Law & Psychiatry: Hall v. Florida: Defining Intellectual Disability in the Shadow of the Death Penalty

Defendants with severe intellectual disabilities have always been excluded from the death penalty. Under English common law, they were referred to as “idiots,” and, along with “lunatics,” they would have been considered non compos mentis and exempted from criminal punishment (4). The death penalty was still a possibility for capital defendants with lesser degrees of disability, although their impairment could be taken into account as a mitigating factor. As late as 1989, when the U.S. Supreme Court decided Penry v. Lynaugh, a majority of the justices found no constitutional basis for rejecting this approach (5).

Just 13 years later, though, the balance on the Court shifted. When Daryl Atkins challenged the death penalty for people with mental retardation as a violation of the 8th Amendment, the Court rejected the traditional approach. Citing the principle that what constitutes cruel and unusual punishment should be determined in light of “the evolving standards of decency that mark the progress of a maturing society,” the ruling in Atkins created an absolute bar to death sentences for people with mental retardation. The six justices in the majority pointed to shifts in public opinion and legislation since Penry as indicating a progressive rejection of the legitimacy of the death penalty in these cases. In addition, Justice Stevens’ opinion noted that the impairments associated with mental retardation undermined the retributive and deterrent rationales for a capital sentence, because people with mental retardation were less culpable for their behavior and less likely to be deterred by the prospect of the ultimate punishment.

However, the Court pulled back from defining the practical implications of its ruling. Atkins left open a host of important questions about the procedures for determining whether a defendant is exempt from the death penalty, including the standard of proof to be applied, whether the defendant or the state bears the burden of proof, and whether a judge or the jury makes the decision (2). Perhaps most critically, the justices declined to specify criteria for determining when a defendant is mentally retarded. “Not all people who claim to be mentally retarded will be so impaired as to fall within the range of mentally retarded offenders about whom there is a national consensus,” Justice Stevens wrote, and hence, “we leave to the States the task of developing appropriate ways to enforce the constitutional restriction upon its [sic] execution of sentences” (1).

In the wake of Atkins, in addition to a plethora of procedural approaches to determining a capital defendant’s status, states adopted a variety of standards for distinguishing defendants who lacked sufficient cognitive capacity. Most states drew on either the DSM-IV or American Association for Mental Retardation (now the American Association on Intellectual and Developmental Disabilities [AAIDD]) criteria that were cited by the Court in Atkins. Both criteria sets focused on three elements: significantly subaverage intellectual functioning, impairments in adaptive functioning, and onset before age 18 (6). But states varied the definitions. Perhaps most significantly, some states set a hard cutoff for IQ scores indicative of “significantly subaverage intellectual functioning,” refusing to consider any defendant with an IQ above 70 to have mental retardation (7).

This was the situation confronting Freddie Lee Hall, a long-term resident on Florida’s death row, who sought to prove that he was intellectually disabled and hence should have his sentence commuted to life in prison. Hall was convicted in 1979, along with an accomplice, of having kidnapped, beaten, raped, and murdered a 21-year-old pregnant newlywed, Karol Hurst. After the murder, he and his partner drove to a convenience store, where they killed a sheriff’s deputy who attempted to arrest them. Hall was sentenced to death for the murders, but in what is a familiar story in the United States, he has spent the subsequent 35 years appealing that decision. His initial death sentence was overturned because he had not been afforded an adequate opportunity to present mitigating evidence about his abusive childhood and intellectual impairment. However, after a rehearing in which this evidence was presented, he was again sentenced to death.

Hall’s series of subsequent appeals was given new life when the Supreme Court decided Atkins in 2002 and applied the ruling retroactively to all prisoners on death row. After Florida adopted new procedures consistent with Atkins, Hall filed a claim for his death sentence to be vacated on the grounds that he had mental retardation. It took another five years before Hall’s case was heard by the Florida courts. He presented testimony from family members regarding his impairment in childhood, along with results from nine different IQ tests taken over 40 years. However, the two tests with scores under 70 were excluded because raw data were not available, leaving only tests showing IQs ranging from 71 to 80. Consistent with the “bright-line” approach (absolute IQ cutoff) adopted by the Florida courts, Hall’s claim was dismissed. His subsequent appeal to the Florida Supreme Court challenging the bright-line rule was rejected on the grounds that the state’s approach was consistent with Atkins. Hall then appealed to the U.S. Supreme Court, which agreed to consider the case.

Hall’s objections to the use of an absolute IQ cutoff were based on its rejection by the mental health professions (8). IQ tests, like all measurement techniques, are imperfect, with an intrinsic degree of error. For most IQ tests, the standard error of measurement (SEM), an estimate of the inherent inaccuracy of the score, is about 2.5 IQ points. To identify the range within which a person’s actual IQ falls with 95% certainty, clinicians add two SEMs to either side of the score. Hence, DSM-5 notes, “Individuals with intellectual disability have scores of approximately two standard deviations or more below the population mean, including a margin for measurement error (generally +5 points).” Thus a score of 71, which was Hall’s lowest score on the tests that were entered into evidence, is best understood as a 95% probability that his actual IQ is between roughly 66 and 76.

