and as not only superfluous but also dangerous. By way of example, in practice, the GBMI defendant is not ensured treatment beyond that available to other offenders. Thus, Christopher Slobogin (one of the leading current scholars in this area of the law) suggests, it is “not only misleading but dangerous to characterize the [GBMI] verdict either as a humane advance in the treatment of mentally ill offenders or as a more effective way of identifying offenders in need of treatment.” The GBMI verdict, he concludes, is “a verdict in name only” (p. 515).

Insanity Defense Myths

The empirical research has revealed that at least half a dozen myths about the insanity defense had arisen and have been regularly perpetuated but were all disproven by the facts. The research showed that the insanity defense opens only a small window of nonculpability, that defendants found that NGRI does not “beat the rap,” and, perhaps most important, that the tenacity of these misbeliefs in the face of contrary data is profound.

Myth 1: The insanity defense is overused. All empirical analyses have been consistent: the public, legal profession, and, specifically, legislators dramatically and grossly overestimate both the frequency and the success rate of the insanity plea. This error undoubtedly is abetted by media distortions in presenting information on persons with mental illness charged with crimes.

Myth 2: The use of the insanity defense is limited to murder cases. In one jurisdiction where the data have been closely studied, slightly less than one third of the successful insanity pleas entered over an 8-year period were reached in cases involving a victim’s death. Furthermore, individuals who plead insanity in murder cases are no more successful in being found NGRI than persons charged with other crimes.

Myth 3: There is no risk to the defendant who pleads insanity. Defendants who asserted an insanity defense at trial and who were ultimately found guilty of their charges served significantly longer sentences than defendants tried on similar charges but did not assert the insanity defense. The same ratio is found when exclusively homicide cases are considered.

Myth 4: NGRI acquittees are quickly released from custody. Of all the individuals found NGRI over an 8-year period in one jurisdiction, only 15% had been released from all restraints, 35% remained in institutional custody, and 47% were under partial court restraint following conditional release.

Myth 5: NGRI acquittees spend much less time in custody than do defendants convicted of the same offenses. Contrary to this myth, NGRI acquittees actually spend almost double the amount of time that defendants convicted of similar charges spend in prison settings and often face a lifetime of postrelease judicial oversight.

Myth 6: Criminal defendants who plead insanity are usually faking. This is perhaps the oldest of the insanity defense myths and is one that has bedeviled American jurisprudence since the mid-19th century. Of 141 individuals found NGRI in one jurisdiction over an 8-year period, there was no dispute that 115 were schizophrenic (including 38 of the 46 cases involving a victim’s death), and in only three cases was the diagnostician unable to specify the nature of the patient’s mental illness.

Abolition and Limitation Proposals

In the past two decades, state legislatures in Idaho, Montana, Kansas, and Utah have abolished the insanity defense, and in those jurisdictions, state supreme courts have subsequently held that abolition of the defense did not violate due process. Arizona stopped barely short of abolishing the insanity defense by creating a “guilty except insane” verdict that eliminates the “nature and quality of the act” prong from the M’Naghten test. In one instance (Nevada), such abolition was struck down as unconstitutional in Finger v. State of Nevada (2001), with the majority of the sharply divided court finding that legal insanity was a “fundamental principle” entitled to due process protections. The court reasoned as follows:

Mens rea is a fundamental aspect of criminal law. Thus it follows that the concept of legal insanity, that a person is not culpable for a criminal act because he or she cannot form the necessary mens rea, is also a fundamental principle. (p. 80)

The U.S. Supreme Court recently addressed questions raised in Arizona’s insanity defense:

Whether due process prohibits Arizona’s use of an insanity test stated solely in terms of the capacity to tell whether an act charged as a crime was right or wrong; and whether Arizona violates due process in restricting consideration of defense evidence of mental illness and incapacity to its bearing on a claim of insanity, thus eliminating its significance directly on the issue of the mental element of the crime charged (known in legal shorthand as the mens rea, or guilty mind). (Clark v. Arizona, 2006)

In both instances, the Court held there was no violation of due process.


  1. American Law Institute. (1962). Model penal code. Philadelphia: Author.
  2. Clark v. Arizona, 548 U.S._126 S. Ct. 2709 (2006).
  3. Durham v. United States, 214 F.2d 862 (D.C. Cir. 1954) (overruled in United States v. Brawner, 471 F. 2d 969 (D.C. Cir. 1972).
  4. Finger v. State of Nevada, 27 P.3d 66 (Nev. 2001).
  5. Goldstein, A. S. (1967). The insanity defense. New Haven, CT: Yale University Press.
  6. Low, P. W., Jeffries, J. C., & Bonnie, R. J. (2000). The trial of John W. Hinckley, Jr.: A case study in the insanity defense (2nd ed.). New York: Foundation Press.
  7. Morris, N. (1984). Madness and the criminal law. Chicago: University of Chicago Press.
  8. People v. Wells, 202 P. 2d 63 (Cal. 1949).
  9. Perlin, M. L. (1989/1990). Unpacking the myths: The symbolism mythology of insanity defense jurisprudence. Case Western Reserve Law Review, 40, 599-731.
  10. Perlin, M. L. (1994). The jurisprudence of the insanity defense. Durham, NC: Carolina Academic Press.
  11. Perlin, M. L. (1997). “The borderline which separated you from me”: The insanity defense, the authoritarian spirit, the fear of faking, and the culture of punishment. Iowa Law Review, 82, 1375-1426.
  12. Perlin, M. L. (2002). Mental disability law: Civil and criminal (2nd ed., Vol. 4). Newark, NJ: Matthew Bender.
  13. Regina v. Oxford, 173 Eng. Rep. 941 (1840).
  14. Rex v. Arnold, 16 How. St. Tr. 695 (1724).
  15. Slobogin, C. (1985). The guilty but mentally ill verdict: An idea whose time should not have come. George Washington Law Review, 53, 494-527.
  16. State v. Pike, 49 N.H. 399, 442 (1870).
  17. United States v. Brawner, 471 F.2d 969 (1973).
  18. Weiner, B. A. (1985). Mental disability and criminal law. In S. J. Brakel, J. Parry, & B. A. Weiner (Eds.), The mentally disabled and the law (3rd ed., pp. 693-773). Chicago: American Bar Association.

See also: