The Supreme Court Rejects Fake Facts in Capital Cases

The Supreme Court recently decided Moore v. Texas, in which it considered a decision by the Texas Court of Criminal Appeals (“the CCA”) (the highest court there that hears criminal cases) affirming a death sentence. The condemned had argued that he was intellectually disabled and therefore ineligible for execution. The state habeas court accepted the argument, based on current clinical standards found in the 11^th edition of the American Association on Intellectual and Developmental Disabilities clinical manual and the 5^th edition of the Diagnostic and Statistical Manual of Mental Disorders, published by the American Psychiatric Association, and made a recommendation that Moore be granted relief.

The CCA rejected the habeas court’s recommendation and, instead of appealing to up-to-date clinical standards, utilized a test for intellectual disability that came from a 2004 Texas case, Ex parte Briseno, and that incorporated a now-superseded manual on intellectual disability coupled with factors that the Texas court appeared to have invented based on stereotypes about intellectual disability (in part perhaps coming from John Steinbeck’s 1937 novel, Of Mice and Men, which the opinion cites). The U.S. Supreme Court held that the CCA erred in applying the Briseno factors coupled with an older clinical standard and that it should have used up-to-date clinical standards to determine intellectual disability. In this column, I will consider the implications of this decision for other areas of law in which factual truth does not always inform legal practice.

What Was Wrong With the CCA’s Approach?

Ordinarily, courts rely on precedents to determine how to approach a question of fact or law. The CCA did just that in appealing to the 2004 Briseno case to determine whether Moore was intellectually disabled. The CCA was correct, moreover, in observing that the state habeas court—which had recommended a finding of intellectual disability—seemed to be substituting its own judgment about what standards to apply for the requirements of the Briseno precedent. Utilizing up-to-date clinical standards to assess intellectual disability was not part of the applicable precedent; yet the state habeas court made its recommendation on the basis of such standards. So what was wrong with what the CCA did?

The answer lies in the Supreme Court’s view of what Atkins v. Virginia required: that no one who is intellectually disabled be executed. Using the Briseno precedent to judge intellectual disability would, the Court found, yield a class of intellectually disabled defendants and convicts that would constitute only a subset of the actual number of intellectually disabled defendants and convicts. In other words, up-to-date clinical standards would give us a factually more accurate determination of the universe of intellectually disabled defendants and convicts. To rely on an inferior and factually questionable standard instead would leave some number of intellectually disabled people vulnerable to execution. And that would violate the Eighth Amendment ban on cruel and unusual punishments. For the Court in Moore, then, the reality of who is and who is not intellectually disabled mattered, and the Texas CCA approached the question in a non-reality-based manner.

Other Zones Where Facts Matter Less

At first glance, the Court’s ruling seems very sensible. If we have a more accurate and better informed way of distinguishing intellectually disabled defendants, it does seem misguided and unconstitutional to choose instead to use a standard coming from a legal precedent that gives us inferior results. But we do not always see the law pursuing the most factually accurate outcome. Consider some examples.

In the law of evidence, which one would expect—from the name—to be very evidence-based, we have an important fiction which we deploy. The fiction is that when jurors are given an instruction that defies common sense, the jurors will nonetheless be both capable of following and willing to follow the instruction as given to them by the judge. Studies have shown that this is truly a fiction. In one study, researchers discovered that jurors ignore instructions to consider a criminal defendant’s prior convictions only on credibility and not as evidence of guilt, with jurors openly admitting that they did not consider the convictions on the credibility question but did consider them evidence of guilt. The effect was especially strong when the prior conviction was for a crime similar to that being tried. In another study, the authors determined that when jurors in a civil trial are instructed not to talk about insurance, they do so anyway and even make assumptions about the plaintiff having insurance, though they have heard no evidence supporting that assumption. Instructions, in many cases, do not work and may even tempt the jury to do precisely what they are told not to do (just try to follow my instruction now: “Don’t think about pink elephants”). Yet in most cases, criminal defendants convicted after juries were given (likely ineffective) instructions do not have their convictions reversed by a higher court.

