Concept
Limits and Successor Tests
Irresistible Impulse Test
Limits and Successor Tests
The control test earned its place by reaching the defendant who knew the act was wrong but could not stop, and that same design drew criticism from two opposite directions. One group of critics thought it reached too few people, because it divides the mind into a faculty that knows and a faculty that wills, while modern psychiatry treats the mind as an integrated personality in which reason is only one element.1 Another group thought it reached too many, because a strong but resistible urge is the kind of impulse that the threat of punishment may still deter.2 The first line of criticism produced two replacements, the product-of-disease test and the substantial-capacity test, and the second helps explain why courts later pulled back from volitional tests altogether. The discussion below takes the narrowness criticism first and then each replacement, keeping to the reasoning that moved courts past the traditional rule, because later readings teach each successor in full.
The Total-Incapacity Problem
The word irresistible suggests that only a complete loss of control counts, and that suggestion supports the second major criticism of the traditional test, after the complaint about suddenness addressed earlier in this chapter. Under the usual reading, M'Naghten asks whether the defendant wholly failed to know that the act was wrong, and the control supplement asks whether the defendant wholly failed to resist it, so each test demands the extreme form of one faculty's collapse. The Model Penal Code formulation reflects a judgment that no test is workable when it calls for complete impairment of the ability to know or to control, because clinical experience showed a graded scale and a disorder could be severe without being total.3 The practical consequence is a test that few defendants can satisfy. Conditions in which the inability to resist is absolute are rare, and a person in such a condition would probably be out of touch with reality enough to meet M'Naghten anyway, so a control test requiring total incapacity would add almost nothing to the cognitive rule it was meant to supplement.4
The harshest version of the test asks whether the defendant would have committed the crime even with a police officer standing at the elbow, and the defense fails unless the answer is yes.5 That version measures the wrong thing, because a person may lack substantial capacity to control conduct in ordinary circumstances and still manage to summon restraint under so extraordinary a threat.6
How harshly the label operates depends on how a jurisdiction defines the test. Juries are usually instructed in terms of loss of self-control or free choice, evidence of lost control is freely admitted, and experts may explain what they mean by it, so jurors are unlikely to conclude that an absolute inability to resist is required.7 The control test is therefore not one test but many: at its strictest it differs little from M'Naghten, and at its most expansive it approaches the product test discussed next.8 A broad definition meets the opposite objection, because it may excuse a defendant who could have been deterred, and psychiatrists often find it hard to distinguish an impulse that was not resisted from one that could not be resisted.9 Students lose this issue when they assume every jurisdiction demands a total loss of control, so the first step on an exam stem is to see how the jurisdiction defines the test.
Durham's Product Test
In Durham v. United States the D.C. Circuit replaced both the right-wrong test and the irresistible impulse test with a rule that a defendant is not criminally responsible if the act was the product of mental disease or defect.11 The court was reviewing the combined rule that many jurisdictions used, since in the District of Columbia the right-wrong test had been the exclusive standard until 1929, when the irresistible impulse test was added as a supplement.12 It found each half inadequate. The right-wrong test rests on a single symptom and ignores conduct that is the product of altered moods, brooding, or other noncognitive disturbances.13 The irresistible impulse test suggests only sudden, momentary inclinations, although many illnesses produce prolonged alterations in mood and motivation that lead to carefully planned acts that are nonetheless the product of disease.14
The product test trades a list of symptoms for a causal question. The jury asks whether the accused acted because of a mental disorder, psychiatrists inform the jury of the character of the disease, and evidence that the accused did not know right from wrong, acted under an irresistible impulse, or lost the power of will remains relevant without defining the defense.15 The D.C. Circuit later explained, in an opinion quoted in Blocker v. United States, that Durham did not bar the older criteria, so a jury could still consider whether the accused understood the nature of the act and whether the conduct resulted from a failure, because of mental disease, to control it.16 The design keeps the jury's traditional role of applying inherited ideas of moral responsibility while giving it a broader picture of mental life than any single symptom supplies.17
The product test asks about causation, while the control test asks about capacity, so evidence that the disease produced the act matters even when the defendant kept some knowledge and some control. Because the test names no particular incapacity, the line between a disorder that produced the act and one that merely accompanied it falls to the jury. Several federal circuits that left M'Naghten chose the A.L.I. formulation, discussed next, in place of the product test.18
