Criminal Responsibility

Mental health professionals are frequently asked to evaluate criminal defendants to assist the courts in determining whether those individuals may have been legally insane (i.e., not criminally responsible) at the time of their crimes. This section discusses the legal concept of criminal responsibility and criteria for insanity, as well as the challenges that forensic experts face in conducting these evaluations.

In contemporary Western society, prohibited behaviors are typically codified in the criminal law, and most citizens are held to be responsible to obey these laws. Individuals who violate the law may be prosecuted and, if convicted, punished for their behavior. Such individuals are said to be “criminally responsible,” a label that reflects the moral and legal judgment that the person had neither a justification nor an excuse for his or her behavior, should have known better, and must endure the punishment as a corrective mechanism intended to discourage the recurrence of such behavior.

As the last sentence indicates, behavior that on its face appears to be criminal may in some circumstances not warrant the legal and moral conclusion that the actor is “guilty” or “criminally responsible.” For example, under most circumstances, it is unlawful to take the life of another person, yet doing so in “self-defense” (the victim was threatening the life of the actor) may constitute a justification that precludes a finding of guilt. Similarly, taking money from another person (robbery), when performed under duress (a third-party threatens to kill the actor’s child unless the money is taken), may be seen as justified because of the greater harm (death of the child) that was avoided by robbing the victim.

There are other individuals, or classes of individuals, who may be exempt or excused from judgments of “criminally responsible,” not because of extraordinary or justifying circumstances, but because of individual characteristics or features that render them, in society’s eyes, incapable of making the appropriate moral and legal judgments required to behave appropriately and (perhaps also) incapable from benefiting from punishment as a corrective measure. For our purposes, two such classes of individuals will be mentioned, both of which have been recognized in Western cultures, literally for centuries, as inappropriate targets for judgments of moral and legal culpability.

The first class of individuals is children, who, because of youthful age, lack of life experience, and mental or emotional immaturity, are considered not accountable as moral actors in the way that adults are held accountable for their behavior. Although exceptional cases may be found, the law has generally considered it an unrebuttable assumption that children at the age of 7 years and younger may not be held to adult standards of criminal responsibility, whereas there is a rebuttable assumption that children between the ages of 7 and 14 years are not moral agents to be held to adult standards of criminal responsibility.

The second group or class of individuals, and the one of primary focus here, comprises individuals with significant mental disorders whose symptomatology contributes to their “criminal behavior” in specific ways that society deems excuses them from moral culpability (criminal responsibility). In more common legal parlance, these individuals are considered “legally insane” or “not guilty by reason of insanity.”

The insanity defense is generally unpopular in public opinion because of misperceptions about how it is used. Survey studies reveal public beliefs that the insanity defense is both frequently used and often successful. Such beliefs are likely distortions that result from the high degree of publicity surrounding notorious cases such as those of Patty Hearst, David Berkowitz (“Son of Sam”), Jeffrey Dahmer (none of whom was judged legally insane), and John Hinckley (who was found insane). However, neither belief is correct. Research shows that the insanity defense is asserted in less than half of 1% of criminal cases and more often than not it fails. Furthermore, when the insanity defense is successful, it is not the result of clever lawyers pulling the wool over the eyes of naive jurors. Most successful insanity defenses are not seriously challenged by the prosecution; more often it is clear to all parties that the defendant was insane at the time of the offense (according to criteria discussed below) and should not be held legally and morally responsible for his or her actions. Thus, many successful insanity defenses result, in effect, from plea agreements.

The following sections discuss (a) the criteria for legal insanity, (b) the legal calculus for determining when a defendant meets these criteria, and (c) the methods used by mental health professionals to gather evidence and formulate opinions about the mental state of a defendant at the time of an offense.

Criteria for Legal Insanity

Historians of Western law point to the influence of Henry de Bracton, whose writings on English law in the 13th century introduced notions of mental capacity and intent into deliberations about guilt and moral culpability. Early language referenced notions such as “infancy” or reasoning capacity not far removed from that of a “wild beast” as potentially exculpating mental states. The most influential English case was the M’Naghten case (1843), which established as the test for insanity that the accused was laboring under such 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.

