Settled Insanity: Substance Use Meets the Insanity Defense

In November 2013, Aldo Dunphe, then 23 years old, was voluntarily admitted to a psychiatric unit at UMass Memorial Medical Center in Worcester, Massachusetts, for treatment of an acute psychotic episode. Dunphe, who had been married less than 3 weeks before, had reportedly been a heavy cannabis user for at least the past 3 years. In the months leading up to the wedding, he appeared increasingly paranoid, accusing his fiancé of infidelity. The accusations continued after their wedding, along with increasingly fragmented thoughts, staring into space, and smiling at inappropriate times. He heard a radio station tell him that his wife was committing adultery. After he became convinced that he was being followed by a biker gang and spoke of taking an overdose to end his life, he agreed to hospitalization (1).

Shortly after admission, Dunphe reported having seen his biological father, who had abused him as a child but who lived in Guatemala, on the ward. Pointing out another patient, a 31-year-old immigrant from Bhutan, Dunphe told a nurse, “The little guy is my father. He’s been in a time travel. He’s been gone for 2,500 years. I’ll do that, too, but I can do it in 10.” The next day, as described in the court opinion, Dunphe “suffocated the victim with a pillow, punched him repeatedly in the face, smashed his head against the cement floor, and finally stuffed cloth into the victim’s nose and mouth.” When the victim died a few days later, Dunphe was charged with first-degree murder.

At trial, Dunphe pleaded not guilty by reason of insanity. The forensic psychologist who testified for the defense said that Dunphe had perceived the victim as threatening him and thought that he was acting in self-defense. He opined that Dunphe’s delusions, hallucinations, and disorganized behavior were the result of paranoid schizophrenia, which, in the psychologist’s view, deprived him of the ability to appreciate the wrongfulness of his behavior or to conform his behavior to the requirements of the law—thus meeting Massachusetts’ standard for legal insanity. In contrast, the psychiatrist hired by the prosecution testified that Dunphe did not have schizophrenia but was experiencing psychotic symptoms as a result of cannabis withdrawal. Because withdrawal from a substance that was voluntarily consumed does not serve as an acceptable basis for an insanity defense, the prosecution expert concluded that Dunphe was not legally insane.

After hearing from both experts and receiving a charge from the judge, the jury decided that Dunphe was guilty of first-degree murder, having acted with “extreme atrocity or cruelty.” He was sentenced to life imprisonment without the possibility of parole. Dunphe appealed his conviction on the grounds that the judge had not properly charged the jury. But to understand the basis for his appeal, a more detailed exploration is needed of the complicated relationship between substance use and the insanity defense.

The belief that an extreme disturbance of mental state—what the law has traditionally termed “insanity”—precludes punishment for criminal behavior has historical roots that in the Anglo-American legal tradition can be traced to medieval times (2). The precise standard used to determine legal insanity has varied over time and jurisdiction but generally involves either a failure to recognize that one is committing a wrongful act or an inability to control ones’ behavior as a result of a mental disorder. Various justifications for the insanity defense have been offered, including the unfairness of punishing someone who did not grasp the wrongfulness of their behavior or lacked the ability to control it and the lack of a deterrent effect when such people are punished. Although the U.S. Supreme Court recently decided that an insanity defense is not constitutionally required, 46 states and all federal jurisdictions retain the possibility of defendants pleading insanity (3).

States of intoxication and withdrawal, however, have always been excluded as qualifying conditions for an insanity defense, even though these states may lead to impairments in the ability to understand the wrongfulness of one’s actions or to control one’s own behavior that are comparable with impairments seen, for example, with psychotic disorders. As a matter of legal principle, it is generally assumed that people should be held responsible for the choices they make, including the decision to consume an intoxicating substance, which they know or should know can lead to altered states of consciousness and behavioral disinhibition. From a policy perspective, it is often argued that allowing intoxicated behavior to result in exculpation may increase the propensity to consume substances to excess, thus encouraging criminal acts, or lead offenders to feign intoxication to immunize themselves from punishment. The only situation in which voluntary intoxication has a partially exculpatory impact is when the effect of a substance renders a person unable to form the state of mind required for a “specific intent” crime (e.g., when intoxication leaves a person too disorganized to formulate the intent to kill required for a charge of first-degree murder), in which case the defendant would be convicted of a lesser charge (4).

However, U.S. courts, beginning in the mid-nineteenth century, have recognized one circumstance in which behavior resulting from the consumption of intoxicating substances can be the predicate for an insanity verdict. When consumption of a substance results in an exculpatory mental state that persists beyond the period of acute intoxication, the courts have allowed insanity verdicts, deeming such cases instances of settled insanity. The rationale for this deviation from the general rule that voluntary consumption of intoxicating substances never excuses criminal actions appears to be that, in contrast to those experiencing acute intoxication, persons consuming substances neither seek nor desire the remote effects that impair their mentation or behavior. At least 29 states and the District of Columbia have accepted settled insanity as the basis for an insanity defense, and only one state, Colorado, has explicitly rejected it (5).

