SCOTUS’s Unclear Reversal of Capital Case It Deems “Unclear”

The U.S. Supreme Court has finally acted on Alabama’s petition in the case of murderer Joseph Clifton Smith. The high court’s repeated relisting of this case for consideration in an unprecedented number of conferences has drawn considerable speculation as to what was going on.

The high court did not take the case up for full briefing and argument but instead sent it back to the Eleventh Circuit for a do-over. That is most unfortunate, because they failed to clean up a mess of their own creation.

In the 1989 case of Penry v. Lynaugh, the Supreme Court ruled correctly that its precedent in Lockett v. Ohio requires capital sentencing juries to consider mental retardation (as it was then known) as a mitigating circumstance. (Whether Lockett itself was correctly decided is another question. See this article.) In 2002, the high court decided that wasn’t good enough, and it made mental retardation (as it was still known then) a categorical exclusion. Along with the constitutional problems, there is a huge practical problem. The Court tried to draw a bright-line rule with a paint roller.

Intelligence is a continuous spectrum, and the breakdown into categories is entirely a human construction. There are no natural dividing lines set by objective science. The lines are therefore subject to manipulation, as discussed in this post.

Maturity is also a spectrum, of course, but the Supreme Court’s bright-line rules on sentencing for juveniles have a crisp dividing line set by very long tradition. For most purposes, people are legally juveniles before their 18th birthday, and after it they are adults. (In 1837, 18 was old enough to be Queen of the UK without a regent.) There is no such crisp line for intellectual disability, as it is now known.

One portion of the definition has to do with “intelligence quotient.” IQ is now defined so that people with IQ 70 or below are at least two standard deviations below the mean, i.e., in the bottom 2.5%. (The original definition of mental age / chronological age went into the dustbin of science long ago.) The IQ tests have their own margin of error (standard error of measurement) so that for some tests it is possible (though highly unlikely) for a person to have a test as high as 75 and still qualify as having an actual IQ of 70.

The Smith case involves a more precise test, so three points are allowed for lowest possible real IQ for a given score. When a person has a number of tests, should we just subtract three from the lowest of all the scores? Of course not.

From today’s opinion:

The Eleventh Circuit’s opinion can be read in two ways. On the one hand, the Eleventh Circuit’s opinion might be read to afford conclusive weight to the fact that the lower end of the standard-error range for Smith’s lowest IQ score is 69. That analysis would suggest a per se rule that the lower end of the standard-error range for an offender’s lowest score is dispositive. On the other hand, the Eleventh Circuit also approvingly cited the District Court’s determination that Smith’s lowest score is not an outlier when considered together with his higher scores. That analysis would suggest a more holistic approach to multiple IQ scores that considers the relevant evidence, including as appropriate any relevant expert testimony.

Okay, so other than the impossibly vague “more holistic,” what is the correct way to go about this? Here is what the Court said today:

The Eleventh Circuit’s opinion is unclear on this point, and this Court’s ultimate assessment of any petition for certiorari by the State may depend on the basis for the Eleventh Circuit’s decision. Therefore, we grant the petition for certiorari and Smith’s motion for leave to proceed in forma pauperis, vacate the judgment of the Eleventh Circuit, and remand the case for further consideration consistent with this opinion.

Thanks a heap. It took your honors over a year to decide that?!

This issue, if not this very case, will be back. If the Court sticks to the statistical question, it should recognize that multiple tests increase the precision, not reduce it, and the range of possible real IQs gets narrower, not wider. A person like Smith with multiple tests from 72 to 78 is definitely not below 70 in reality. Smith’s claim should not have made it to first base.

Hopefully, though, by then the Court will be prepared to look beyond the narrow statistical issue and consider whether this mess is really required by the Constitution at all. Here are some possibilities:

1. Limit the bright-line rule of Atkins to cases of moderate retardation or greater, as defined by the DSM-IV in effect at the time, and restore the Penry rule for cases of mild intellectual disability. That would resolve nearly all cases, as very few murderers are even arguably within the moderate range.

2. Overrule Atkins and go back to Penry.

3. Apply original understanding to the entire capital punishment line of cases, as the Court has to other provisions of the Bill of Rights and some Eighth Amendment issues. With that done, Lockett v. Ohio and all its progeny are obviously inconsistent with the original understanding and can be overruled. And good riddance.