In July 2011, Anders Breivik, a 32–year-old Norwegian man, began a killing rampage that left 77 people dead. His first attack occurred in a cluster of government buildings in downtown Oslo, where he set off a car bomb made from fertilizer and fuel oil, killing eight people. Breivik then drove to a nearby lake, taking a ferry to Utoya Island, the site of a summer camp run by the youth wing of the country’s ruling Labor Party. Dressed as a police officer and armed with a carbine and a pistol, Breivik roamed the island for over an hour, killing 69 additional people and wounding 66 others. When police finally appeared, he surrendered without a struggle (1).

At his trial for murder nine months after the crimes, Breivik readily admitted to the killings but argued that they were “the most spectacular sophisticated political act in Europe since the Second World War” (2). Breivik said that his actions were taken in “self-defense” to protect Norway from the growing influence of Muslim immigrants and multiculturalism. Two pretrial psychiatric reports reached opposite conclusions about his mental state at the time of the crime: one described him as psychotic due to paranoid schizophrenia (3), whereas a second evaluation concluded that he had not been psychotic but did have a narcissistic personality disorder (4). Because there was no question that Breivik had committed the acts with which he was charged, the trial centered on whether he had been legally insane at the time of the crimes.

Although the insanity defense is uncommon—it is seriously entertained in well under 1% of felony prosecutions in the United States (5)—it is often the choice of defendants and their counsel when no other defense is plausible. However, Breivik’s trial presented the unusual spectacle, apparently unprecedented in Norway, of a defendant who insisted that he was sane while the prosecution argued for a finding of insanity. In essence, the prosecution was attempting to have an insanity verdict imposed over the objections of the defendant, in the face of conflicting evidence about his state of mind. The possibility, for which there are precedents in American law, raises complex questions about whether defendants should be allowed to opt out of an insanity defense, even when that seems to be the appropriate verdict.

Given that defendants and their attorneys ordinarily get to decide trial strategy, including such consequential decisions as pleading guilty in exchange for a lesser sentence, what justification might exist for imposing insanity pleas on defendants who insist that they were not insane at the time of their crimes? Perhaps the most compelling argument is that the prospect of convicting defendants who were not responsible for their actions offends our basic notions of justice. The leading case to offer this rationale in affirming the involuntary imposition of an insanity defense was Whalem v. United States, a 1965 decision of the U.S. Court of Appeals for the District of Columbia Circuit (6), in which the court offered the following defense of the practice: “[T]he trial judge must uphold [the foundation of the criminal law in the concept of individual responsibility] by refusing to allow the conviction of an obviously mentally irresponsible defendant, and when there is sufficient question as to a defendant’s mental responsibility at the time of the crime, this issue must become part of the case.”

However, the decision in Whalem does not stand unchallenged. The judges in Frendak v. United States, another case from the nation’s capital, this time from the District of Columbia Court of Appeals (the equivalent of a state supreme court in other jurisdictions), took a very different view (7). Relying on U.S. Supreme Court decisions on other aspects of criminal procedure, including the right to act as one’s own attorney (8), the Frendak court concluded that a competent defendant could waive the right to an insanity defense as long as that waiver was intelligent and voluntary. By “intelligent,” the court was clear that it did not mean that the waiver of an insanity defense had to be wise as long as it was based on a rational thought process. Only in the case of a defendant who failed this test could an insanity verdict be imposed. The ability of a competent defendant to manage his or her own defense took precedence in Frendak over whatever societal interest might exist in accurately identifying defendants who were not responsible for their behavior.

Where do the states stand on whether to allow an insanity plea to be considered over a defendant’s objections? A 1996 survey, apparently the most recent, found that 17 of the 48 jurisdictions with an insanity defense would permit nonconsensual pleas to be imposed by the court (9). Interviewing 50 consecutive defendants who had pursued an insanity defense in Colorado, the same authors reported that 32% of insanity defenses appeared to have been pled against the defendant’s will, a much greater frequency of imposed insanity defenses than anyone had previously suspected. One scholar noted, however, that what was likely at issue here was the failure of attorneys to consult with their clients regarding the defense (which another study had indicated was common attorney behavior [10]) rather than a strategy adopted over the defendants’ overt objections (11).