The emphasis in M’Naghten on impairment in the ability to “reason” or to “know” clearly made judgments about a defendant’s cognitive capacities central to the legal determination of insanity. This emphasis has survived in modern formulations that reference a defendant’s “ability to appreciate the wrongfulness of his conduct.” However, over the course of time modern psychiatry has influenced the law to consider volitional as well as cognitive impairments as potential bases for legal insanity. Thus, formulations in some jurisdictions refer to insanity resulting from criminal behavior that was due to an “irresistible impulse” or impaired capacity “to conform one’s conduct to the requirements of the law” irrespective of the presence of cognitive impairment.

Finally, some formulations specify quantitative, albeit imprecise, levels of impairment for the insanity test to be met. Thus, it may be that not merely “impaired capacity” but “substantially impaired capacity” must be established to excuse the defendant from being criminally responsible for his or her behavior.

The Calculus of Legal Tests for Insanity

In jurisdictions that allow the insanity defense, judges and juries must apply the relevant legal test. The structure of all such tests requires three findings by the trier of fact. First, there is a predicate mental condition from which the defendant must have been suffering at the time of the offense. In the M’Naghten formulation above, the requisite condition is a “disease of the mind.” More modern formulations reference the presence of “mental disease or defect” or similar language. It is important to note that in virtually all formulations, the legal definition for the predicate mental condition is neither highly precise nor tied explicitly to clinically recognized categories of mental illness or other mental impairments. (It is the case, however, that in many jurisdictions certain clinical conditions are explicitly barred by law as a basis for an insanity defense, most commonly (a) states of intoxication due to voluntarily consumed drugs or alcohol and (b) “disorders” defined almost exclusively on the basis of a history of antisocial behavior, such as antisocial personality disorder or sociopathy.)

The second component in the legal test is that the criminal act was affected by (loosely, “caused” by) the predicate mental condition. In other words, merely having a “disease of the mind” or a “mental disease or defect” alone is not sufficient to excuse the defendant from being criminally responsible. Mentally ill people may commit crimes for all the noncrazy reasons that other people do; they may steal because they are greedy, fight because they are angry, or drive poorly because they are intoxicated. To sustain an insanity defense, the mental impairment must have contributed to the occurrence of the criminal behavior.

Finally, the link between the predicate mental impairment and the criminal behavior must specifically be of the type prescribed in the legal test. As noted above, depending on the legal jurisdiction, the test for insanity may reference either cognitive or volitional impairments. Imagine a scenario in which a person with a well-established diagnosis of generalized anxiety disorder (“mental disease”) suddenly felt extremely anxious in a situation where the only means of escaping was to take another person’s car and drive away. He is not confused as to the ownership of the car (i.e., he “knows” that it belongs to another person), does not think that he has that person’s permission to take the car (i.e., has no illness-related delusion that he has the authority or approval to take the car), and is aware of, and maybe even consciously anxious about, the possibility that he could be arrested for taking the car. Under a purely cognitive insanity formulation that focuses on “knowing” or “appreciating” the wrongfulness of his behavior, there is little to suggest that the actor’s illness (acute anxiety symptoms)—although it motivated his decision to take the car—impaired his cognitive abilities in the way prescribed by the legal test. Alternatively, under a volitional formulation that referenced impaired control of impulses or capacity to conform conduct, the sudden strong urge to flee, arguably animated by his anxiety disorder, might support a finding of insanity.

Clinical Assessment of Criminal Responsibility

When the defense decides to pursue a defense of insanity, mental health professionals, commonly psychiatrists or psychologists, are hired by the prosecution and defense and/or appointed by the court to evaluate the defendant’s mental condition and to provide reports and/or testimony as to the defendant’s criminal responsibility. This is one of the most challenging types of evaluation for mental health professionals in their roles as forensic examiners because it is different, in so many ways, from the ordinary evaluations that they conduct in clinical (nonlegal) settings.

