Why is it So Hard for Courts to Adjust to Advancements in Knowledge of Human Behavior?
A Death Penalty Case Study
We are thrilled to welcome this week Katie Kronick as a guest essayist on the Sentencing Matters Substack. Katie is an Assistant Professor of Law at the University of Baltimore School of Law. She is also the Founding Director of the law school’s Criminal Defense and Advocacy Clinic, where students represent indigent people charged with misdemeanors and low-level felonies in Baltimore City courts. Katie previously served as an Assistant Deputy Public Defender in Hudson County, New Jersey, before becoming the Practitioner in Residence at American University’s Washington College of Law. Katie's scholarship focuses, in part, on the intersection of intellectual disability and sentencing.
- - -
Death penalty cases do not always make headlines, but when John Grisham, the main detective on the case, and other prominent individuals are raising the alarm that an innocent person is about to be killed by the state of Texas, people pay attention and headlines are made. This has been the case with Robert Roberson since October 2024.
This case has all the twists, turns, and intrigue of a Grisham novel. After exhausting (and losing) every legal path to relief, the Texas Supreme Court stayed Mr. Roberson’s October 17 execution because, in a last-minute turn, the Texas legislature subpoenaed Mr. Roberson to testify in a committee meeting. Then, just about a month later, the same court held that a legislative subpoena cannot supersede a death warrant. In mid-November, the state court judge, who needs to issue the new death warrant, recused herself from this case. As of today, Mr. Roberson does not have an execution date, but he is scheduled to go before the legislature on December 20, 2024.
And also like a Grisham novel, this case is filled with complex characters and situations that raise questions about the efficacy of the criminal legal system. Mr. Roberson, who has maintained his innocence since this tragic incident more than twenty years ago, was convicted of killing his two-year-old daughter. To prove their case, the prosecution relied on Shaken Baby Syndrome, a now debunked diagnosis—often deemed “junk science.” Mr. Roberson’s daughter suffered from several chronic illnesses, which were more likely the cause of her tragic death.
Rightly, much attention in the legal pleadings and the media around this case has been the prosecution’s reliance on junk science to form the basis of the jury’s guilty finding and, ultimately, the death sentence. Yet, many also argue that a significant reason Mr. Roberson is facing the death penalty is that he has autism.
In 2003, at the time of the investigation into and prosecution of this case, Mr. Roberson did not have an autism diagnosis. The diagnosis came more than a decade into his death sentence in 2018. But back in 2003, he was exhibiting some characteristics of a person with autism. And people involved in this case (nurses, police detectives, and the prosecution) misinterpreted these characteristics and behaviors as being indicative of his guilt. According to the reports, he had dressed his daughter prior to bringing her to the hospital and paused to make a sandwich while showing police around his home. A detective found him “just not right.” All of this was used at trial to paint Mr. Roberson as a “callous and remorseless” killer. This behavior, we know now, is far more likely to be reflective of his autism—not guilt.
Much of the conversation around Mr. Roberson’s autism diagnosis centers around whether the outcome would have been different if Mr. Roberson had the diagnosis at the time of his conviction in 2003 rather than 2018. Maybe it would have been different, but I am doubtful. This belief assumes that when lawyers, judges, and other actors in the criminal legal system know about a person’s mental disability, they act with nuance and understanding.
Unfortunately, the reality is often far crueler. What we know is that outcomes for persons with mental disabilities in the criminal legal system are worse than for similarly situated individuals—they are more likely to be arrested, held in jail pretrial, receive worse plea deals, and serve longer sentences. They are also more likely to be wrongly convicted. Mr. Roberson’s case of injustice is just one in a long line of cases in which our criminal legal system has failed persons with mental disabilities.
Studies indicate that 40% of people facing charges in the criminal legal system has one or more mental disability. Despite their disproportionate presence in this system, most lawyers, judges, and other system actors are unequipped to identify whether a person has a mental disability or how to best accommodate them.
Thus, time and again, we end-up in situations like Mr. Roberson’s—misinterpretations that result in grave harm. In his case, the prosecutors relied on behaviors that were manifestations of his autism to prove his guilt. More broadly, persons with mental disabilities are more likely to be arrested and convicted because actors in the criminal legal system routinely mistake behaviors that directly relate to their disability as indications of criminality or dangerousness when they are neither more dangerous nor more prone to criminality.
This problem not only infects the guilt phase, but likely more frequently affects what happens at sentencing. Just as his “callous and remorseless” behavior may have been indications of guilt, they may have also spurred a death sentence instead of life without parole.
