The determination of competence to stand trial has a rather inglorious history. Under English common law in the medieval period, the proceedings of a trial were terminated by the failure of a defendant to plead to the charges. Therefore, officials had to resort to means of convincing the defendant to enter a plea. The procedure commonly used was referred to as peine forte et dure; it consisted of placing increasingly heavy rocks on the body of the defendant until either he voiced a plea or the necessity for the trial disappeared. Although many of those who refused to plead were undoubtedly criminals seeking to avoid capital punishment, some of those who succumbed to the weight of the court’s persuasion were probably mentally ill and intellectually disabled individuals who did not have sufficient capacity to understand what was required of them.

As the law worked its way into more enlightened times, the rocks were dispensed with in recognition that there might be some categories of individuals who were incapable of pleading and who, in addition, might not be able to participate adequately in the proceedings. These individuals were arraigned before a 12-person jury, and evidence of their mental state was presented. If the jury determined that the defendant was not able to plead, the defendant was sent to prison, to remain there until he became competent again. For many, that meant a lifetime of incarceration. Those who were thought to be malingering were tried despite their failure to plead.

The grounds on which the jury was to make its determination of competence were at first not made clear. Beginning in the late 1700s, a series of cases led, by virtue of somewhat more explicit judicial decisions, to a consensus on the legal standard to be applied: the defendant needed to be possessed of sufficient reason to understand the charges against her and to participate in the trial in her own defense. The heavy cognitive component of this standard, insofar as it stressed understanding to the exclusion of impairments of rationality or affect, was characteristic of the time and is also reflected in the nearly contemporaneous M’Naghten rule (see Sec. II-B-2-a) for criminal responsibility determinations.

Modern standards for competence to stand trial differ little, in most jurisdictions, from the nineteenth-century common-law rule. Each state determines the standards that govern most criminal trials in the state, except those under federal jurisdiction. In those cases, the standard enunciated in the federal case of Dusky v. United States applies: “The test must be whether he has sufficient present ability to consult with his lawyer with a reasonable degree of rational understanding, and whether he has a rational as well as a factual understanding of the proceedings against him.”

A tripartite definition of competence applies in most jurisdictions: the defendant must have an understanding of the nature of the charges against her and of the nature and purpose of court proceedings, and she must be able to cooperate with an attorney in her own defense.

A long-standing area of uncertainty is the relationship between competence to stand trial and competence to make specific decisions that arise in the course of defending against a criminal charge. For example, is someone who has been found competent under Dusky-type criteria necessarily competent to decide whether to plead guilty in exchange for a reduced sentence, to waive the right to a jury trial, or to dismiss his attorney and defend himself? Most courts, eager to avoid multiple competence hearings for defendants, lump these capacities together and assume that someone who meets a Dusky standard can make whatever decisions are required. A small number of courts and some commentators, however, criticize this approach, arguing that the possession of the minimal level of capacity envisioned by Dusky does not imply the presence of abilities to make complex, defense-related decisions. In 1993, the US Supreme Court addressed this question in Godinez v. Moran, holding that further inquiry into the defendant’s capacities was unnecessary once she met the Dusky criteria. However, Justice Thomas’s majority opinion indicated that, when defendants were making decisions with regard to waiver of constitutional rights, the “trial court must satisfy itself that the waiver…is knowing and voluntary.” The court characterized this not as an assessment of competence but as an evaluation of whether the defendant “actually does understand the significance and consequences of a particular decision” and makes the decision in an uncoerced manner. Although not free of ambiguity, the Godinez decision opened the door to individualized assessment of defendants’ abilities when decisions about waivers of rights are involved, albeit not under the rubric of a competence evaluation. More recently, the 2008 decision of the US Supreme Court in Indiana v. Edwards held that states could impose a higher standard for competence to waive counsel and could require representation by legal counsel for an unwilling defendant when that person was so impaired as to be unable to participate meaningfully in trial proceedings.

The rationale that underlies competence to stand trial standards has also undergone considerable evolution since the days of peine forte et dure. No longer is the law strictly concerned with the ritualistic aspects of lodging a plea. Rather, the requirement that a defendant be competent for a trial to take place is seen as protecting the fundamental fairness of the proceedings, guaranteed by the US Constitution, and ensuring that all relevant information that the defendant can provide is made available to the court. It can be argued that some aspects of the trial as a ritual persist: to the extent that public faith in the criminal justice system requires a perception of inherent fairness in the process, the exclusion of those whose competence is impaired makes the entire system more legitimate. Thus, judges are constitutionally required to order competence evaluations whenever the defendant’s behavior or the representations of any party raise a bona fide doubt about the defendant’s competence to stand trial.

Unlike the English courts of several centuries ago, contemporary American courts do not rely on the unassisted lay judgment of a jury to decide a defendant’s competence to stand trial. Although a few jurisdictions still use juries, the defendant is usually first examined by a clinician appointed by the court, and the clinician’s report then provides an important basis for the jury’s decision. In most jurisdictions, the trial judge determines the defendant’s competence, again almost always with clinical assistance. Some states make use of panels of experts to assess competence.

Regardless of who the ultimate decision-maker is, empirical studies of courtroom procedure show that the opinion of the forensic evaluator is highly influential in the final determination. However, it is not always clear what standards are being used by the clinician in making his assessment.

a. Standards for assessment. Until the 1970s, little attention was given to how mental health professionals should operationalize the standards used in assessing defendants’ alleged incompetence. The impression of researchers who have conducted field studies of hearings on this issue is that clinicians, as well as attorneys and judges, frequently confuse the question of competence with the questions of the presence of psychosis, suitability for civil commitment, and responsibility for criminal acts, determinations that should be clearly distinguished from one another. Evaluators’ opinions have tended to be stated in a conclusory manner with little or no justification given for the opinion that the defendant is not capable of standing trial. Although evaluations have improved somewhat with the development of training courses for certified examiners and the increasing use of specialized forensic services to perform the assessments, quality remains variable and, in some places, problematic.

