Elements of the Test

Elements of the Test

The traditional rule has three working parts. A defendant is entitled to an acquittal if, as a result of mental illness, the defendant could not control his actions or could not conform his conduct to the requirements of the law, and the lack of control may develop or exist over time rather than arising suddenly.¹ The first part is the causal link, which ties the loss of control to a mental illness. The second part is the inability itself, the defendant's incapacity to govern his behavior or bring it into line with the law. The third part is the timing clause, which removes a requirement of suddenness that the word "impulse" might seem to suggest. Each part answers a different question about the defendant, so the test is easier to apply when the parts are taken one at a time than when a single question is asked about whether the defendant could not help it.

The division also reflects the different jobs the parts do. The causal link works as a gate. Almost any offender who acts on a strong want can say that the want felt overpowering, and if a strong desire alone were enough, the law would have no means of separating a disease-driven loss of capacity from ordinary willfulness. The defense would then swallow the rule of responsibility. The inability inquiry fixes attention on the capacity to choose. Because the defendant may well have known the act was wrong, the question turns on whether the illness took away the power to refrain, and how the act looked from the outside does not decide it. The timing clause prevents a mistaken inference in the other direction, since an act carried out slowly and with preparation can still come from an illness that disabled the will throughout.

The causal link comes first in the analysis because the other two parts matter only once an illness is in the picture. The next subsection therefore begins with mental illness as the source of the lost control, and the two that follow take up the inability and the timing clause in turn.

Mental Illness as the Source

The control test excuses a defendant only when mental illness is the reason control failed. The traditional rule says so in its opening clause: the defendant must have been unable to control his actions or to conform his conduct to the law as a result of mental illness.² A defendant who cannot point to an illness therefore never reaches the question of control, however sincerely he insists that he could not stop. The same ordering appears in the Parsons formulation, which begins by asking whether the defendant was afflicted with a disease of the mind and closes by requiring that the crime be connected to that disease as cause and effect.³ A trial court's charge reproduced in Fisher v. United States gives jurors the point in plain terms: a lack of willpower and control must be the result of a disease or disorder of the mental faculties, and "[m]ere loss of moral restraints leading to a surrender to criminal thoughts and passions is not enough."⁴

The requirement has two parts, and each does separate work. The first is the existence of an illness, which ties the defense to a condition that is gross and verifiable, because the criminal law does not adjust its standards to an individual's capacity to meet them unless the disability can be established the way mental disease or defect can.⁵ The second is the causal link, which matters because a defendant may be ill and still have acted for reasons the illness did not supply. Underneath both lies an incentive problem. A claim that an urge could not be resisted rests on an inner state that only the defendant can describe, and nearly everyone who commits a crime on a strong want has reason to describe it that way. A rule that excused any overpowering want would excuse the very impulses criminal sanctions exist to deter, so the law demands a recognized illness as outside evidence that the disability is real. That is why a defendant who says he would not have committed a violent crime had he been a less excitable person offers no defense, and why a thief cannot excuse a theft by pointing to an acquisitive nature.⁶ Students lose this issue when they treat a strong urge, standing alone, as if it were the disability the test requires.

What counts as mental illness is a legal question rather than a clinical one. Insanity is a legal standard, and the working term mental disease or defect covers disease of the mind, congenital lack, and damage from traumatic injury, while excluding the excitement or stupefaction that liquor or drugs produce.⁷ Lawmakers outside the criminal law draw the same connection: a Washington civil-commitment statute quoted in Washington v. Harper defines mental disorder as an impairment with substantial adverse effects on cognitive or volitional functions, which treats impaired volition as an effect of a disorder.⁸ In United States v. Lyons, a defendant who had become addicted to prescribed narcotics offered expert evidence that the addiction left him unable to conform his conduct, a claim made under the broader substantial-capacity formulation discussed later in this chapter. The Fifth Circuit, sitting en banc, stated that addiction standing alone, without other physiological or psychological involvement, raises no issue of mental disease or defect, while actual physical damage to the brain does fall within the term.⁹