The Substantial-Capacity Test
The A.L.I. formulation in Model Penal Code § 4.01 answers the narrowness criticism while keeping a defined legal standard. A person is not responsible for criminal conduct if, as a result of mental disease or defect, the person lacks substantial capacity either to appreciate the criminality (wrongfulness) of the conduct or to conform the conduct to the requirements of law.20 A second subsection provides that the terms mental disease or defect do not include an abnormality manifested only by repeated criminal or otherwise antisocial conduct.21
Each feature of the formula answers a defect in the older tests. Substantial capacity means a capacity of appreciable magnitude when measured against people generally, and it excludes the trivial remnant that survives even the most severe afflictions, so the test no longer demands total impairment.22 The commentary reasons that a question of degree is administrable because the law routinely asks juries to decide such questions, as it does with recklessness and negligence.23 The volitional prong also sits inside the definition beside the cognitive prong, joined by either/or, so a defendant who appreciates wrongfulness but lacks substantial capacity to conform qualifies without any supplement. The exclusion for repeated antisocial conduct answers a worry raised about control rules, that almost anyone with an antisocial personality might claim an inability to control impulses.24
The volitional prong then came under pressure of its own. In Blake v. United States the Fifth Circuit, sitting en banc in 1969, adopted this formulation and noted that five other circuits had already done so.25 In United States v. Lyons the same court, again en banc, withdrew the volitional prong in 1984, reasoning that it did not comport with current medical and scientific knowledge.26 The majority relied on the view that most psychiatrists lack an accurate scientific basis for measuring self-control, that the line between an impulse that could not be resisted and one that was not resisted is no sharper than the line between twilight and dusk, and that volitional testimony invites fabrication and jury confusion.27 Judge Rubin's dissent answered that a verdict of guilt is a moral judgment of blameworthiness, so punishing a person for conduct the person had no ability to avoid imposes punishment without fault, and that the line between cognitive and volitional illness rests on faulty premises.28 The volitional approach remains controversial and a minority position, and public reaction to John Hinckley's acquittal by reason of insanity after the 1981 attempt on President Reagan's life led many legislatures and courts to adopt stricter approaches.29 Whether a jurisdiction asks about both prongs, as the A.L.I. formulation does, or only about appreciation, is therefore a matter of its own law, and the next readings take up the product test and the substantial-capacity formulation in full.
The Total-Incapacity Problem
The word irresistible suggests that only a complete collapse of self-control qualifies, and that suggestion is the basis of the second major criticism of the control test, which sits beside the complaint about suddenness taken up earlier. Under the usual interpretation, M'Naghten asks whether the defendant's capacity to know was completely impaired, and the control supplement asks whether the capacity for self-control was completely impaired, so each test is satisfied only at the far end of its own faculty.31 The definition drawn from Davis v. United States, which the Fifth Circuit long followed, shows how absolute the wording could be. It reached a defendant whose will, "the governing power of his mind," had been "so completely destroyed" that his actions were beyond his control.32 A standard worded this way turns the trial into a search for the last remnant of capacity, and that search is where the difficulty lies.
The standard is hard to meet because of what clinicians actually observe. Psychiatrists told the drafters of the Model Penal Code that a disorder could be extreme and still not total, and that experience showed a graded scale with marks along the way.33 A person with schizophrenia, for example, may be badly cut off from reality and yet respond to an order from someone in authority inside a hospital, which means some capacity to conform to a norm survives even in grave illness.34 A rule that demands zero capacity therefore excludes the defendants whose control was gravely reduced and reaches only the rare few whose control vanished entirely. Those few probably qualify under the cognitive test already, because a person with an absolute inability to resist would likely be out of touch with reality, so a control test read this way adds almost nothing to the rule it was meant to supplement.35
Proof creates a second difficulty. An expert who testifies about the defendant's condition is reconstructing it after the fact, from history, examination, and reported behavior, at a time the expert did not observe. An honest witness in that position can say that capacity was gravely impaired, but a statement that none remained is a claim of complete absence that the evidence rarely supports. The Model Penal Code's drafters made the same point when they explained that substantial capacity is about all that candid witnesses, inferring a situation they did not observe, can ever confidently say, even when they know the disorder was extreme.36 Under a total-incapacity standard, then, the honest expert's best testimony falls short of the legal requirement.