To focus clinicians’ attention to those areas of functioning that are most relevant to a determination of competence to stand trial, and to fill the void left by the absence of clear guidelines from the courts, a variety of instruments and checklists have been devised. The most carefully constructed of the early instruments is the work of McGarry et al. (see Suggested Readings), which was empirically validated by cross-reference to the ultimate judicial disposition of the defendants on whom it was tested. The McGarry scale attempts to operationalize the competence assessment questions by looking at 13 areas of functioning, including the defendant’s:

Although the McGarry scale comes with an elaborate scoring system for each of these factors, the average clinician outside of a research setting will probably find the use of the checklist to structure her own interview is sufficient to ensure thoroughness (see Sec. III-B-1). It appears that responses to several of the criteria depend on previous experience with the legal system (i.e., factors 1, 4, 5, 6, and 9). Clinicians should not regard poor showings on these factors as determinative of incapacity, unless the defendant demonstrates an inability to be educated about these issues.

In addition to ensuring a thorough and systematized evaluation, the use of a scale, whether the McGarry checklist or one of the other published works, enables the clinician to organize the report to the court in such a way that the basis for his decision about the patient’s competence or incompetence is immediately clear. This avoids the problem of offering blanket statements that judges and attorneys have to either reject or accept on their face. Unfortunately, there are not good data comparing the available instruments, which include the MacArthur Competence Assessment Tool-Criminal Adjudication (MacCAT-CA), the Georgia Court Competency Test, and the Interdisciplinary Fitness Interview, to establish which may have higher levels of reliability and validity. Clinicians may want to explore the use of several instruments to assess which best meets the needs of their setting.

b. Answering the ultimate question. Because competence to stand trial is, in the end, a legal determination, many experts believe that a clinician should not offer an opinion as to the ultimate legal question or “ultimate issue.” That is, they maintain that, although all the relevant data that a judge might need to decide if a defendant is competent (e.g., those factors in the McGarry scale) should be presented clearly, the evaluator should not offer an opinion as to whether she considers the defendant competent or incompetent. These experts assert that the question of whether a defendant should stand trial is not a clinical decision but one based on a balancing of values
concerning fairness to the defendant and the public’s interest in resolving the criminal charges. Clinical evaluators have no special expertise in resolving these questions and are thus likely to rely on their own sociopolitical biases in reaching a decision, an outcome that is unfair to the defendant. In addition, it is argued that the clinical professions lose credibility with the public when they attempt to usurp the role of society’s appointed fact finders in this manner.

Conversely, some forensic clinicians advocate that evaluators who work in court settings should in fact view themselves as an arm of the court and should attempt to be as useful to the court as they possibly can. If answering the ultimate question would be helpful for the court, then that is a function mental health professionals should perform. These forensic clinicians would probably argue that objectivity in assessment is illusory and that, to the extent that bias exists, openly recognizing it is more likely to mute its effect than ignoring it altogether.

Something can be said for both sides of the issue, but, in the end, the clinician is an expert in human behavior, not in law. Regardless of professional considerations, it is unfair to the defendant for his legal status to be decided by clinical, rather than by judicial, fiat. Because the presentation of the bare clinical facts, even the use of a scale such as McGarry’s, leaves the judge or jury with the task of making the conceptual leap from the clinical formulation to the legal desiderata (an extremely difficult chore that they prefer to avoid), attorneys and judges frequently press evaluators for a more direct response to the legal issues. Clinicians in this situation are faced with three options: (1) caving in to the pressures exerted by the judge, (2) stating to the judge that their expertise lies in assessment of the patient and does not extend to legal matters and therefore they would prefer not to answer the question, or (3) taking the intermediate position of addressing the language of the legal test (e.g., “Based on the data I have presented, I believe that the defendant has a rational as well as a factual understanding of the proceedings against him”) but without reaching the ultimate judgment of whether the defendant is competent to stand trial.

Several empirical studies have demonstrated that a large number, in some cases a majority, of the defendants ostensibly referred by the courts for competence examinations are in fact being referred for other reasons. The prevalence of this phenomenon may have decreased somewhat in recent years, but it is still not uncommon. It usually occurs when the prosecutor or the court is interested in finding an alternative to holding a defendant in jail until the time of the trial or releasing him on bail. When the competence assessment is done on an outpatient basis or in the courthouse, the judge may use it as a way of finding a reason for denying bail (e.g., as a consequence of a finding that the defendant is mentally ill and dangerous) or for committing the defendant to a state hospital or other psychiatric facility (e.g., as a consequence of a finding that the defendant needs further evaluation for competence).

Referrals to inpatient facilities for competence assessments are even more subject to abuse. Most states permit involuntary commitments of defendants for periods of up to several weeks, with renewals for periods of up to several months. Judges are most likely to use such alternatives when defendants have a history of treatment for mental illness or have been behaving oddly, even if there is no good evidence that they are incompetent to be tried. Commitment for evaluation allows judges to avoid criticism for returning likely recidivists to the streets on bail. It also serves a judge’s wish to sequester people whose behavior is so deviant, without actually violating the law, that they are a perpetual discomfort to society. These individuals can be arrested on a petty charge such as disorderly conduct and then, without a trial or a chance at obtaining bail, be committed by the judge to a hospital for a substantial period.

This kind of commitment by the judiciary obviously deprives defendants of their constitutional rights to bail and to a speedy trial and often results in unnecessary detention. On the other hand, the competence referral is frequently used for a more benevolent, although still extralegal, purpose, namely, for the treatment of persons with mental illness who will not accept voluntary hospitalization but who are in need of treatment. Restriction of civil commitment statutes in many states to criteria emphasizing dangerousness has left a large number of severely ill, but nondangerous, people roaming the streets with no means for the state to provide the care that they need. Benevolent judges often seek to thwart the intent of such statutes by committing these people after arrests on minor charges; formally, a competence evaluation is requested, but in actuality, the reason for commitment is the judge’s wish that these individuals receive psychiatric care, even if only for a
short period. This misuse of the competence assessment laws is a result of the pressures exerted on society by civil commitment statutes that are entirely governed by dangerousness criteria (see Chap. 2, Sec. II-D-2). It is another confirmation of the need for involuntary treatment laws that permit treatment for the patient’s own welfare, not merely because someone else is being threatened by the patient’s behavior.