Clinical diagnostic assessments are imperfect even under optimal conditions—that is, when the clinician is working with a voluntary, candid, and willing client and the focus is on present mental functioning and treatment planning. Such conditions are almost never present when evaluations for criminal responsibility are being conducted. Because the insanity evaluation focuses narrowly on a specific point in time in the past, this inevitably diminishes the utility of commonly used clinical measures, such as psychological tests or other diagnostic procedures. Instead, insanity evaluations rely to a large extent on reviews of investigative evidence collected by the police, interviews with defendants, and information collected from third parties who may have knowledge relevant to the defendant’s behavior and functioning at or near the time of the offense. Thus, investigative reporting, rather than traditional clinical assessment, is perhaps a better conceptual model for criminal responsibility evaluations.

The challenges faced in conducting criminal responsibility evaluations include the following:

The Evaluation Is Retrospective

It is not uncommon for insanity evaluations to be conducted weeks or months after the defendant’s arrest. Furthermore, the time window between the crime and the clinical evaluation may be extended considerably if the arrest is made only after a prolonged investigation. Much can happen during this interval to distort the reconstructed picture of the defendant’s prior mental state, including the following:

  1. The defendant has a mental illness that has deteriorated over time; the clinician interviews the defendant in this more disturbed state and may attribute more psychopathology at the time of the offense than was actually present.
  2. The defendant has a mental illness that improves either spontaneously, due to the cyclical nature of the disorder, or to treatment received (e.g., in jail); the clinician interviews the defendant in this less disturbed state and may attribute less psychopathology at the time of the offense than was actually present.
  3. Although not symptomatic at the time of the offense, the defendant may have developed symptoms subsequent to the offense (e.g., a reaction to the nature of the crime itself, to events that occurred at arrest or in the jail, or in anticipation of serious consequences); the clinician may attribute some of this symptomatology as being present at the time of the offense.
  4. Information obtained by interviewing the defendant or third-party sources (e.g., witnesses, family members), even if offered “honestly,” may be less accurate due to deterioration of memory over time.

Concerns about Information Validity

Most people who provide information to the forensic examiner have a personal or professional interest in the opinions and findings that the examiner will reach. Thus, concerns about the validity of information are greater in insanity evaluations (and other forensic assessments) than with evaluations conducted for standard clinical and therapeutic purposes.

A defendant may view a successful insanity defense as his or her only hope for avoiding a lengthy prison sentence and thus be motivated to exaggerate or fabricate symptoms of mental disorder in describing behavior and motivations at the time of the offense. Family members sympathetic to the defendant’s plight may distort information in ways that they believe are helpful to the case. The attorney(s) may be selective in the investigative information made available to the clinician, withholding that which they believe might lead the clinician to an unfavorable opinion. Evidence may be gathered and provided by the police in ways that provide a misleading picture of the defendant’s prior mental functioning. For example, a defendant who is mentally confused and verbally incoherent may be cajoled into signing a “confession,” drafted in perfectly organized and sensible language by an arresting officer, that belies the extent of psychopathology present at the time of arrest.

Translating Clinical Findings for Legal Consumers

Based on information gathered from the defendant, the police, and available third-party sources, the forensic examiner attempts to reconstruct an account of the defendant’s mental state at the time of the offense that considers whether, and the extent to which, symptoms of mental disorder may have contributed to the alleged crime. However, as noted above there is no direct translation of clinically recognized mental disorders, which can vary from relatively benign (e.g., nicotine use disorder) to severely incapacitating conditions (e.g., schizophrenia, manic disorder), into legal terms such as disease of the mind or mental disease or defect.

Similarly, various formulations of the legal criteria for insanity require qualitative or quantitative determinations of either the nature of the functional legal impairment (e.g., ability to “know” or to “appreciate” wrongfulness of conduct) or the extent of impairment (e.g., categorically “did not know” vs. “lacked substantial capacity to know”) for which there is no clinical or scientific technology.