The deeper objection is that the all-or-nothing line sorts defendants by the wrong fact. Many cases of advanced mental disorder involve rudimentary capacities that remain intact, and those cases are plainly inappropriate occasions for criminal punishment.37 Punishment rests on blame, and blame follows the capacity a person had to act otherwise, so a trace of remaining capacity, treated as if it were full capacity, converts a gravely impaired defendant into a fully culpable one. The Supreme Court has accepted the clinical premise in a neighboring setting. In Kansas v. Crane, which concerned the civil commitment of a dangerous sexual offender rather than the insanity defense, the Court held that no showing of total or complete lack of control is required, observed that most severely ill people retain some ability to control their behavior, and demanded proof of serious difficulty in controlling behavior.38 The Court also relied on the psychiatric profession's view that the line between an irresistible impulse and an impulse not resisted is probably no sharper than the line between twilight and dusk.39 Because the setting there is civil commitment, Crane leaves the criminal test untouched, and each jurisdiction's control test still has to be read on its own terms.
In Blake v. United States, the defendant offered psychiatric testimony that he suffered from schizophrenia with psychotic episodes, and that during such an episode his actions would not be subject to his will.40 The Fifth Circuit nevertheless found that the record did not show complete mental disorientation, because he was not unconscious, he could distinguish right from wrong, and his will was not completely destroyed in the terms of the Davis definition.41 Under that charge he could not prevail, although a charge asking only about substantial lack of capacity might have let him.42 His evidence that the illness impaired his control, even severely, could not carry the defense, because the standard asked whether the will was destroyed, and impairment short of destruction did not count.
The first defendant in that pair sits at the rare extreme the traditional test was built to reach, so the criticism has no bite for such a defendant.43 The criticism concerns the second defendant and the large middle ground that the first leaves behind.
How much the criticism matters depends on how a jurisdiction defines the test. Courts usually state the rule as inability to control conduct, juries are instructed in terms of capacity for self-control or free choice, and evidence of lost control is freely admitted, so the label may overstate the burden in practice.44 At the strict end sit formulations that ask whether the defendant would have committed the offense even with a police officer at the elbow. That framing tests capacity in an artificial, closely supervised setting, so a defendant who lacked substantial control in ordinary life may still muster restraint under so extreme a prompt, and the defense fails for a reason unconnected to how the defendant actually lived and acted.45 These two ends explain why the control test is better described as many tests than one: at its strictest it differs little from M'Naghten, and at its broadest it approaches the product-of-disease approach.46 Students lose this issue when they treat partial loss of control as automatically sufficient, or as automatically fatal, without first checking how the jurisdiction defines the test.47 The substantial-capacity formulation, taken up below, is the response that removes the total-incapacity demand directly.
Durham's Product Test
The two criticisms of the control test taken up above, its suggestion of total incapacity and its association with sudden impulse, led some courts to ask whether the law should isolate any single faculty of the mind. In Durham v. United States the D.C. Circuit reviewed its own rule, which paired the right-wrong test with the irresistible impulse supplement, and found both inadequate.49 The right-wrong test, the court reasoned, rests on one symptom, cognition, while modern psychiatry treats the mind as an integrated personality of which reason is only one element, so the test ignores conduct that grows out of altered moods, brooding, or other noncognitive disturbances.50 The irresistible impulse supplement recognizes that disease can strip a person of the will power to resist, but its wording suggests only sudden and momentary inclinations, whereas many illnesses produce prolonged changes in mood and motivation that lead to carefully planned acts that are still the product of disease.51 Each older test named one symptom and treated its absence as the end of the inquiry, and that shared feature is why the court concluded that a broader test was needed.
