The US Supreme Court will hear Hamm v. Smith on Wednesday, December 10, a death penalty case testing how courts should treat multiple IQ scores when deciding whether a death-row prisoner is intellectually disabled and therefore ineligible for execution under the Eighth Amendment.
Joseph Clifton Smith was sentenced to death in Alabama for a 1997 robbery-murder. A federal district court later concluded, after a full Atkins hearing, that Smith is intellectually disabled and vacated his death sentence. The Eleventh Circuit affirmed. Alabama has now persuaded the Supreme Court to review how the lower courts evaluated Smith’s IQ evidence and what to do when a defendant has several IQ scores clustered just above 70, all within the tests’ margin of error.
What is the law at issue?
At the center of the case is the Eighth Amendment’s ban on cruel and unusual punishments as interpreted in Atkins v. Virginia. In Atkins, the Court held that executing people with intellectual disability is unconstitutional, but left to the states “the task of developing appropriate ways” to enforce that rule.
The Alabama Supreme Court described intellectual disability using a three-part framework: significantly subaverage intellectual functioning, significant deficits in adaptive behavior, and onset of those limitations during the developmental period. Alabama adopted that basic structure and, for the first prong, drew a firm numerical line: a defendant is considered intellectually disabled only if he proves an IQ of 70 or below, plus substantial adaptive deficits and onset before age 18.
The US Supreme Court has already pushed back against rigid number-driven rules. In Hall v. Florida, the Court invalidated a Florida statute that treated any IQ score above 70 as automatically disqualifying, emphasizing that “intellectual disability is a condition, not a number” and that courts must account for the standard error of measurement and consider adaptive functioning when a score falls near the cutoff. In Moore v. Texas, the Court went further, insisting that states align their practices with contemporary clinical standards and not rely on lay stereotypes.
What Hall and Moore did not do was explain how to evaluate multiple IQ scores over time. In the 2024 per curiam opinion in Smith’s case, the Court cited Hall describing the issue as “a complicated endeavor” and acknowledged that it had never specified how courts should approach it.
The factual dispute: Smith’s claim of intellectual disability
The contested question is whether Smith is intellectually disabled. Over the course of his life, he has taken multiple IQ tests. As an adult he received full-scale scores of 72, 74, 74, 75, and 78; as a youth, his scores were also in the mid-70s. At a federal evidentiary hearing, experts testified about the standard error of measurement, explaining that a score of 72 on the Wechsler test, for example, is best understood as a range, roughly 69 to 75, rather than a single point. The district court concluded that at least one of Smith’s scores plausibly reflects a true IQ below 70, and then turned to extensive evidence of adaptive-functioning deficits: special education placement, repeated grades, poor literacy, social vulnerability, and limited independent living skills, all dating back to childhood.
What is the procedural background of the case?
Smith was convicted and sentenced to death in Alabama state court, and his conviction was affirmed on direct appeal in 2000. After Atkins was decided, Smith raised an intellectual-disability claim in state post-conviction proceedings. The state courts rejected it, reasoning that his IQ scores, which were consistently above 70, disqualified him under Alabama’s definition.
Smith then turned to federal habeas corpus. The federal district court held a substantial evidentiary hearing, received expert testimony, and considered seven IQ scores along with school records, family testimony, and clinical evaluations. The court treated 70 as the ceiling for “significantly subaverage intellectual functioning” under Alabama law but concluded that, once the standard error of measurement (SEM) was taken into account, Smith’s lowest score of 72 could reflect a true IQ of 69. That finding opened the door to a full Atkins analysis. The court then found significant adaptive deficits with onset in childhood and vacated the death sentence.
In 2023, the Eleventh Circuit affirmed, holding that the district court had not clearly erred in its factual findings and that its analysis was consistent with Atkins, Hall, and Moore.
Alabama sought review in the Supreme Court. In November 2024, the justices issued a per curiam opinion granting the petition and vacated the Eleventh Circuit’s judgment. The Eleventh Circuit’s opinion, the Court said, could be read in two very different ways, either as giving conclusive weight to the lower end of the SEM range for Smith’s lowest score, or as endorsing a more holistic approach to multiple scores and expert evidence.
