Concept
The Durham Rule Announced
Product of Mental Disease or Defect
The Durham Rule Announced
Having found both older tests inadequate, the District of Columbia court replaced them with a single causal question. In Durham v. United States it announced that "an accused is not criminally responsible if his unlawful act was the product of mental disease or mental defect."1 The formulation has two operative parts, a diseased or defective mental condition and a product relationship between that condition and the act, so a case that shows the condition without the relationship falls outside the rule. Applied to a verdict, the rule requires an acquittal when the charged offense resulted from the condition, and an offense qualifies when the defendant would not have committed it but for the disease or defect.2 That but-for reading was a later gloss, because the original opinion made no effort to explain what "product" means, a gap that returns in the criticisms near the end of this chapter.3
The wording does work that a list of symptoms could not do. The court tied the rule to the legal and moral tradition that requires responsibility where acts are committed willfully and with evil intent and withholds blame where acts stem from mental disease or defect, and the phrase "product of" carries that tradition straight into the test.4 If the illness produced the act, the act did not come from the willful wrongdoing that justifies punishment, so excuse follows the same causal line that blame would otherwise have followed. Because the question is why the accused acted rather than which symptom the accused displayed, psychiatrists can describe whatever the disorder looked like in the particular patient, and the jury asks whether the accused acted because of a mental disorder, knowing that medical science does not expect any single symptom to accompany every serious disorder.5 The jury still applies inherited ideas of moral responsibility, so the rule changes what the jury learns about mental life and leaves with the jury the decision whether blame attaches.6 The design was meant to maximize the contribution of psychiatrically trained witnesses and to widen the group of mentally ill defendants who could use the insanity defense.7
The court did not write on a blank slate. It described its new rule as not unlike the one the New Hampshire court had followed since 1870.8 The New Hampshire decisions, State v. Pike and State v. Jones, asked whether the crime "was the offspring or product of mental disease."9 After New Hampshire broke with M'Naghten, no other American jurisdiction adopted that rule or gave it serious consideration for eighty-three years, until the District of Columbia did so in 1954.10 Critics of the New Hampshire rule complained that its ambiguity left juries with too little guidance on the critical question of responsibility, and Durham inherited that exposure because it too left "product" unexplained and defined disease and defect only in relation to each other.11
The court also had to say by what authority it could change the test. In the District of Columbia the formulation of tests of criminal responsibility is entrusted to the courts, and the court invoked its inherent power to make the change prospectively, so that the new rule governed the retrial in Durham's own case and cases to come.12 The choice of authority follows from the history just reviewed. The same court had approved the right-wrong test in 1882 and the irresistible impulse supplement in 1929, so both older tests were judicial creations and the court could replace them without waiting for a legislature.13 A test made by judges can also be unmade by them, and the D.C. court later overruled Durham in favor of a modified Model Penal Code test, a development taken up in the closing sections of this chapter.14
Announcing the rule did not entitle every defendant who mentions mental trouble to an instruction on it. The court set a threshold: whenever there is "some evidence" that the accused suffered from a diseased or defective mental condition at the time of the unlawful act, the trial court must provide the jury with guides for determining criminal responsibility.16 The threshold is low because of who decides. Under the product test, whether the disease existed and whether it produced the act are questions of fact for the jury, so a trial judge who demanded proof of disease before instructing would be deciding the very question the rule hands to jurors.17 The trigger still has content, because it requires evidence of an actual diseased or defective condition at the time of the act, and a defendant who offers none gives the guides nothing to operate on. Since the court could not frame a single instruction suited to every case, it described the substance any instruction should convey, which is a sequence of questions the jury answers in order.
Each step places the demanding standard on the side of conviction. The jury may convict only if it holds that degree of belief about the absence of the condition or about the absence of causation, and otherwise it must acquit by reason of insanity.18 The defendant's task at the threshold is therefore to bring forward some evidence, and the instruction then requires the jury to be persuaded beyond a reasonable doubt of one of the two negatives before it may convict. Students lose this issue when they demand proof of mental disease before the jury receives the guides, since the trigger asks only for some evidence.19
The two defendants differ in one fact, whether any evidence of a diseased or defective condition exists, and that fact alone decides whether the jury receives the guides. Once the guides are given, the product question belongs to the jury. The next two subsections take up the two phrases of the rule that need their own explanation, first what counts as a mental disease or defect and then the moment at which the condition must exist, and the causal link on which the second and third steps turn receives its own section after them.
Mental Disease or Defect
The product test has two elements, and the first the jury meets is an abnormal mental condition. The causal question has nothing to work on until such a condition is shown, which is why the first step of the jury instruction asks whether the accused was suffering from a diseased or defective mental condition at the time of the act.21 On what that phrase means, the Durham court said very little. It used disease to mean a condition capable of improvement or deterioration, and defect to mean a more static condition that may be congenital, due to injury, or residual from disease.22 Nothing else in the opinion defines either word, and that silence matters for reasons developed below.23
The two definitions sort conditions by how they behave over time and say nothing about which symptoms a condition produces, which fits the court's refusal to confine the jury to arbitrarily selected symptoms.24 A disease moves: it may improve with treatment or worsen without it. A defect holds still, whether the person was born with it, acquired it through injury, or was left with it after an illness ran its course. Because the rule reaches an act produced by either kind of condition, the choice between the two labels never decides a case by itself, and the threshold asks only that some abnormal mental condition of one kind or the other exists.25 Together the two words cover conditions that change and conditions that do not, so no abnormality falls outside the element merely because of its course over time.