The causal link also screens out explanations that sound like causes but are not illnesses. In United States v. Alexander, the defendant shot a Marine in a hamburger shop after a racial insult, and the defense offered expert testimony that an irresistible impulse to shoot came from the defendant's socially and economically deprived childhood.¹⁰ The trial judge told the jury to disregard the testimony about that background and to consider only the insanity plea, and the court of appeals affirmed over Judge Bazelon's dissent.¹¹ The outcome shows what the causal link demands. A history of deprivation may help explain why a person's controls were weak, but the test asks whether a mental illness produced the loss of control, and an upbringing is not itself a disease of the mind. Had the same history left the defendant with a diagnosable disorder that disabled his will, the disorder would have been the source and the inquiry would have moved on to whether it did.

Voluntary intoxication is the most common way the causal link fails. A defendant is not insane just because he is intoxicated, since insanity requires a disease of the mind and drunkenness cannot supply one, and being legally sane he is ineligible for the insanity defense, including its control version.¹⁴ The reason for refusing to treat self-induced impairment as illness appears in Montana v. Egelhoff, where the Court's opinion upheld a state rule excluding voluntary-intoxication evidence on the mental-state question. The opinion observed that the rule fits society's moral perception that a person who voluntarily impairs his own faculties should answer for the consequences, and that it deters people from drinking to the point of violence.¹⁵ Two nearby doctrines are only flagged here because each rests on its own grounds: most jurisdictions treat involuntary intoxication like insanity, so an involuntarily intoxicated defendant may invoke the control inquiry,¹⁶ and intoxication may also negate a mental-state element of the offense.¹⁷

A last boundary separates the control test from unconsciousness. In People v. Grant, an Illinois appellate court treated acts committed during a psychomotor epileptic seizure as a matter of automatism, the state of a person who is capable of action but not conscious of what he is doing, and not as insanity.¹⁸ That defense needs only a showing of lack of volition, while insanity also requires a mental disease or defect.¹⁹ The Georgia Supreme Court classified sleepwalking the same way in Smith v. State, holding that the trial court erred by forcing a sleep-disorder defense into the insanity framework, because the defendant claimed he had not acted voluntarily at all.²⁰ A claim of that kind runs outside the mental-illness requirement, so it does not need the control test and cannot borrow from it.

Inability to Control Conduct

Once an illness is established as the source, the inquiry turns to what the illness did to the defendant's capacity for self-command. Under the traditional rule, the defendant is entitled to an acquittal if, as a result of mental illness, he could not control his actions or could not conform his conduct to the requirements of the law.²² This question is separate from the one the right-wrong test asks. The control test calls for a verdict of not guilty by reason of insanity even when the defendant knew what he was doing and knew that it was wrong, and his mental condition need not satisfy the right-wrong test as well.²³ The inquiry therefore picks up exactly where the cognitive questions leave off. The defendant understood the act and its wrongfulness, and the only issue left is whether that understanding could be turned into restraint.

Everything in this inquiry depends on the difference between could not and did not. A defendant who failed to resist has not thereby shown that he was unable to resist, because every wrongdoer who gives in to temptation has, in the plain sense, failed to hold back. The test excuses only the person whose illness removed the power to hold back, so the evidence must show a capacity that the illness disabled, which is a different thing from a capacity that simply went unused. The law's own description of an uncontrollable impulse makes the same point: it is an impulse so overwhelming that it cannot be resisted.²⁴ That vocabulary of impossibility is what separates the defense from ordinary weakness of will.

The reason the test aims at capacity comes from what punishment presupposes. Criminal sanctions work by deterring people who can respond to the threat, and they are justified as blame for people who could have chosen otherwise. A person whose illness truly removes the ability to hold himself back cannot be steered by the threat of punishment, and blaming him for failing imposes punishment without fault, which is the concern Judge Rubin pressed in his dissent in United States v. Lyons.²⁵ The same reasoning supplies the limit. A person who could have stopped and did not is exactly the person the threat of punishment is built to reach, and some argue that sanctions matter most when the pull toward the crime is strong but not totally overpowering.²⁶ A rule that excused every strong desire would loosen the law's hold on the very people it most needs to deter, so the test asks whether capacity was lost, and the intensity of the desire is only evidence on that question.