The replacement rule asks whether the accused's unlawful act was the product of mental disease or mental defect.52 The idea came from New Hampshire, where State v. Pike had pioneered a product approach in 1869, and the D.C. court described its own rule as not unlike the one New Hampshire had followed.53 The jury's task has two parts. A finding that the accused suffered from a mental disease or defect does not finish the inquiry, because the jury must also consider whether the abnormality caused the act, and an accused whose act was not the product of the condition may be convicted.54 The jury therefore asks whether the accused acted because of a mental disorder, without regard to whether any particular symptom was displayed, since medical science recognizes that symptoms do not necessarily accompany even serious disorders.55
The older criteria did not vanish from the courtroom under this rule. They changed status. The jury's inquiry may still include whether the accused did not know right from wrong, acted under an irresistible impulse, or had lost the power of the will.56 The D.C. Circuit later made the new status explicit in McDonald v. United States, holding that capacity to distinguish right from wrong and ability to refrain from wrongdoing may be put to the jury, but only as factors bearing on whether the act was the product of the disease, and not as independently controlling or alternative tests.57 Evidence of lost control thus enters the product test as one way of proving causation. Students lose this issue when they treat the product test as a version of the control test, because the question put to the jury is causation, and loss of control is only one way of proving it.58
Two reasons stand behind the shift from capacity to cause. The first is informational: a test that asks whether the disorder produced the act lets the jury see the accused's mental life as a whole, so testimony about mood, brooding, and compulsion all reaches the jury on one question instead of being sorted into separate symptom tests.59 The second concerns the experts. Under M'Naghten, psychiatrists sympathetic to a defendant who was not in fairness responsible felt pressed to testify that the defendant did not know the act was wrong, even when the defendant's own words showed otherwise, and the product test ended much of that stretching by letting experts describe the illness candidly.60 The ultimate question also stays with the jury, which keeps its traditional role of applying inherited ideas of moral responsibility on the principle that the community does not permit punishment where it cannot impose blame.61
Because the question is causal, the test reaches defendants whom the older pair left out. On the reasoning in Durham, a defendant whose illness produced prolonged brooding and a carefully planned offense can be acquitted even though the defendant knew right from wrong and no sudden impulse appeared, since the jury asks only whether the illness caused the act.62 The dissent in McDonald objected to exactly this breadth, arguing that a person who can distinguish right from wrong and can refrain from the act should not be excused because of some other mental infirmity.63 The breadth also explains why the control test, at its most expansive, approximates the product approach, while at its strictest it differs little from M'Naghten.64 States that adopted a product test did so out of the conviction that M'Naghten was too restrictive, the same motive that produced the volitional supplement.65
That outcome follows the instruction the Durham court contemplated, under which a jury that believes the accused had a mental condition, but believes beyond a reasonable doubt that the act was not its product, may convict.66 The diagnosis supplies the first element only, and the causal link supplies the second.
The product test drew criticism of its own, which is why it appears here as one response to the narrowness of the older tests. The Second Circuit in United States v. Freeman praised Durham for dropping the right-wrong dichotomy and freeing expert testimony, yet observed that the demand for a product of disease raised near-impossible problems of causation resembling those of the older tests.67 It added that the test gave the fact-finder no standard for measuring the accused's competency, so that psychiatrists' labels in effect decided cases, as when a hospital staff conference in 1957 reclassified psychopathic personality as a mental disease and defendants once treated as responsible became candidates for acquittal.68 Another reading teaches the product rule and those difficulties in full. Here Durham matters as an answer that changes the question from capacity to cause, while the A.L.I. formulation, taken up next, keeps the capacity question and loosens it.
The Substantial-Capacity Test
The second response to the narrowness of the control test keeps the capacity question that Durham abandoned and loosens it. About a year after Durham, the American Law Institute's Model Penal Code project passed over a product test, even though its own psychiatric advisory committee had recommended one, and approved instead a modernized combination of M'Naghten and the irresistible impulse test.70 The drafters rejected Durham because of the ambiguity of the word “product,” and they kept cognition and volition as the two impairments that matter when the law tries to exclude people who cannot be deterred from penal sanctions.71 The result is the substantial-capacity test of Model Penal Code § 4.01(1), under which a person is not responsible for criminal conduct if, “as a result of mental disease or defect he lacks substantial capacity either to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the requirements of law.”72
The two prongs are alternatives, so a defendant who understood that the act was wrong can still prevail by showing the control incapacity, and satisfaction of either prong is enough to excuse.73 Both prongs sit beneath the opening words “as a result of,” which build in a causal requirement: exculpation follows only if the incapacity results from the disease, so the disorder of a kleptomaniac, presumably, produces no incapacity to conform to the law against rape.74 Three departures from the traditional control test then matter, and each has its own reason.