Abuse of the competence referral sometimes emanates from the defendant’s attorney, too. In addition to not protesting extralegal referrals made by the courts, defense lawyers sometimes request competence evaluations when the real issue is their desire to delay the court hearing, whether for their own convenience or for strategic purposes or to obtain a sounding on the level of the defendant’s psychopathology as it relates to the possibility of later offering an insanity defense. Lawyers may also, without consultation with the client, waive their client’s rights to be released from the hospital after the statutorily permitted period of evaluation has expired.

Clinicians can do little to thwart such abuse of the system. Court clinic personnel can try in their evaluations to be as scrupulous as possible about relating their evaluations to the specific criteria for competence to stand trial (see Sec. II-A-3-a). In addition, clinicians can avoid labeling defendants as incompetent simply because they are mentally ill. Public inpatient services have no option but to accept court commitments, but they can try to accomplish the evaluations as expeditiously as possible with an eye toward returning the defendant promptly to court. The most effective move to prevent abuses, and to make the system function more efficiently, is to require that competence-to-stand-trial evaluations be performed on an outpatient basis, unless the defendant otherwise meets commitment criteria or close observation is required as part of the competence assessment. Some states have successfully implemented such outpatient-oriented assessment systems.

Medieval English courts had a ready solution for the problem of what to do with the defendant who was charged with a crime for which she could not be tried. The defendant was simply sent to jail until her condition improved enough to allow her trial to take place. If the defendant’s condition did not improve, the trial did not occur, and the defendant remained incarcerated indefinitely. Surprisingly, this solution to the problem remained the common one in this country until the US Supreme Court decision in Jackson v. Indiana in 1972. The court held that Jackson, who was severely intellectually impaired and therefore unlikely ever to reach a state of competence for trial on charges of stealing $9 worth of property, could not be given the virtual life sentence to which indefinite detention would amount. The state was ordered to release Jackson when it became clear that he was unlikely ever to become competent.

Most states were forced by the Jackson decision to amend their statutes concerning competence assessments. The usual post-Jackson statute provides for a period of evaluation, followed by a period of treatment, at the end of which, if no improvement has occurred and the defendant is not otherwise committable under state law, the defendant must be released and the charges dismissed. These changes have had the effect of preventing the criminal justice system from using incompetence to stand trial as a means of indefinitely hospitalizing those chronically ill, nondangerous persons whom the civil commitment laws also no longer permit to be detained against their will.

On the whole, this result represents a substantial contribution to the fairness with which the system operates. Patients who have chronic mental illnesses but are harmless, many of whom are probably not likely to benefit from treatment, can no longer be picked up on minor charges and hospitalized for life by the court. It is unfortunate, though, that potentially treatable patients in this group cannot be guaranteed care through the more appropriate channel of the civil system. In addition, a small number of patients are chronically ill, repeatedly violate the law in significant ways, but are incompetent and quite unlikely ever to regain competence. Assuming that their crimes do not render them a danger to themselves or to others, there is no way in most states to deal with the problem that they present. If arrested, they cannot be tried and, after an evaluation period and a brief attempt at treatment, must be released. When released, they invariably become involved in illegal activities again (and may in addition discontinue prescribed medications), resulting in their arrest and a repetition of the cycle. Because they are not dangerous, they cannot, in most places, be committed (see discussion of commitment law in Chap. 2, Sec. II-E-1). Our society’s devotion to individual rights leaves no means of dealing with the majority of this admittedly small, but troublesome, class of offenders. It may be the price that society must pay to prevent worse abuses.

One other aspect of the posttreatment disposition of incompetent defendants should be mentioned. A not-insignificant number of defendants who have been restored to competence and transferred back to jail to await trial relapse in jail. The reasons for their deterioration are varied but typically include the minimal psychiatric services available in many jails, the reluctance of jails to enforce involuntary treatment orders, and sometimes just the stress of awaiting trial in an overcrowded and underfunded facility. As a result of their relapse, these defendants are again found incompetent and returned to the hospital, only for the cycle to repeat itself. Potential solutions to the problem of revolving-door admissions for restoration of competence include retaining the defendants in question in the hospital until trial and beefing up psychiatric support in county jails. Many systems, however, have failed to do either, so the problem continues.

We have suggested in this book (see Chap. 5, Sec. III-A-7) that insofar as it is the result of a psychopathologic process, incompetence ought to evoke a search for the underlying illness and an effort at treatment. Some facilities have programs aimed at “treating” incompetence through active education concerning the role and function of the court, including opportunities for defendants to rehearse actual participation in a trial. Although these efforts at education are commendable and perhaps should be made routine for all defendants, not just the ones with mental illness, they do not constitute genuine treatment of the underlying disorder that has led to the incompetent state, because simple ignorance of courtroom procedure should never in itself be the basis for a finding of incompetence. A genuine treatment for incompetence involves intervention directed at the defendant’s psychopathologic state.