California courts have identified four criteria to determine whether a condition qualifies as settled insanity. The condition must be fixed and stable, last for a reasonable duration (although not necessarily permanently), not be solely dependent upon the ingestion and duration of effect of the drug, and meet the jurisdiction’s legal definition of insanity (6). The most straightforward example of such a condition is what is now referred to in the DSM-5 as substance-induced neurocognitive disorder. Along with cognitive impairment, associated features may include “loss of emotional control, including aggressive or inappropriate affect, or apathy” (7). Substance use can also induce psychotic states that endure beyond acute intoxication, although the DSM-5 suggests that if such symptoms persist for more than a month after intoxication or withdrawal, they may be more likely to result from an independent psychotic disorder. A large proportion of people with substance-induced psychosis go on to develop schizophrenia or bipolar disorder, suggesting that an underlying psychotic diathesis interacted with the substance to produce the psychotic state (8).

Aldo Dunphe’s appeal of his conviction was based in large part on a challenge to the instructions that the trial judge gave to the jury that Dunphe argued precluded a finding of settled insanity. The jurors were instructed, in part, “Where a defendant lacked substantial capacity to appreciate the criminality or wrongfulness of his conduct or to conform his conduct to the law solely as a result of voluntary intoxication, then he is criminally responsible for his conduct.” Although Dunphe’s expert witness diagnosed Dunphe as having paranoid schizophrenia without explicitly linking his psychosis to his cannabis use, Dunphe argued that the testimony of the prosecution’s psychiatrist could have led a reasonable jury to conclude that there was a causal relationship between his consumption of cannabis and his psychosis. However, if the jurors reached that conclusion, this wording might have led them to believe that a verdict of not guilty by reason of insanity was precluded, because Dunphe’s condition was the result—albeit a remote one—of voluntary intoxication. That is, Dunphe claimed that the jury instructions deprived him of the opportunity for a settled insanity defense, which was otherwise available under Massachusetts law.

When his initial appeal was denied, Dunphe took his case to the state’s highest tribunal, the Supreme Judicial Court (SJC). The justices, in a unanimous opinion, began by noting that the trial judge faithfully conveyed the model jury instructions that the SJC itself had developed (1). However, in a rare example of judicial self-criticism, they agreed with Dunphe that the language of the instructions might have misled the jurors to conclude that any consequence of voluntary use of intoxicating substances precluded an insanity claim, which was contrary to Massachusetts law. Moreover, the potential impact of the misleading language was heightened by the prosecutor’s closing statement, in which he suggested that the defendant should bear the consequences of his drug use, whatever they were. However, the justices underscored that this was simply incorrect. “What our case law declares, but our model jury instructions do not, is that if a defendant has a mental disease or defect, its origins are irrelevant: it does not matter whether the disease or defect arose from genetics, from a childhood disease or accident, from lead poisoning, from the use of prescription medication, or from the chronic use of alcohol or illegal drugs.”

To remedy the problem, the justices modified the model jury instructions in several ways, including inserting a new sentence specifying that “a mental disease or defect might be caused by or result from a defendant's earlier chronic use of alcohol or drugs.” Thus, settled insanity was reinforced as settled law in Massachusetts, as it is in the majority of U.S. jurisdictions. As for Dunphe, given the likelihood that he had been subjected to “a substantial miscarriage of justice” in being denied a settled insanity defense, his conviction was vacated, and his case was remanded for retrial.

Notwithstanding the rarity of claims of insanity based on the remote effects of intoxicating substances, the doctrine has evoked considerable debate. Arguments for its abolition note that, as with acute intoxication, the long-term effects of substance use result from volitional choices of the defendant for which he or she should bear responsibility. “There is no principled basis to distinguish between the short-term and long-term effects of voluntary intoxication by punishing the first and excusing the second. If anything, the moral blameworthiness would seem to be even greater with respect to the long-term effects of many, repeated instances of voluntary intoxication occurring over an extended period of time” (9). It has also been suggested that imposing punishment in such cases would help to deter substance abuse and its criminal consequences, however unlikely that seems (5). Proponents of the doctrine respond with the claim that remote effects of intoxicants, unlike acute intoxication, are variable in their occurrence and are rarely considered when substances are consumed.

Perhaps the strongest argument from a psychiatric perspective for the retention of a defense based on settled insanity, however, stems from growing knowledge of the complex roots of mental disorders (10). Psychosis after exposure to substances, whether brief or prolonged, appears to occur more commonly among people with predispositions to developing psychotic disorders, as demonstrated, for example, by their family histories and subsequent course (8). The same may be true for substance-induced neurocognitive disorders, which are more common among older people, in whom vascular changes and other causes of dementia may also be playing a role. Given the difficulty—and in many cases the impossibility—of separating syndromes induced by endogenous predispositions from conditions caused entirely by substances and given the likelihood that in most cases both sets of factors are at play, focusing on a defendant’s clinical state rather than its causes is probably the fairest and most practical approach.