On remand, the Eleventh Circuit issued a short opinion making clear that it meant the latter. It explicitly “rejected” any per se rule based on the lowest endpoint of the lowest score’s range and reaffirmed that it had looked at all of Smith’s scores, his adaptive deficits, and the expert testimony together. The court again upheld the district court’s intellectual-disability finding.
Alabama returned to the Supreme Court. On June 6, 2025, the Court granted certiorari a second time, this time limiting review to a single, tightly framed question.
What specific legal issue is the Supreme Court being asked to decide?
Whether and how courts may consider the cumulative effect of multiple IQ scores in assessing an Atkins claim. The justices are not being asked to overrule Atkins or Hall. Instead, they are being asked to clarify how those decisions constrain state and federal courts when they confront multiple IQ scores in the borderline range.
What is Smith’s position?
Smith defends the rulings below by leaning heavily on the Supreme Court’s own language in Hall and Moore and on the record developed in the district court.
On the law, his position is that once a valid, properly administered IQ test yields a score whose confidence interval includes a value below 70, courts are required to move beyond the raw number and consider evidence of adaptive deficits. States, in his view, may not use either a strict IQ cutoff or a framework that functionally replicates such a cutoff by insisting on a literal score of 70 or below with no regard for SEM. The Eleventh Circuit’s approach, he argues, adhered to Hall and Moore by treating SEM as real, rejecting mechanical rules, and insisting on a holistic assessment that included both IQ and adaptive functioning.
On the facts, Smith emphasizes that the district court did not simply grab his lowest score and stop there. Rather, the court heard from experts for both sides, considered the entire sequence of IQ tests, and concluded that the lowest score was not an outlier when viewed in that context. It then evaluated decades of evidence about his functioning in school, at work, and in daily life and found pervasive deficits of the sort clinicians associate with intellectual disability. The Eleventh Circuit, applying deferential clear-error review, saw no basis to overturn that factual determination.
Smith and his supporting amici, among them disability-rights organizations and professional groups like the American Psychological Association, warn that Alabama’s position would tilt the system toward treating IQ scores as dispositive, contrary to both clinical standards and the Eighth Amendment’s command to avoid an “unacceptable risk” that people with intellectual disability will be executed.
What is the government’s position?
Alabama reads the case very differently. For Alabama, the starting point is its own substantive definition of intellectual disability. State law requires a defendant to prove an IQ of 70 or below, significant adaptive deficits, and onset before age 18. On that first prong, the State stresses that Smith has never recorded a single full-scale score of 70 or less. Every test fell between 72 and 78. That pattern, Alabama argues, is fundamentally inconsistent with the conclusion that his true IQ is at or below 70.
Alabama contends that the lower courts misread Hall by treating any SEM-adjusted value that dips below 70 as effectively sufficient to satisfy the intellectual-functioning prong. In the State’s view, this creates a de facto “one-low-score” rule: as long as a defendant can derive a sub-70 value from a single test once error is considered, the state-law threshold is met, even if all the other scores point higher. That, Alabama says, distorts both Atkins and Hall, which were meant to preserve meaningful state discretion rather than to impose a “one-way ratchet” favoring relief whenever there is any statistical uncertainty.
The State’s briefs analyze it from a statistical perspective. If a person takes multiple tests and consistently scores in the mid-70s, Alabama argues, the cumulative evidence suggests that his true IQ is above 70, even if each individual score has a margin of error. It claims the lower courts were fixated on the possibility that one score might be as low as 69 while ignoring the information contained in the rest of the distribution.
The United States, in an amicus brief, largely shares this view. It urges the Court to make clear that states and the federal government may place the burden of proof on the prisoner, may require courts to consider multiple IQ scores together rather than favoring the lowest possible estimate, and are not constitutionally compelled to treat the lower end of the SEM range as determinative. So long as states do not use the kind of rigid, single-score cutoffs rejected in Hall, the Solicitor General argues, Atkins leaves them substantial room to shape evidentiary rules.
What could happen at oral argument?
Oral argument is likely to bounce between technical questions about psychometrics and institutional questions about the roles of state courts, federal habeas courts, and the Supreme Court itself.