The treatment of defendants who are incompetent to stand trial has become a topic of controversy, often stemming from a misunderstanding on the part of many lawyers and judges of the nature of action of psychotropic medications. Their objections affect voluntary and involuntary treatment.

a. Voluntary treatment. Although one would expect that a defendant who wanted to take psychotropic medication that would ameliorate his psychosis and simultaneously restore him to competence would be permitted, even encouraged, to do so, a number of objections to this practice have been raised. Some members of the legal profession think that the state induced by the medications represents “artificial competence,” which differs in some way from the “true competence” that the patient would attain if only he were allowed to improve without medications. People who offer such arguments generally do not recognize the normalizing effect of psychotropic medications: these medicines do not create an artificial state of functioning but return the patient in the direction of his nonpsychotic baseline. The issue is restored competence, not artificial competence. In addition, to ask that many chronic patients attain competence without medication is to require the impossible; if these patients are to be tried at all, it can only be while under treatment. Because the Jackson rule has precluded the indefinite detention of incompetents, objections to voluntary treatment have fallen off. The criminal justice system is usually eager for a trial to take place before the defendant must be released.

b. Involuntary treatment. Involuntary treatment remains a controversial area. The objections to treating incompetent defendants against their will include those that are applied to involuntary treatment in general (see Chap. 3, Sec. II-C). Additional arguments that are offered include (1) the “artificial competence” argument, (2) the belief that it is unethical to infringe on the patient’s autonomy merely to serve society by readying her for trial, and (3) the fear (on the part of the defendant’s attorney) that an insanity defense will be undermined by the defendant’s anticipated nonpsychotic appearance if treated.

The argument regarding the ethics of the decision is a complicated one but seems to lean toward the decision to treat. Not only does society’s interest in trying the defendant outweigh the defendant’s desire to maintain autonomy, but because the defendant’s exercise of autonomy has already been massively impeded by his arrest and incarceration, it comes down to a choice of means of societal coercion, not to a question of whether coercion will take place. Adding to this the interest of society in maintaining the integrity of the criminal justice system by preventing those who would avoid its grasp (in this case, by refusing medications) from doing so, the balance seems to tip in favor of treatment.

A careful explanation to the jury of the effect of the medications should counter the objection that treatment undermines the insanity defense. Although one can never be certain that the jury will be able to discount the effect of seeing an apparently normal defendant before them, our system entrusts the jury with many more complicated assessments; it does not seem unreasonable to ask them to take the effect of medications into account.

Attempts to resolve this issue in the courts have been underway for several decades. Some states allowed defendants to refuse treatment that would restore competence; others did not. The US Supreme Court’s first decision on this issue, in the 1992 case of Riggins v. Nevada, left many of the most important questions unanswered. So, when the court returned to this question in 2003 in Sell v. U.S., its decision was awaited with considerable anticipation. Writing for a six-person majority, Justice Breyer held that the Constitution permitted involuntary medication of incompetent defendants but with several significant caveats. “Important governmental interests” must be at stake, which usually means that the defendant must be charged with a serious crime. Administration of medication must be likely to render the defendant competent and be “substantially unlikely to have side effects that will interfere significantly with the defendant’s ability to assist counsel…thereby rendering the trial unfair,” not determinations that are usually easy to make in advance of treatment. Other means of restoring competence must not exist, and the treatment itself must be in the defendant’s best medical interests. Finally, if other grounds for involuntary treatment exist, such as dangerousness or incompetence to make medical decisions, they should be pursued in preference to a Sell hearing.

Although Sell may define the constitutional minima for involuntary treatment of defendants, states are free to develop more restrictive approaches. How Sell will work in practice is still being defined by the courts. The Supreme Court decision, with its many limitations on the government’s power to treat incompetent defendants, reflects the Court’s hesitance to grant this power to the trial-level criminal courts, which may not have great familiarity with mental illness and the means available for treatment. Justice Breyer indicated his belief that only “rare” cases would meet the criteria he laid out. The extent to which Sell resolves the problems created by incompetent defendants who refuse treatment for their illnesses may not be evident for some years.

The abuses of the competence-to-stand-trial evaluation process plus the enormous expense of evaluating several tens of thousands of defendants each year have led to a number of proposals for altering the system. The most innovative of these involve removing the bar to trying defendants who may lack competence. Two versions of these proposals have been offered.

It has been suggested, and actually implemented in some jurisdictions, that incompetent defendants be permitted to stand trial, with their attorneys making the best defense they can. If the defendant in such a trial is acquitted, she is permitted to go free. If the defendant is convicted, the trial is vacated and she is committed for restoration of competence, and, ultimately, retrial. This approach has the virtue of maintaining adherence to a core value underlying the rules on competence to stand trial: fairness to the defendant. It also diminishes the need to evaluate and treat those defendants who would be found innocent anyway, thus reducing costs and making abuse of the competence assessment system more difficult. On the negative side, it runs the risk of permitting some trials in which the defendant’s inability to assist in her defense or behave properly in the courtroom mocks the adjudication process. If many second trials were required, they might also consume whatever money is saved as a result of reducing evaluations and hospitalizations.

A second, related suggestion goes even further in mitigating the effect of a finding of incompetence. An incompetent defendant would be permitted to waive his right not to be tried in an incompetent state, as long as his attorney, acting, in effect, as a limited guardian, concurred in the decision. The defendant would then take his chances at trial or in a plea bargain. Arguments in favor of this approach emphasize that costs are reduced and that many defendants play little role in preparing or presenting their defense, with their lawyers making the decisions for them, including the decision to keep them off the stand. Not much would be lost in such a case were the trial to proceed even with the defendant still incompetent. This would allow defendants and their attorneys, as a matter of strategy, to risk a finite, often short, sentence rather than accept long-term hospitalization for restoration of competence. Substantial constitutional, ethical, and
practical objections to trying incompetent defendants, however, stand in the way of adoption of such proposals.

An alternative approach now followed in some states results in misdemeanants who are found incompetent having the charges against them dismissed. Although this undoubtedly reduces costs and burden on the system, it creates the possibility of repetitive criminal behavior (e.g., shoplifting) by someone who has learned that she cannot be punished for her behavior.

All of the formulations of the insanity defense require that the impairment claimed in mental functioning be a result of mental disease or defect. Defect is usually understood to refer to intellectual disability. Defining disease is problematic. Few of those in either the legal or the mental health professions would disagree that any psychotic illness, as long as it is not deliberately induced by the defendant’s use of mind-altering agents, legitimately may be considered a disease for the purpose of criminal responsibility determinations. On the other hand, the personality disorders, especially antisocial personality disorder, and other disorders present problems a good deal more controversial.