Because the question is about capacity, courts usually frame it that way and seldom use the phrase irresistible impulse itself. The jury is ordinarily told to ask about the defendant's inability to resist or to control his acts, and it is not told that the act must have been sudden or that the impulse must have been totally irresistible; the instruction speaks of the capacity for self-control or free choice.²⁷ The Parsons formulation discussed earlier asked whether the disease had so far destroyed the power to avoid the act that free agency was gone.²⁸ A charge reproduced in Fisher v. United States put it as a disease that so far destroyed the will that the defendant's actions were beyond its control, which left him not responsible even though he understood the acts and knew they were wrong.²⁹ The Supreme Court's summary of the volitional incapacity test in Clark v. Arizona is shorter still: it asks whether a person was so lacking in volition because of a mental defect or illness that he could not have controlled his actions.³⁰ How complete the loss must be to qualify is a further dispute that a later section takes up.

The second scene matches the failing pole drawn in Washington v. United States, a defendant with the same knowledge and the same urge as to whom the evidence showed an ability to refrain.³² Proof of which side a defendant falls on is the subject of a later section, since psychiatrists often find it hard to tell an impulse that was not resisted from one that could not have been resisted.³³ A long record of bad conduct does not supply the missing capacity either. A dissent in Kansas v. Hendricks, written in the civil-commitment setting, distinguished an abnormality that consists simply of a long course of antisocial behavior from one that includes a specific, serious, and highly unusual inability to control one's actions, and it observed that the law has traditionally treated the second kind as akin to insanity for purposes of confinement.³⁴

On an exam, a fact pattern that depicts outlandish or impulsive-looking behavior does not establish insanity by itself, and the behavior has to be matched against the elements of the jurisdiction's test. For this element, look for facts showing that the defendant understood the wrongfulness of the act, struggled against the urge, was overpowered, and has expert or treatment evidence tying that powerlessness to the illness. Students lose this issue when they treat an impulse that was not resisted as if it were one that could not be resisted, because the whole difficulty of proof lies in telling the two apart.

Sudden or Developed Impulse

The inability to control conduct that the traditional test requires does not have to arrive suddenly. Under the rule as usually stated, the lack of control may develop or exist over time, so a defendant whose illness wore away the capacity for self-restraint across many months stands in the same position as one overtaken in a moment.³⁶ The question of timing arises at all because of the label. In ordinary usage an impulse is a sudden urge or inclination that prompts an unplanned action.³⁷ A test named for impulses therefore invites a listener to picture a flash of compulsion followed at once by the act. The test itself asks about capacity for self-control, and capacity is a standing condition of the defendant that can persist for months. A sudden urge is only one way that condition can show itself, and a rule demanding suddenness would tie the excuse to the appearance of the act instead of to whether illness had disabled the will.

Courts that rejected the control test as a free-standing rule built their objection on this word. In Durham v. United States the D.C. Circuit reasoned that "irresistible impulse" misleads because it suggests only sudden, momentary inclinations, while many mental illnesses produce prolonged alterations in mood and motivation that lead to carefully planned acts which are nonetheless the product of disease.³⁸ The court held the test inadequate for failing to recognize illnesses marked by brooding and reflection, and it replaced both the right-wrong test and the irresistible impulse test with a rule asking whether the unlawful act was the product of mental disease or defect, a rule taken up later in this chapter.³⁹ In United States v. Freeman the Second Circuit made the same objection: the label carries the misleading implication that an impulsive crime must have been committed in a sudden and explosive fit, which excludes the far more numerous crimes committed after excessive brooding and melancholy by a person unable to resist sustained psychic compulsion, and a criminal act may be coolly and carefully prepared yet still the result of a diseased mind.⁴⁰ The majority in Wade v. United States, a Ninth Circuit decision, likewise observed that the test's language impliedly refers to sudden, explosive, fit-like actions, when the acts of a person unable to control his conduct more often follow excessive brooding and melancholy.⁴¹