Several justices can be expected to press Smith’s counsel on how far Hall goes. Does Hall truly require courts to treat SEM-adjusted scores as if the lowest possible value were the best estimate of a defendant’s true IQ, even in the face of a long run of higher scores? If not, what does a “holistic” approach demand in concrete terms? At the same time, Alabama will be asked to explain why its approach does not simply revert to the rigid number-based rule Hall rejected, especially in cases where scores cluster very close to 70.
The Court is also sitting as a federal habeas court, reviewing a judgment that, at least formally, arises under a deferential standard. That posture may prompt questions about how much room the Supreme Court has to overturn a district court’s fact-intensive conclusion in a “close case” where experts disagreed and the court of appeals found no clear error. Some justices may be wary of turning a dispute about how to weigh IQ tests into a broader invitation for federal courts to re-try capital sentencing issues.
Expect those familiar ideological and methodological divides on the death penalty to be on full display. With Chief Justice Roberts and Justices Thomas and Alito having dissented in Hall and Thomas also having dissented in Atkins, they are natural candidates to look for ways to cabin Hall and increase deference to state court judgments, especially now that three members of the Hall majority are gone and only Justices Sotomayor and Kagan remain from that majority. By contrast, Sotomayor, Kagan, and (very likely) Justice Ketanji Brown Jackson are poised to emphasize the risk that a more state-friendly rule will allow people with genuine intellectual disability to be executed. That leaves Justices Gorsuch, Kavanaugh, and Barrett in the middle, and listeners should pay close attention to how they discuss respect for state criminal judgments here.
Key signals to watch for during oral argument
Listen to how the justices talk about Hall and Moore. Do they describe those decisions as binding constraints that strongly favor Smith, or as narrow holdings that can be reconciled with Alabama’s approach without much difficulty? Watch for whether members of the Court, particularly the justices appointed in the last decade, cast Hall as a modest limit on extreme cutoffs or as a broader endorsement of clinically grounded flexibility.
Watch the Court’s comfort level with statistical reasoning. Some justices may probe whether it makes sense to speak in terms of the probability that Smith’s true IQ is below 70 given all of his scores, and whether the law should recognize that multiple tests can sharpen, rather than obscure, that estimate. Others may resist importing such formal methods into Atkins litigation, preferring to keep the focus on legal standards and burdens of proof. The tone of those questions will signal whether the Court is inclined to push lower courts toward more rigorous quantitative analysis or to leave those choices largely to state law and expert practice.
Take note of how much attention the justices pay to adaptive functioning. To the extent the argument dwells on Smith’s life story, his schooling, work history, social functioning, and everyday skills, it will reflect a view of intellectual disability consistent with clinical practice and the Court’s own rhetoric in Moore. If instead the discussion remains largely about numbers and thresholds, that will suggest an appetite for a more constrained, IQ-centric reading of Atkins.
How this case can impact future cases
If the Court agrees with Alabama’s approach, the ripple effects will be immediate and concrete. Capital defendants whose IQ scores cluster in the low-to-mid-70s could find the courthouse door narrowing, especially in states that, like Alabama, tie Atkins to a strict numerical threshold. A decision that foregrounds aggregated scores and emphasizes the prisoner’s burden of proof would give states a clearer green light to deny Atkins relief whenever the overall testing record trends above 70, even if one or two scores, viewed in isolation and adjusted for SEM, could plausibly support a sub-70 estimate. In practical terms, that would make it easier for states to argue that “borderline” defendants fall on the wrong side of the constitutional line.
A very different future emerges if the Court instead doubles down on Hall and Moore and agrees with the Eleventh Circuit’s holistic analysis. That kind of opinion would cement the message that courts must look past raw numbers and graphed score distributions and give real weight to adaptive-functioning evidence whenever SEM places a defendant in the borderline range. Trial judges, state supreme courts, and federal habeas courts would almost certainly cite such a ruling in capital cases across the country, treating IQ tests as diagnostic tools to be interpreted alongside life histories and clinical assessments.
The stakes are not limited to state systems. The federal government confronts the same intellectual-disability questions under its own capital statutes and has aligned itself with Alabama here, so whatever framework emerges from Hamm v. Smith will also apply to federal prosecutions and habeas review.