The first departure is the word substantial. Under the usual reading of M'Naghten and the irresistible impulse test, a complete impairment is necessary, whereas the A.L.I. test requires only a lack of substantial capacity.75 The reason, developed earlier in this chapter, is that clinical experience reveals a graded scale of impairment rather than an on-off switch, so a test confined to total incapacity makes the inquiry unreal for the expert who must testify.76 The drafters defined the word as a capacity of some appreciable magnitude when measured by the standard of humanity in general, in contrast to the small and trivial remnant that survives even the most severe afflictions of the mind.77 The Second Circuit in United States v. Freeman drew the practical consequence: incapacity of any degree is insufficient to avoid responsibility, yet total incapacity is also unnecessary.78 The word must still separate incapacity from mere indisposition, because a standard addressed to volition has to distinguish a defendant who could not resist from one who did not.79
In Blake v. United States, the Fifth Circuit saw the difference on the record before it. The evidence, if believed, showed a severe mental disease that the jury might have found impaired the defendant's control, but his will was not completely destroyed, so he could not prevail under a charge that spoke in absolute terms, although he might have prevailed under a substantial-lack-of-capacity charge.80 The court responded by adopting the Model Penal Code standard.81 A defendant whose capacity to conform is gravely reduced but not gone therefore satisfies the substantial-capacity test and fails the strict total-incapacity reading.
The second departure is the verb conform. The irresistible impulse label suggests a sudden, explosive fit, yet the acts of people who cannot control their conduct more often follow excessive brooding and melancholy.82 The A.L.I. wording asks about capacity to conform conduct to law, and so it avoids the implication that lost volitional capacity can appear only in sudden or spontaneous acts as distinguished from acts accompanied by reflection.83 An act can be coolly and carefully prepared and still be the result of a diseased mind, which is why a planned offense is not disqualified from the control prong.84 The capacity is also measured at the time of the conduct and in the unstructured conditions of ordinary life, so evidence that the defendant would have held back with an attendant or police officer at the elbow does not answer the question.85
The third departure is paragraph (2) of the test, which provides that the terms “mental disease or defect” do not include “an abnormality manifested only by repeated criminal or otherwise anti-social conduct.”86 The drafters aimed the clause at the so-called psychopathic personality, a condition that differs from the normal person only in degree rather than in kind and whose status as a disease is disputed.87 The incentive problem becomes clear once it is named. If repeated offending alone could establish the disease, each additional offense would strengthen the defense, and the behavior the defense is meant to excuse would be the only proof offered to excuse it. The clause therefore excludes a condition manifested only by the very behavior that must, by hypothesis, result from disease before irresponsibility is established, and it spares courts a contest over terminology.88 Students lose this issue when they let repeated criminal conduct standing alone supply the mental-disease element, because the clause removes exactly that showing.
The formulation spread widely. The core rule was adopted, with variations, by all but one of the federal courts of appeals, and a significant minority of states accepted it, usually by statute.89 A later count found fourteen jurisdictions using an amalgam of the volitional incapacity test and some variant of the moral incapacity test, with either one enough to excuse.90 The volitional half proved the more fragile part. The Fifth Circuit, which had adopted the test in Blake, withdrew its recognition of the volitional prong in United States v. Lyons after finding that a majority of psychiatrists believed they lacked accurate scientific bases for measuring capacity for self-control.91 In 1984 Congress went further and repudiated the A.L.I. approach for federal prosecutions in favor of a M'Naghten-style statute.92 A separate reading teaches the Model Penal Code test in full, including how courts have varied its wording; the point here is where it departs from the control test it grew out of.