With the exception of the American Law Institute (ALI) test for legal insanity, the law has no definitive clarification of what it means by disease. It is clear that, in theory, disease is not limited to psychosis (although the actual practice may be quite different). A prominent court offered the following definition: “A mental disease or defect includes any abnormal condition of the mind which substantially affects mental or emotional processes and substantially impairs behavior controls.” Clearly, such a definition sets virtually no limits on what is classified as a disease.

An embarrassing example of the consequences of not clarifying the terms of disease and defect arose in the mid-twentieth century in the well-known Blocker case. The defendant was originally convicted of murder despite his claim of insanity because, although his psychiatrists diagnosed him as having a “sociopathic personality disturbance,” they asserted that he did not have a mental illness. Shortly after his trial, the doctors at St. Elizabeth’s Hospital in Washington, where Blocker had been examined, decided that henceforth they would testify that sociopaths were indeed victims of a mental disease. Blocker’s conviction was overturned on appeal on the basis that he deserved “a verdict based upon the most mature expert opinion available”—in other words, the latest swing of the pendulum.

The issue is clearly not academic. Every clinician who testifies in court in a case in which the insanity defense is used is required to state her opinion as to whether the defendant has a mental disease. As of now, the answers of any set of examiners who concur in the diagnosis of, for example, a personality disorder or an adjustment reaction, do not necessarily agree with each other. The resulting confusion discredits the professions and perplexes the courts.

The inherent circularity of the definition of disease in this context should not be ignored. Mental disease is a prerequisite for exculpation presumably because society recognizes something in the popular understanding of disease that reduces moral culpability. The law could, in fact, define disease in just that way: a state of reduced moral culpability. One judge (Bazelon, concurring in U.S. v. Brawner) made a suggestion along these lines.

To the extent that any psychiatric definition of disease deviates from the tacit moral principles that underlie the insanity defense, it is likely to be rejected by the courts. For example, if organized psychiatry declared that every disorder in the current Diagnostic and Statistical Manual of Mental Disorders (DSM) were actually a “disease,” courts would still properly refuse to accept “tobacco use disorder” as a disease for the purposes of criminal law. As new disorders are defined, the courts must determine whether they more nearly resemble schizophrenia or tobacco use disorder in their potential effects on moral culpability. This process is now occurring for disorders such as pathologic gambling and posttraumatic stress disorder.

Despite this apparent contradiction—the law asks for a definition of disease to aid in determining moral responsibility but only accepts definitions limited to those conditions that appear to impair moral responsibility—there is a point in attempting to formulate a psychiatric definition of disease. First, clinical testimony, for whatever use the courts wish to make of it, will then be consistent and reproducible. Second, in that consistency the law will find a starting point from which to begin the process of clarifying its own approach to mental disease. As things stand, any individual evaluator remains free to decide for himself whether the defendant he is evaluating is mentally ill, as long as he can defend that definition in court.

Mental disease alone does not exculpate a defendant from responsibility for her criminal acts. Something more must be evident. The mental disease, the presence of which has been established by the threshold test, must bear such a relation to the criminal behavior that the perpetrator appears to be morally nonculpable. Exactly what that relation should be has been the subject of voluminous discussion in the legal and psychiatric literature, the result of which has been the formulation of a number of tests of criminal responsibility. Those tests are examined in the following sections.

a. M’Naghten test. The most durable of the tests of criminal insanity (the term “insanity” in the law describes a state of nonresponsibility and is not related to the presence or absence of psychosis) is the M’Naghten rule (pronounced “Mac-naw-ten”), first formulated in Great Britain in 1843. It states that “to establish a defense 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.”

A literal reading of M’Naghten leaves little room for applying it to most mentally ill individuals. Few defendants manifest a “defect of reason” such that they “know” neither the nature and quality of their act nor that it was wrong. But some courts have been more generous in their interpretations, removing “know” from a strict cognitive context and allowing it to connote an
affectively infused appreciation of one’s acts. A schizophrenic patient, in this more generous reading, convinced that her landlord was an agent of the devil bent on turning her mind to thoughts of sex, and who therefore had to be killed, could be held not to “know” the wrongness of her act of murder, although she might respond in the abstract that to murder was wrong.

“Wrong” is the other key word in the M’Naghten test. Courts are split on whether the knowledge that an act is wrong refers to a recognition that the behavior is forbidden by the law or an awareness of its moral wrongfulness. The latter is more consistent with the usual legal holding that ignorance of the law is no excuse, but it allows a significant degree of subjectivity to enter into the assessment by focusing on the offender’s capacity for distinguishing varying shades of moral error. Assuming an impairment in moral awareness is discovered, the forensic examiner has the added burden of determining whether it is a result of the mental illness.

M’Naghten has been the subject of criticism for two reasons: first, by focusing on “knowing” that an act is “wrong,” it is said to limit the scope of expert testimony to cognitive assessments alone; second, in so doing it is alleged to adopt the outdated “faculty” psychology of the nineteenth century, which divided the mind into several discrete, nonoverlapping compartments, thereby ignoring modern understanding of the interrelationship between a variety of mental functions. Nonetheless, some variation of the M’Naghten rule is still the law in the majority of American jurisdictions.

b. Irresistible impulse test. The second standard test of criminal responsibility, the “irresistible impulse” test, is usually used in conjunction with the M’Naghten test. It suggests that the defendant is not liable if, by reason of mental illness, he is unable to exert control over his acts. Once again, this test of volitional control can be construed narrowly, as its eponym suggests, to be applicable only to suddenly arising impulses that are acted on without reflection, or it can be given a broader interpretation, to encompass all acts that result from defective control regardless of the length of time between the initial thought and the act.