Imposition of an insanity defense, whatever the incidence of the practice, has been disfavored by most commentators (9,12,13), who offer an imposing array of arguments that fall into two broad categories. First, there are the pragmatic arguments as to why a defendant who qualified for an insanity defense might not want to invoke the plea. Defendants found not guilty by reason of insanity (the precise term for the verdict differs from state to state) almost always undergo automatic commitment for evaluation, usually to a forensic facility, with the prospect of indefinite hospitalization until they are found no longer dangerous or mentally ill. When facing charges with relatively short sentences, defendants might reasonably prefer to be found guilty and serve their time in jail or prison rather than risk confinement without end in a forensic hospital—a choice commonly seen in the criminal courts. They might also simply prefer being in a prison to being on a psychiatric unit, or they may want to avoid the stigma of being found to have a mental illness, the notoriety of a public trial, or the acknowledgment of having committed the act with which they are charged—even while in a mental state that rendered them not responsible.

A second set of reasons offered to shun an imposed insanity defense derives from legal notions of fundamental fairness for a defendant. The Sixth Amendment to the Constitution has been interpreted as reflecting a right to personally guide one’s defense to criminal charges. Therein lies the basis for the U.S. Supreme Court’s decision in Faretta v. California (8), which recognized a defendant’s right to waive the assistance of counsel and undertake his or her own defense (although the Supreme Court subsequently limited that right in Indiana v. Edwards for defendants who lack mental competence to perform that role [14]). Indeed, although the courts have been somewhat ambiguous on the question, the right to control one’s own defense may be so central to fairness in the criminal process as to be protected by the Fourteenth Amendment’s guarantee of due process (13).

In addition to these arguments, there is one other consideration that militates in some cases—including the prosecution of Anders Breivik—against an imposed insanity defense. At his trial, Breivik and his attorneys argued that his actions were undertaken solely with a political motive in mind: he was defending Norway against “Islamic colonization” (15). Regardless of how one feels about the motive, it is clear that the effect of finding Breivik not guilty would have been to discredit his actions as the work of a “madman.” Precisely because of the value of an insanity label in resisting any challenge to the dominant political model, totalitarian regimes such as the Soviet Union and China, the latter even today, prefer to impose psychiatric hospitalization on dissenters rather than respond to their dissatisfactions. Many observers thought Norway’s prosecutors were seeking a similar goal in the Breivik case, trying to exclude anti-immigration sentiment from any place in political discourse. Even in a democracy, the temptation to use an imposed insanity defense to discredit opponents is real, and it constitutes another reason for caution in allowing a court to substitute its preference for the choice of the defendant.

No one relishes the prospect of a defendant with a serious mental illness who might qualify for a finding of nonresponsibility being found guilty after foregoing a viable insanity defense. But the arguments against imposing the defense are substantial. At the same time, there is no question that some defendants shun an insanity defense for reasons that reflect an underlying psychiatric disorder. They may lack insight into the presence of an illness or may base their refusal on delusional beliefs. For example, some psychotic defendants are convinced that they are certain to be acquitted if only they can explain their motives and the righteousness of their cause to a jury (16). With our general concern for protecting criminal defendants with psychiatric illness from the negative consequences of their disorders, an approach is needed that accommodates both kinds of concerns.

For several decades, scholars have pointed to the determination of competence to stand trial as the solution to this dilemma. Once defendants are found competent, they should have the right to decide what defenses to abandon or pursue, including foregoing an insanity defense (12,13). But in determining whether a defendant is competent to proceed, courts should take into account the defendant’s beliefs about a defense of insanity (11). When a defendant cannot recognize the presence of a serious mental disorder or bases a choice to forego a nonresponsibility defense on delusional premises, the defendant should be considered to be incompetent to stand trial. In those cases where a defendant’s choice has a rational basis—for example a preference for the determinate sentence that will follow a finding of guilt compared with indeterminate commitment to a forensic facility—the defendant should be free to shun pursuit of an insanity verdict, if that is his or her choice.

In cases such as Anders Breivik’s, there would have been an additional benefit to such an approach. Despite prosecutors’ arguments that Breivik was insane, the court found him guilty and sentenced him to the maximum possible time in prison under Norwegian law (17). Hence, Breivik’s political motives, in the end, were taken seriously. Not to have done so would have risked steering Norway and other European countries away from the genuine threat represented by fringe—but real—political views. Similar threats, albeit reflecting different perspectives, materialized in violence in the Oklahoma City federal building bombing in 1995 and the London Underground attacks in 2005. To have cast these acts as manifestations of “insanity” would have undercut the appropriate responses that ensued, based on recognition that right-wing militia members in the United States and Islamic militants in Britain both posed risks of terrorism and needed to be monitored to prevent future violence. This is an instance in which good law not only guards the rights of defendants but also protects the well-being of the public.

The author reports no competing interests.