With more than 90% of cases resolving with plea bargains rather than trials, sentencing becomes the critical moment in which a defense attorney can argue for their client before the judge. Yet, people with mental disabilities have a harder time providing mitigating information to their attorneys such that it can be used effectively at sentencing. And, in some cases, providing information about a mental disability at sentencing can result in an increased rather than decreased sentence. For example, in a review of sixty-three federal cases, when the court considered the defendant’s intellectual disability, the court found it either irrelevant to the sentencing determination or an aggravating factor.
Then, while serving a sentence, a person with mental disabilities might still experience worse outcomes. If on probation or parole, they might struggle to make meetings, find and maintain employment, and be unable to obtain services because of illiteracy. If incarcerated, they might do more time because they are susceptible to infractions and may be ineligible for certain programming. In addition, jails and prisons are ill equipped to serve persons with mental disabilities, sometimes with deadly consequences.
These disparately negative sentencing outcomes occur for persons with mental disabilities even though the Americans with Disabilities Act (ADA), which applies to the states, and the Rehabilitation Act (RA), which applies to federal entities, require that the court systems make accommodations for persons with disabilities.
The ADA/RA make clear that persons with mental disabilities should not have worse outcomes because of their disabilities—the courts and related systems should provide accommodations as long as doing so would not cause undue burdens or pose a significant risk to health or safety. Despite the ADA/RA’s applicability, which the Supreme Court affirmed in Tennessee v. Lane, the ADA/RA have historically provided only rare relief at any stage of the criminal legal system.
Mr. Roberson’s case, though in some ways extreme, is reflective of this wider problem—a problem with which some courts are finally willing to reckon. In fall 2024, a federal judge for the District of Columbia issued a preliminary injunction, requiring the United States Parole Commission and Court Services and Offender Supervision Agency to accommodate persons with disabilities. This judge is among a very few to force the criminal legal system to provide accommodations for persons with mental disabilities.
The agencies’ failures were not small inconveniences or slights—their actions have directly resulted in men going back to prison. As the court noted in issuing his order, “Absent immediate relief, the Parolees will face irreparable harm; namely, obstacles to success on supervision solely because of their disabilities, which expose them to downstream harms like revocation and reincarceration.”
For example, one of the plaintiffs in the D.C. litigation, Mr. Kennedy Davis, has mental disabilities that sometimes result in him missing appointments. Despite his requests for accommodations, he is now back in prison because of these technical violations. He has committed no new offenses. Though this ruling, so far, only applies to the named plaintiffs, this litigation could result in broader improvements for persons with disabilities under supervision in D.C.
In contrast, just a few years earlier in a courthouse down the street, an attorney in a delinquency matter in D.C. Superior Court asked the judge to require the shelter home at which the child was detained pretrial to provide accommodations for his mental disabilities. The court barely entertained the request before flatly rejecting it. Unsurprisingly, the child did not receive necessary accommodations and was removed from the shelter home.
The largely conspicuous absence of accommodations is attributable to many factors including a failure of attorneys and government officials to identify whether a person has a disability, a lack of training amongst those who work in the criminal legal system, and the systemic focus on efficiency over nuance.
Compounding these problems is the lack of a clear mechanism for requesting accommodations in the criminal legal system. Should an attorney request accommodations in advance of the sentencing hearing? During the hearing? Separately in a civil action? Unlike in the employment or education contexts, the court systems do not have clear pathways to ensuring accommodations. The result is that persons with mental disabilities might choose not to alert anyone to their disability because they are unlikely to receive accommodations and risk they will face even more bias and discrimination in a system already rife with it.
Returning to Mr. Roberson’s case, even if he had an autism diagnosis back in 2003, I am quite skeptical that the system would have reacted differently. When a person has a mental disability, we (both the public and the law) expect them to comport with what a “reasonable person” would do. I am doubtful that the nurses, detectives, and prosecutors would have had the training and time to understand how these “just not right” behaviors related to his autism rather than a guilty conscience.
If knowledge of a mental disability does not result in a more just outcome, how do we get there? How do we ensure that persons with mental disabilities are not experiencing worse outcomes because of their disabilities? That is a complicated question in an underfunded system that places a high priority on efficiency and finality.
The experiences of Mr. Roberson and Mr. Davis exemplify the failures to adequately address the needs of persons with mental disabilities. Sentencing, though, represents a moment in which some of the wrongs can be righted. If bias infected the policing, led to pretrial detention, and resulted in a harsh plea, the court can make the choice to sentence in a way that takes into account not only the disabilities but also how the system might have produced less just outcomes throughout the process. And where individuals are sentenced to probation, courts can stop incarcerating people just because they miss a meeting or fail to find employment, frequent manifestations of disability.
Robert Roberson has a brief but tenuous reprieve to continue his fight. But the injustice and cruelty of his experience should not be forgotten—it should spur us to think more deeply about how we treat and sentence persons with mental disabilities in the criminal legal system.
 | A guest post by
|