The problems for forensic evaluators should be evident immediately. First, it is difficult to differentiate between an uncontrolled act and an uncontrollable act; the post hoc temptation is to assume that every act that has occurred was inevitable. But even more difficult is the acceptance of the test’s underlying premise that disorders of the mind exist that so impair the will as to leave the individual a prisoner of her impulses. Much of the approach of modern psychiatry is built on a negation of that assumption. Even psychotic patients are encouraged to bear full responsibility for their acts and are discouraged from attributing their behavior to impulses over which they have no control. Behavior modification programs, often used with the most regressed or intellectually disabled patients, work from similar premises.

Because the defense is frequently used, however, forensic evaluators must exist who disagree and who believe that people do, in fact, lose control over their behavior. One suspects that in many of these cases the irresistible impulse test is superfluous because such individuals would fail the “knowing” test of M’Naghten (e.g., an individual in a toxic delirium). But for those instances in which this test is relied on exclusively, and the limited data available suggest that a significant number of defendants are reported to qualify for a volitional but not a cognitive, standard of non-responsibility, the question remains open whether any mental illness so impairs behavioral controls as to render the defendant helpless to exert any domination over his impulses.

c. American Law Institute standard. The third major test of criminal insanity, the standard devised by the ALI and published as part of its Model Penal Code, gained in popularity through the early 1980s. It was adopted in the majority of federal jurisdictions and a significant number of states. It reads, in full:

In drafting the Model Penal Code, an effort was made to incorporate the two older standards in their broadest forms. Thus, the “know” requirement of M’Naghten was replaced by the “appreciate” test, an alternative designed to suggest the affective as well as the cognitive awareness of the
nature of one’s acts that the law requires. Similarly, the capacity “to conform one’s conduct to the requirements of the law” allows the broadest possible reading of the irresistible impulse standard.

Unique to the ALI test is the requirement that the defendant have only a lack of “substantial capacity” either to appreciate the wrongfulness of her conduct or to control her behavior, this in place of the total deficits that the older tests seemingly required. Psychiatrists, reluctant to attribute a complete absence of behavioral controls to the defendant, might feel more comfortable testifying to a partial impairment, subject, of course, to the inevitable quibbling over what constitutes “substantial capacity.” In a similar vein, a partial lack of appreciation of wrongful conduct might more plausibly be asserted than a complete one.

The full ALI test pointedly excludes any mental disease or defect “manifested only by repeated anti-social or otherwise abnormal behavior.” Although not all jurisdictions that have adopted the ALI standard have accepted this qualification, it is clear that the intent is to exclude the defendant with antisocial personality disorder from the advantages of the insanity defense. The potential impact of changes in psychiatric diagnosis on the law is nowhere more evident. At the time that the Model Penal Code was published, many psychiatrists argued and many judges accepted that “as the majority of experts use the term, a psychopath is distinguishable from one who merely demonstrates recurrent criminal behavior.” The subsequent DSM-II definition of antisocial personality confirmed this, stressing elements such as selfishness, impulsiveness, absence of loyalty, inability to experience guilt, and a low frustration tolerance and stating expressly, “A mere history of repeated legal or social offenses is not sufficient to justify this diagnosis.” The DSM-5 diagnostic criteria, on the other hand, allow the diagnosis to be made on the basis of a history of legal and social infractions alone; plainly, in any jurisdiction that recognizes the second half of the ALI test, most defendants with DSM-5 antisocial personality disorder must look elsewhere to establish a workable defense. Despite the many protestations of the judiciary that the law, and not the psychiatric profession, determines who is eligible for the insanity defense, this is clearly one example in which the interaction between the two is decisive.

i. Modified American Law Institute standard. In the wake of the trial of John Hinckley, Jr., who attempted to assassinate President Reagan and was found NGRI, a great deal of discussion took place about narrowing defendants’ opportunities to use an insanity defense. One of the most widely endorsed approaches, which received the approval of the American Bar Association and the American Psychiatric Association (APA), involves removing the so-called volitional prong (or modified irresistible impulse standard) from the ALI test, leaving the cognitive or appreciation test on its own. The rationale for this change was the belief that volitional impairments are particularly difficult to ascertain and account for much conflicting testimony by experts.

Persuaded by this argument, although without empirical data to support it, a Congress intent on restricting the use of the insanity defense altered the federal test for criminal responsibility in 1984 to include only a defendant who, “as a result of a severe mental disease or defect, was unable to appreciate the nature and quality or the wrongfulness of his acts.” The use of the phrase severe mental disease or defect was designed to limit the use of the defense by persons with all personality disorders, not just antisocial personality disorder. Dropping the volitional test was thought to eliminate the most dubious cases in which the defense would be used. Some states have followed suit. Whether any of the desired results have been achieved is unclear.

ii. Abolition of the insanity defense. The furor after the Hinckley trial led some states to experiment with an attempt to abolish the insanity defense. The defendant’s mental state could not be removed from consideration entirely, because the law has always required that the criminal act (in legal terminology, the actus reus) be accompanied by an evil intent (the mens rea). The nature of the intent required differs from one crime to another. Montana, Idaho, and Utah, however, abolished an independent defense of NGRI, allowing expert testimony on mental state only when, in the words of the Montana statute, “it is relevant to prove that the defendant did or did not have the state of mind which is an element of the offense.” Data from Montana (see Suggested Readings, Steadman et al.) indicate that, although NGRI findings disappeared after the statutory change, an equivalent number of defendants were found incompetent to stand trial and diverted to the state hospital for long-term treatment, with their charges ultimately dismissed. Thus, a de facto insanity
defense was created in the face of a statute nominally abolishing that option. As the Hinckley trial recedes into the background, the momentum for abolition appears to have faded.