Beyond the law of evidence, we can look to how the U.S. Supreme Court has handled issues in other criminal procedure contexts. Take the case of the “consent search.” A consent search occurs when police officers, having neither warrant nor probable cause, simply ask a suspect for permission to perform a search and receive the requested permission. In one case, Schneckloth v. Bustamonte, the respondent argued that in order to prove that a search truly was the product of voluntary consent (and its results therefore properly admissible in evidence), the government should have to prove that the suspect knew there was the option of saying no in response to the request for consent. This seems eminently reasonable—very much like the requirement that a court use up-to-date clinical standards to determine that a convict is not intellectually disabled. Yet the Supreme Court rejected the claim here and held that knowledge that one has the right to refuse consent is just one factor to consider in determining whether a consent to search was voluntary. Stated differently, there could in theory be voluntary consents to searches in which the consenters were unaware that they had the option of refusal. To the extent that the Court was using the word “voluntary” in the ordinary manner, to mean that a person freely chose to say yes to a search, this ruling flies in the face of the facts. If a suspect believes that he must say yes in response to a request for consent to search, then his “yes” is not voluntary. Yet the Court found otherwise.

Consider also the context of custodial interrogation. When police interrogate a suspect in custody, they are required to give Miranda warnings, alerting the suspect to the right to remain silent, the right to counsel, etc. Under Edwards v. Arizona and its progeny, if a suspect requests an attorney during an interrogation, the interrogation must stop and not resume, so long as the suspect remains in custody, until the suspect has consulted with counsel and has counsel present at the interrogation. The reason for this rule is that people who ask for an attorney are understood to be feeling pressured and coerced and unable to handle the interrogation alone. That is why a request for counsel is more powerful than a request for silence (which only yields, under Michigan v. Moseley, a several-hour break in interrogation, after which police may return with a new set of Miranda warnings). Understanding that suspects who ask for lawyers feel vulnerable and subject to coercion, one Davis brought a case to the Supreme Court saying that an equivocal request for counsel (such as “Maybe I should talk to a lawyer.”) ought to trigger the Edwards right to have interrogation stop. If a person who unequivocally requests a lawyer feels vulnerable, it would seem to follow that a person who equivocally does so feels even more vulnerable and unable to be assertive and demand an attorney. Janet Ainsworth discusses the phenomenon of women and minorities being less assertive in just this way in her article, “In a Different Register: The Pragmatics of Powerlessness in Police Interrogation.” Yet the Supreme Court, ignoring this rather obvious reality, chose to hold, in Davis v. United States, that absent a clear request for counsel, interrogation may continue.

Fake Facts

One could say, perhaps uncharitably, that the law in general and the Supreme Court in particular are prepared in a variety of contexts to assume that fictions are true on the way to reaching their rulings. The law pretends that consent is voluntary even when consenters believe they must say yes. The law pretends that people pose equivocal requests for counsel because they are not sure they really need assistance. And the law pretends that juries follow counterintuitive instructions, even though we know that they do not. This is unfortunate, because basing rulings on reality generally yields superior outcomes. But there are reasons for relying on “fake facts.” The Supreme Court likes consent searches and wants to be able to say that more searches are consensual; a fictional account of voluntariness allows it to do so. And the Court also likes when police are able to interrogate suspects; pretending that equivocal invocations of counsel signify something other than vulnerability and a need for support allows the Court to do so. And in the law of evidence, we want to be able to cross-examine defendants about their prior convictions when they take the witness stand; pretending that a limiting instruction prevents jurors from relying on those convictions as proof guilt permits us to do so. And more generally, limiting instructions may be a necessary evil because we cannot simply exclude all relevant evidence that has a downside risk attached.

In the context of executing the intellectually disabled, however, the Court has made the wise and honorable decision to reject fake facts. Texas will not be allowed to pretend that people who are intellectually disabled are not, simply because they fail to measure up (or down) to a standard that combines old clinical approaches with invented factors. In at least this one case, the facts will matter. And we can perhaps hope that this thinking will spread beyond the particulars of this case to the many other contexts in which the law engages in pretense and fictional assumptions.