The sparse definition looks harmless until the product test is compared with the right-wrong test it replaced. Under M'Naghten a particular effect had to be discernible, an inability to distinguish right from wrong, so courts had little reason to press on what counted as a disease of the mind, because the required effect did the filtering. Under Durham the condition had to be nothing except a factor causing the crime, so the classification of the condition carried the weight that the required effect had carried before.26 Durham then defined disease and defect only in relation to each other, which left the question the jury meets first with almost no content beyond whether the condition could change.27
The consequence was a threshold that moved with medical labels. In Blocker v. United States the D.C. court held that a defendant with a sociopathic personality was entitled to a new trial when, after his conviction, part of the psychiatric profession changed its view and began to call sociopathy a mental disease.28 A related episode followed a 1957 staff conference at St. Elizabeth's Hospital, which reclassified psychopathic personality as a mental disease. Because that single hospital supplied most of the psychiatric witnesses in the District's courts, juries were abruptly told that defendants once considered responsible were now to be acquitted.29 The defendants' conditions had not changed; the label had. The same divergence appears whenever experts classify one person. In Atkins v. Virginia, which concerned a different legal question, a defense psychologist found the defendant mildly intellectually disabled, and the State's rebuttal expert instead diagnosed antisocial personality disorder.30
Later decisions of the District of Columbia court supplied content that the original opinion withheld, and only the outline of that development belongs here. In McDonald v. United States the court recognized that mental disease means one thing to a physician bent on treatment and something different to a court of law, and it defined the term to include any abnormal condition of the mind that substantially affects mental or emotional processes and substantially impairs behavior controls.31 That definition belongs to the test's later history. The rule as announced rests on the improvement-or-deterioration distinction alone.
Satisfying this element is only the threshold. The jury's task is not completed by a finding of mental disease or defect, because the causal relationship between the abnormality and the act must also be considered.32 Students lose this issue when they treat a diagnosis as an acquittal, since a defendant with a genuine disease or defect remains responsible for an act that did not result from it.33 The next subsection fixes the moment at which the condition must exist, and the causal link follows in its own section.
Condition at the Time of Act
The product test asks about the accused's mental condition on a single date, and that date is the moment of the unlawful act. The Durham court wrote the date into the rule: the trial court must give the jury guides for determining responsibility whenever there is some evidence that the accused suffered from a diseased or defective mental condition "at the time of the unlawful act," and the first step of the suggested instruction asks whether the accused was suffering from such a condition at that time.35 The second step carries the same date forward, because it asks whether the act was the product of the abnormality, and an act can be the product only of a condition that existed when it was done.36 Causation runs forward in time, so an illness that began after the act cannot have produced it, and an illness that had already ended could not have operated on it.
The date also tracks the purpose of the excuse. Responsibility is a judgment about a particular act, and the product test excuses an accused whose act flowed from mental disease or defect, so the condition that matters is the one that shaped the act. A defendant who was ill when acting is excused if the illness produced the act, even though treatment has restored the defendant to health by the time of trial. A defendant who was healthy when acting made the choice in full health, and a collapse after the act cannot change what kind of choice it was. A test keyed to the trial date would reverse both results, convicting the person whose illness drove the act and acquitting the person whose illness arrived too late to have any bearing on it.
Because the inquiry is pinned to the date of the act, evidence from other dates counts only as proof of the condition on that date. The Durham opinion leaves the jury free to consider testimony about symptoms and manifestations, together with other relevant evidence, on the ultimate questions of fact, so earlier or later signs of illness can serve as circumstantial evidence of the condition when the act occurred.37 Courts reason the same way outside criminal law. In Barnes v. Marshall, a will contest, the defendants argued that the testator, who suffered from manic-depressive psychosis, had normal periods between episodes and was in one when the will was executed, and the court answered that his condition at the precise time of execution was a question for the jury.38 A diagnosis on some other day therefore settles nothing by itself; it is one piece of evidence about the day that counts. Article 31 of the Rome Statute anchors its mental-disease ground for excluding responsibility to the same date, since the incapacity must affect the perpetrator when the act was committed and not afterwards.39
A different mental-health doctrine looks at a different date and has to be kept apart from the product test. Incompetency to stand trial concerns the defendant's condition during the proceedings: a defendant whose mental disease or defect leaves that defendant unable to understand the proceedings or to assist in the defense may not be tried, convicted, or sentenced while the condition lasts, and is ordinarily committed until recovery.40 The date is the present because this rule protects the fairness of the proceeding, and whether a proceeding is fair depends on what the defendant can do now. A related doctrine concerns the condition after conviction. In Ford v. Wainwright the opinion records no suggestion that Ford was incompetent when he committed the offense, at trial, or at sentencing, yet delusions developed on death row years later, and the Court held that the Constitution forbids executing the insane.41 Neither doctrine decides responsibility for the act. Their consequences differ as well: incompetency suspends the case until the defendant recovers, while a successful product defense ends in a verdict of not guilty by reason of insanity.42
The two defendants differ in a single fact, the date of the illness, and that date decides which doctrine governs. Students lose this issue when they treat an illness that arrived after the act as a defense under the product test, or when they assume that recovery by the time of trial erases a condition that existed when the act was done.43