iii. Durham standard. The Durham test, which prevailed in the District of Columbia from 1954 until 1972, allowed a finding of NGRI if the accused’s unlawful act “was the product of mental disease or defect.” This “product” test was designed to allow psychiatrists maximal leeway in introducing evidence relevant to the accused’s mental state. A similar test has long been used in New Hampshire. As clarified in a later case, it represented a “but for” approach to causation: “But for” the existence of mental disease, the act in question would not have occurred. This standard is perhaps most compatible of all with the exculpation of many of the personality disorders, for almost everyone would acknowledge that the “inflexible and maladaptive traits” (DSM-5) that are their characteristic elements contribute substantially to the resultant behavior. The Durham standard died, however, not because of the breadth of its potential scope but because the court believed that the psychiatrists who testified under it were defining “product” in conclusory terms that eroded the fact-finding function of the jury.

iv. Diminished capacity. “Diminished capacity” is a defense based on impairment of the mind that has evolved in several states, beginning in California (which has now abandoned it). It supplements, rather than replaces, the insanity defense, allowing evidence of any interference with the normal functioning of the mind (although in some incarnations of the defense such interference must once more be the result of mental disease or defect) to be introduced to prove that the defendant did not have the ability to formulate one of the specific mental elements required for the crime charged. Thus, an intoxicated individual accused of assault with intent to murder might reasonably claim that she was too drunk to formulate an intent to murder the victim of her assault; the result would be a guilty finding on the reduced charge of assault and battery. The defense has somewhat arbitrarily been limited to crimes that are said to require a higher-level “specific intent” rather than the “general intent” common to all criminal acts.

v. Guilty but mentally ill. Guilty but mentally ill (GBMI) is another variation on the usual techniques for dealing with the mentally disordered offender. Juries can find a defendant pleading insanity GBMI if they believe that his illness existed at the time of the crime but did not contribute to the act to such an extent that a finding of NGRI is warranted. The mental health profession’s role is not much different in this instance from that in an insanity defense; the contribution of the individual’s illness (again, the problem of defining the illness arises) to his act must be assessed. Persons found GBMI are then referred for evaluation and are supposed to receive psychiatric treatment, if that is warranted.

The GBMI verdict became more popular after the Hinckley trial, in large part because it was seen as a way of persuading juries that they could recognize a defendant’s mental illness without finding her NGRI; when applied for this purpose, it is something of a fraud. Those found GBMI are treated the same as those who are convicted in an ordinary manner; any prisoner can be evaluated for treatment and should receive it if needed. To the extent that jurors are led to find a defendant GBMI in the belief that such a verdict differs materially from a finding of guilty and is akin to a verdict of NGRI, they have been tricked. Nonetheless, approximately one-fourth of the states adopted a GBMI option, largely in the apparent hope that just such misunderstandings would occur.

Studies suggest that GBMI verdicts tend to occur in cases involving the most violent crimes and often result in periods of detention exceeding those either of persons found guilty or those found NGRI for those offenses. Insofar as a GBMI option places a defendant who pleads NGRI at risk of a more punitive response than a straightforward guilty finding, it may further reduce the attractiveness of an NGRI plea (see Suggested Readings, Steadman et al.). If there is any advantage to the plea, it may be the dubious one that defendants and their families alike can attribute the criminal act to the consequences of illness, rather than taking full responsibility for it.

a. Raising the defense. The procedural details of implementing an insanity defense vary from jurisdiction to jurisdiction, but in general any party in a criminal proceeding can raise the issue of a defendant’s criminal responsibility at any point in the process. Defense attorneys
who intend to raise an insanity defense are required in many states to notify the prosecution in advance of the trial or forfeit the option. Prosecutors often use the request for an examination of responsibility as a technique for detaining without bail those defendants whom they prefer not to see released before trial; judges often go along with these requests for similar reasons. Examinations of this sort are usually conducted in inpatient facilities whose security varies with the seriousness of the alleged offense. Allowing the prosecutor and the judge to request responsibility examinations provides an interesting potential for them to redefine “political” crimes as the act of a madman; how often that opportunity is exploited in the United States is unknown, but the highly publicized Anders Breivik case in Norway offers an example from another jurisdiction. Breivik was an admitted mass murderer who the state argued was legally insane, perhaps in an effort to undercut the political statement he intended to make with his acts. In contrast, he argued that his actions, which he did not deny, were justified by the necessity of saving his country from the policies of a left-of-center government. In the end, he was successful in rebutting the insanity defense urged on the judge by the prosecution but was found guilty of murder and incarcerated.

b. Obtaining an Examination. Any defendant is entitled to use his own expert witness and, since the US Supreme Court’s 1987 decision in Ake v. Oklahoma, states have been obligated to provide forensic mental health professionals to defendants facing major charges who cannot afford to hire their own. Which cases this applies to, what level of funding the states must offer, and whether the expert must work only for the defense are among the issues that are still being resolved at the state level. If the defense attorney elects not to use the testimony of a forensic expert who has examined her client, presumably because it does not support an insanity defense, the prosecutor may, in some jurisdictions, later call that expert to testify. Clinicians should be aware of the rules in their own jurisdictions, in part so they can provide accurate information to the defendant they evaluate concerning the level of confidentiality that the examination provides.

c. Burden of Proof. Among states, rules concerning who bears the burden of proof in insanity defense cases vary. After the Hinckley trial, many states and the federal government placed the burden on the defendant to prove legal insanity by a preponderance of evidence. This proved to be the most effective way of decreasing the number of successful insanity pleas and is now the rule in the majority of states. Some states still retain the requirement that the prosecution prove the absence of legal insanity beyond a reasonable doubt, reasoning that, because a mens rea is an essential element in any crime, the prosecution should bear the burden of proving that the defendant was actually criminally responsible. To trigger this burden for the prosecution, the defendant only needs to present evidence to the contrary. In such cases, the job of the expert witness for the prosecution, who must convince a jury beyond a reasonable doubt that a mental condition did not exist, is considerably more arduous than the task of the defense’s expert, whose testimony must only raise some measure of doubt.