The criticism has a practical edge because it concerns what jurors will take from the word. Jurors who hear a test built on impulse may read it as confined to impulsive behavior and as closing the door on planned behavior,⁴² and where the word is stressed, an insane person who brooded over a contemplated crime and carried it out because of his disease would be left without a defense.⁴³ The usual answer is that the label overstates what the test requires in practice. Courts have generally framed the rule as the defendant's inability to resist or to control his acts, the phrase irresistible impulse is very seldom used, and the jury is ordinarily given the rule in terms of capacity for self-control or free choice, without being told that the act must have been sudden.⁴⁴ Even where the word does appear in an instruction, nothing shows that jurors take it to require suddenness, and any evidence tending to show a loss of control from mental disease is freely admitted, so an expert can describe a compulsion that built over time.⁴⁵ A survey of mid-twentieth-century cases decided under the M'Naghten formulation, which says nothing about control, found that trial judges were generous in admitting evidence of loss of control and that appellate courts reversed convictions when a trial judge was too stingy.⁴⁶ Some courts also reworded the test, and as Judge Trask's dissent in Wade describes it, the Ninth Circuit restated the irresistible impulse test as "uncontrollable act," so that the volitional inquiry reaches acts that follow excessive brooding and melancholy as well as sudden, fit-like ones.⁴⁷

What the rule leaves out matters as much as what it includes. A defendant's planning does not, by itself, defeat the defense, because the rule requires no sudden onset and a long-planned act can still be the product of an illness that disabled the will.⁴⁸ The reason follows from the structure of the excuse. The defense rests on a lost capacity for self-control, and a person who cannot stop pursuing a drive may organize elaborate steps to satisfy it, so the care of the preparation says little, taken alone, about whether the person could have refrained. Planning is nonetheless evidence, and what it is evidence of is capacity. A defendant who compares the risks of being caught, picks the safest moment, and goes ahead is steering by consequences, which is the response criminal sanctions are meant to produce, and a jury may infer from that steering that the defendant could have held back. When the jury draws that inference, the defense fails for want of lost control, and the planning was simply the evidence that showed it. Drawing the line is hard even for experts, since psychiatrists often find it difficult to distinguish an impulse that was not resisted from one that could not have been resisted.⁴⁹

A parallel dispute appears in tort law, which also uses an irresistible impulse idea when a negligent defendant is said to have driven another person to self-harm. Under Restatement (Second) of Torts § 455, a defendant who negligently brings about another's insanity is liable for the person's self-inflicted harm if the insanity made it impossible to resist an impulse that deprived the person of the capacity to govern conduct in accordance with reason, because a suicide committed under those conditions is not volitional and so does not break the chain of causation.⁵¹ Applying that rule in R.D. v. W.H., the Wyoming court held that the impulse need not be sudden to count as irresistible or uncontrollable, and that a person may be found to have acted under such an impulse even after earlier attempts or planning, citing New York's decision in Fuller v. Preis among others.⁵² Most courts applying the exception have been narrower, finding an irresistible impulse only in a sudden frenzy or trance-like state and denying recovery where the decedent left a note, bought the means, or otherwise showed planning.⁵³ Those civil decisions concern liability for a suicide and do not state the criminal test, but they show the same fork: courts that equate impulse with suddenness treat planning as proof of control, and courts that follow the Wyoming approach treat planning as one fact for the factfinder to weigh.

The third step does the real work, because planning can point in either direction. The most common error is to treat planning as proof of control, and it is just as mistaken to assume that a diagnosis makes every calculated act an uncontrolled one. The fact that moves the answer is the evidence on whether the defendant could have refrained, and neither the speed nor the care of the act decides that question by itself.