That there is no scientific basis for translating clinical findings into specific legal conclusions poses a challenge to forensic examiners who are often pressured by the attorneys, if not also the courts, to give conclusory opinions under the mantra of “reasonable medical (or scientific) certainty.” Mental health professionals have no “capacimeters” for determining whether the specific nature or extent of impairment in a given case is sufficient to excuse the defendant from his or her moral obligation to obey the rules. These constructs are legal terms of art that, in any individual case, have meaning only as expressed in the eventual social and moral judgment of the judge or jury when the verdict is reached. The status of an individual being legally insane (i.e., “not criminally responsible”) is a social construction that has no meaning prior to, or independent of, the jury’s pronouncement.

This is not to say that clinical evaluations of criminal responsibility cannot be helpful to legal decision makers. Rather, the challenge for forensic examiners is to collect information relevant to a defendant’s legal functioning and to describe it to the triers of fact in ways that facilitates their ultimate judgments, but without offering moral judgments of their own under the guise of scientific expertise.

To illustrate with an example, one defendant who had a long and well-documented history of mental disorder experienced a recurrence of symptoms that included the delusional belief that he had been appointed to the position of deputy director of the FBI (in reality, the individual had worked in a factory for 20 years). On the basis of this belief, and the further notion that he was urgently needed in Washington, D.C., on matters of national security, he boarded a Greyhound bus and, without license or permission, drove it away from the bus depot. He was arrested and charged with unlawfully driving away a motor vehicle.

In this case, a forensic examiner might report that at the time of the offense the defendant experienced symptoms (i.e., delusions—strongly held but erroneous beliefs) of a well-recognized mental disorder (schizophrenia) with which he had been diagnosed for a number of years. Although the objective evidence is that the defendant is a factory worker, the manifestation of his illness at the time of the offense included a set of beliefs that distorted his perceptions of reality with respect to his occupation (i.e., objectively he was not an FBI official) and his rights and duties associated with his occupation (i.e., objectively he did not have the authority to commandeer a public transportation vehicle in the interest of national security). The nature of these distortions impaired his ability to judge correctly with respect to the action of taking control over the bus.

This formulation lacks a conclusive opinion as to whether or not the defendant’s symptoms at the time of the offense satisfy the required predicate condition (“he suffered from a mental disease”). Also absent is any conclusory opinion that the symptoms categorically did, or did not, relate to the criminal act in the prescribed way (i.e., “he did not know that what he was doing was wrong”). However, the formulation does provide a plausible accounting of the relationship between the defendant’s symptoms and the criminal behavior, but leaves it to the jury to “connect the dots,” so to speak, in a fashion congruent with their collective social and moral intuitions as to whether or not an individual so disturbed should be held criminally responsible.


  1. De Pauw, K. W., & Szulecka, T. K. (1988). Dangerous delusions: Violence and misidentification syndromes. British Journal of Psychiatry, 152, 91-96.
  2. Dreher, R. H. (1967). Origin, development and present status of insanity as a defense to criminal responsibility in the common law. Journal of the History of the Behavioral Sciences, 3, 47-57.
  3. Goldstein, A. M., Morse, S. J., & Shapiro, D. L. (2003). Evaluation of criminal responsibility. In A. M. Goldstein (Ed.), Handbook of psychology: Vol. 11. Forensic psychology (pp. 381-1-06). Hoboken, NJ: Wiley.
  4. Morse, S. J. (1994). Causation, compulsion, and involuntariness. Bulletin of the American Academy of Psychiatry and the Law, 22, 159-180.
  5. Morse, S. J. (1994). Culpability and control. University of Pennsylvania Law Review, 142, 1587-1655.
  6. Poythress, N. G. (2004). “Reasonable medical certainty”: Can we meet Daubert standards in insanity cases? [Editorial]. Journal of the American Academy of Psychiatry and the Law, 32, 228-230.
  7. Slobogin, C. (2007). Proving the unprovable: The role of law, science, and speculation in adjudicating culpability and dangerousness. New York: Oxford University Press.