Historically, defendants were loath to raise an insanity defense because a finding of NGRI led almost invariably to an indefinite, potentially lifetime commitment to a psychiatric hospital, usually one for the “criminally insane.” Because all but major crimes (e.g., murder) held out the hope of eventual release from prison and because there was a chance that even the most seemingly guilty defendant could convince a jury of his innocence, it was almost always in the noncapital defendant’s interest to eschew the insanity defense.

The insanity defense became more attractive with a number of court decisions that declared it to be a violation of a defendant’s right to due process to be incarcerated indefinitely regardless of his current mental status or the severity of his crime. More recently, rules in most states allow a period of hospitalization for evaluation, often up to several months, after a verdict of NGRI. Persons who are deemed to be mentally ill and dangerous can be hospitalized for an extended period with periodic reviews of their status. If they remain ill and dangerous despite treatment, recommitment must take place on a regular basis. In some states, this process operates with rules identical to those used for civil committees. In others, variations on the rule (i.e., placing on detainees the burden of proving nondangerousness) are designed to make it more difficult to release acquittees.

The revolution in disposition procedures has heightened the desirability of using the insanity defense and has probably been responsible for an increase in its use. Also contributing to the increased use of the insanity defense is the shift of responsibility for the care of some categories of mentally ill minor offenders from the penal to the psychiatric systems. With a greater sensitivity to the presence of mental illness, the criminal justice system now frequently channels such offenders to state hospitals. Once labeled as psychiatric cases, these offenders are more likely to use and to be successful in using an insanity defense in response to future charges. Obviously the more rapid turnover of NGRI acquittees, mandated by the court decisions requiring a finding of dangerousness for continued incarceration, may also contribute to the increased subsequent use of the defense.

The most important innovation in postacquittal procedures has been the Psychiatric Security Review Board (PSRB) model initiated in Oregon and since adopted, sometimes under different names, in a number of other states. Insanity acquittees in Oregon are committed to the PSRB for the maximum time that they might otherwise have been incarcerated if they had been found guilty. The PSRB can place them in inpatient or outpatient treatment and can follow them after discharge from an inpatient facility. If the acquittee begins to deteriorate or does not comply with discharge conditions, she can be rehospitalized expeditiously, followed by a hearing before the board. This approach, endorsed by the APA as a model for the rest of the country, gets around the problem that exists now in most states: Once an acquittee no longer qualifies for inpatient hospitalization, often according to civil standards, the state loses all control over her. The PSRB model allows long-term monitoring of a class of people about whom society has every reason to be concerned. It appears to be successful, based on recidivism rates, and warrants careful consideration in every state.

Steadman and coworkers’ (see Suggested Readings) landmark study of insanity defense reform showed an unanticipated effect of tighter community monitoring in New York: an increase in the willingness of those involved in the criminal justice process to agree on NGRI verdicts in cases of violent crime. Apparently, judges, juries, and prosecutors are reassured that an NGRI finding does not mean that a potentially violent offender will be released into the community without adequate treatment and supervision.

Many of the issues involved with mental health professionals’ role in determining responsibility are analogous to those discussed previously (see Sec. II-A) with reference to competence to stand trial. Despite the proliferation of legal standards for determining responsibility, little attention has been devoted to translating those standards into clinically meaningful terms. Law and psychiatry have yet to find a common basis for discourse.

The basic legal doctrine can be stated simply: “Our criminal laws are premised on the view that human beings are normally capable of free and rational choice between alternative modes of behavior, that an individual who chooses to harm another is morally blameworthy or guilty, and that he is liable to punishment if his behavior and the resulting harm have been proscribed by the law” (see Suggested Readings, Brakel and Rock). From these basic presumptions, the law formulates its expectations of the mental health professions.

Clinicians, accustomed to speaking their own private language, are forced to speak in legal terms. They are asked whether individuals have acted with appreciation of the nature of their acts or with the free will to avoid violations of the law if they so choose. They are sometimes asked to evaluate the effect of a complex disease process on a remote act that they did not witness and that the alleged perpetrator may maintain has never occurred. They are asked to define mental disease and to apply the definition to the defendant. During the process, they must keep in mind that the system for which they are working (i.e., judicial) denies a fundamental theoretical premise of much of psychiatry (i.e., that all behavior is influenced by unconscious forces beyond the control of the actor and is thus, to some extent, involuntary and predetermined).

It should not be surprising that in this situation a number of studies have cast doubt on the objectivity of forensic evaluators. Their judgments concerning the presence or absence of criminal responsibility have been shown to relate to their political and social views in addition to taking into account the characteristics of the defendant. Part of the public discomfort with the insanity
defense undoubtedly relates to the perception that mental health professionals are allowing their own opinions to influence the outcome of major trials in ways that confuse, rather than clarify, the issues to be decided.

Therefore, a strong argument can be made that forensic clinicians should avoid offering testimony on those aspects of the insanity defense that require more political or moral, rather than clinical, judgments. This means avoiding testimony on the ultimate issue of the defendant’s responsibility; such a limitation was adopted by Congress for the federal courts as part of the post-Hinckley reforms. Clinicians may also benefit, even in this more limited role, from the development of structured assessment instruments to guide their evaluations. Several instruments exist, but they suffer from deficiencies that will, it is hoped, be rectified in subsequent research.

a. Legal criteria. The basic legal requirements for competence to write a will, also known as testamentary capacity, vary somewhat across jurisdictions. A common formulation is that those who would write a will “retain the power to understand the nature and extent of their property, their relationship to those persons who are usually the objects of a person’s bounty, and the nature and operative effect of will making.” In this brief definition, the societal interest in competent will making is quickly evident. It is an interest primarily of the heirs, or (more accurately) the presumptive heirs, in assuring (1) that the estate is not carelessly dissipated (hence, the requirement for an understanding of the extent of the property and of the fact that one is executing a document that controls its disposition) and (2) that they are in fact the beneficiaries (thus, the phrasing “those persons who are usually the objects of a person’s bounty”).