The amicus brief filed in support of Hall’s arguments by the American Psychological Association and the American Psychiatric Association (9), with the support of other mental health groups, noted an additional reason to reject Florida’s approach. Modern concepts of intellectual disability recognize that assessments of intellectual and adaptive functioning are not disjunctive inquiries but should be conducted jointly. Especially for people whose IQ scores cluster just above or below 70, knowledge of their adaptive functioning may be crucial to determining whether they qualify for a diagnosis of intellectual disability. Just as someone with IQ scores above 70 but poor adaptive functioning should be considered intellectually disabled, a person with scores below 70 but good functioning might not meet diagnostic criteria. In the words of DSM-5, “a person with an IQ score above 70 may have such severe adaptive behavior problems in social judgment, social understanding, and other areas of adaptive functioning that the person’s actual functioning is comparable to that of individuals with a lower IQ score. Thus, clinical judgment is needed in interpreting the results of IQ tests.”

On its face, Hall’s argument would seem convincing. The professions involved in the diagnosis of intellectual disability had rejected the approach embodied in Florida law. If the Supreme Court meant what it said in Atkins—that defendants who are intellectually disabled cannot be sentenced to death—it would seem reasonable to ask the clinical professions how to determine who falls into that category. However, a number of considerations potentially weighed against that approach.

It might be, as the Atkins court had suggested, that definitions of intellectual disability that were designed to meet clinical needs are not appropriate for legal purposes. The DSM-5 criteria, for example, might identify a group of people who are in need of clinical or supportive services but who should still be considered fully culpable for their criminal behavior. Hence, unless it could be shown that the dividing line is the same for clinical and legal purposes, states might legitimately choose to define intellectual disability differently than the clinical professions. In addition, considerations of judicial economy often lead judges to shy away from establishing elaborate rules in complex areas of endeavor such as clinical diagnosis. One way to avoid this enterprise to which the justices of the Supreme Court are particularly inclined is to leave it to the states, which has the added advantage of acknowledging state prerogatives under the principles of federalism. Thus it was not a foregone conclusion that the Court would side with Hall in his appeal.

Oral arguments in the case did not start out in a particularly encouraging fashion for Hall (10). The justices appeared to be confused by the statistical arguments and spent much of the first half of the session trying to get an equally bewildered attorney for Hall—a former Solicitor General of the United States—to explain it to them. But they were no less hard on the lawyer representing the state of Florida, probing the limits of a state’s powers to define clinical conditions by its own criteria. A particularly telling moment came when Justice Kagan asked, “Could the state change its statute to say ‘we’re now using a[n IQ] threshold of 60’?” Florida’s attorney, consistent with his argument that Atkins had left these matters to the states, replied that it could, leaving several justices concerned that endorsing states’ powers to define intellectual disability could completely undermine the holding in Atkins.

In the end, the Court resolved its dilemma by relying on the consensus approach of the mental health professions, repeatedly citing the brief of the two APAs and DSM-5. Justice Kennedy, the Court’s swing vote, wrote the opinion for the 5-4 majority. Florida’s “rigid” bright-line rule, he said, “creates an unacceptable risk that persons with intellectual disability will be executed, and thus is unconstitutional.” Since, “[s]ociety relies upon medical and professional expertise to define and explain how to diagnose the mental condition at issue,” it is reasonable here to ask how the relevant professions approach the question. He was careful to say that the Court was exercising its own discretion and not merely relying on the judgment of the medical and psychological communities, and he cited the large number of states with more flexible approaches to the error inherent in IQ scores. Ultimately, he wrote, “This Court agrees with the medical experts that when a defendant’s IQ test score falls within the test’s acknowledged and inherent margin of error, the defendant must be able to present additional evidence of intellectual disability, including testimony regarding adaptive deficits.”

The Supreme Court has had an ambivalent relationship with psychiatry and the other mental health professions. At times, the justices have disparaged the reliability of psychiatric diagnoses (11), accepted diagnostic categories created de novo by the states (12), and rejected psychiatry’s efforts to define the limits of its own powers (13). In Hall, however, the Court took a much more positive view of psychiatric expertise. Just how much reliance the majority placed on psychiatric approaches is evident from the complaint in Justice Alito’s dissent that in its holding “the Court strikes down a state law based on the evolving standards of professional societies, most notably the American Psychiatric Association.” There is, of course, no guarantee that the Court will be similarly deferential to psychiatric views in the future, especially on issues—such as sex offender commitment and prediction of violence risk—where public safety concerns come into play.

The decision in Hall, however, with the deference that it pays to professional expertise, certainly indicates a degree of receptivity of a majority of the justices to input from the mental health professions on matters squarely within the scope of their knowledge. That’s a good thing. And the decision is a good one for Freddie Lee Hall, too, who gets one more chance to prove to the Florida courts that he is intellectually disabled and should not be executed for the crimes he committed 36 years ago.

The author reports no competing interests.