Robert Roberson faces execution in Texas on Oct. 16 for the 2002 murder of his young daughter Nikki Curtis. For years, a growing pool of supporters — from the lead detective in his case to Dr. Phil — have argued Roberson is innocent. They say doctors and prosecutors blamed him for Nikki’s death through a faulty diagnosis of shaken baby syndrome.
But that was not the only diagnosis that sent Roberson to death row. After he was found guilty, a psychologist named Thomas Allen interviewed him using the Psychopathy Checklist-Revised. Sometimes called the “Psychopath Test,” it measures traits like a person’s impulsivity and lack of remorse. Allen gave Roberson a high score, told the jury he was a psychopath, and compared him to Adolph Hitler and Saddam Hussein. The implication was that Roberson would continue to be dangerous, although Allen said the risk was much smaller if he were in prison.
There will always be some mysteries here: There are no available recordings of Allen’s interview with Roberson or the jury’s deliberation about the diagnosis. But as Texas prepares to execute Roberson, a growing number of researchers and lawyers are questioning whether the word “psychopath” is being used too often in courts, terrifying jurors into harsher punishments based on stereotypes from television and movies.
A Marshall Project review of court records found more than a dozen death sentences handed down in cases as far back as 1998 that involved court testimony based on the psychopathy checklist. Because no one tracks such cases in a comprehensive way, there are likely many more. In 2021, the U.S. Supreme Court declined to hear Texas prisoner Quintin Jones’ argument that the checklist had falsely identified him as a psychopath. Jones was executed soon afterward. Like many on death row, he had admitted his guilt. But Roberson has maintained that he did not kill his daughter, raising the question of whether an innocent man was diagnosed as a psychopath on his path to execution.
The checklist in question was designed by Canadian psychologist Robert Hare in the 1970s. The 40-point scale measures personality traits and past behavior. Hare and his students found that the same person could get a similar score from multiple evaluators, even though some categories, like “superficial charm,” might appear subjective.
Not everyone who scored high on the checklist was necessarily violent. Hare himself co-wrote a book called “Snakes in Suits” on why psychopaths succeed in some corporate environments. But the public was captivated by the image of the charming and remorseless killer, as seen most clearly in movies like “The Silence of the Lambs.” A series of studies found that people in prison who scored high on the checklist appeared more likely to commit new crimes after release. In 1991, Hare agreed to sell the test for use by prisons, courts and parole boards.
A 2014 study found the checklist was showing up all over the U.S. It had become so prevalent that it was impossible for Hare or anyone else to track or regulate the checklist’s use. Prosecutors used high scores to send young people from juvenile courts to adult courts. Parole boards relied on high scores to keep people in prison. But the most consequential use was in death penalty trials.
Hare has long advocated for careful standards in the checklist’s use and more research on its influence in courtrooms. He told a Canadian magazine in 2001 that the checklist shouldn’t be used when the death penalty is on the table.
“I would be appalled if a decision of life or death were based on the [checklist],” he told The Marshall Project, adding that it should never be “the sole, or even the primary, basis for a decision with serious consequences for an individual.” He lamented the problem of “hired guns” in the courtroom, but added that he could not be held responsible for every use of his invention, comparing the checklist to a blood pressure monitor.
The checklist emerged after an era in which expert witnesses relied on little other than their own judgment. The most famous was psychiatrist James Grigson of Dallas, often nicknamed “Dr. Death.” In 1977, Grigson diagnosed a man named Randall Dale Adams as an “extreme psychopath,” convincing a jury to send the man to death row, only for him to later be proven innocent and freed.
Critics of the checklist say it has not been a solution. Ideally, the diagnosis should be based on someone’s well-documented life history, but some evaluators fail to get a full picture and rely on what prosecutors tell them. The result is that jurors might get a false impression that the diagnosis is neutral. In 2003, a longtime expert witness named Thomas Ryan disavowed his past use of the checklist for prosecutors in death penalty cases. He acknowledged there was no solid evidence that the checklist could predict violence within prisons, where people would spend the rest of their lives if not executed. Some researchers also began studying treatment options, undermining the older notion that psychopaths could not be helped.
“The checklist makes the testimony sound much less subjective than it really is,” says Texas A&M University professor John Edens, who has studied the checklist for years and been hired by defense lawyers seeking to challenge it in court. “Your assessment is always contingent on what you know or just believe about somebody.”
In 2002, Roberson told the police in Palestine, Texas, that he awoke in the early morning to find his 2-year-old daughter crying and bleeding. He later told police that he wiped the blood from her chin, waited until she seemed well, and went back to sleep.
When he awoke again, Nikki was unresponsive, and he took her to the hospital. Nurses and police later described Roberson’s unusual behavior and lack of emotion, some of which has since been tied to his autism diagnosis. A series of doctors and police concluded Roberson’s daughter died due to physical abuse. In the years since, his lawyers have cited numerous doctors who argue Nikki was ill with pneumonia and fell out of bed, causing her fatal injuries.
After a jury convicted him of murder, Roberson met with Allen, a forensic psychologist hired by the prosecution. Allen gave him a score of 34 on the checklist, above a common cutoff score of 30 for psychopathy. “I think he’s going to explode, be explosive, with people with whom he thinks he can get away with it,” Allen told the jury. He testified that Roberson showed an “absolute lack of remorse” — a well-known feature of psychopathy on the checklist. He focused on how Roberson had fallen asleep after first noticing Nikki’s injury.
At the same time, Allen said, if Roberson were sentenced to life in prison, he would be a “minimal” risk to others, in part because he would no longer be able to spend unmonitored time with children.
But “he’s still a psychopath,” Allen said. To illustrate the concept, he named others he believed to be psychopaths, including Hitler, Hussein, and Ted Bundy. He also named former President Bill Clinton, although he said he was kidding about that example.
In a 2010 petition, Roberson’s lawyers unsuccessfully challenged Allen’s testimony as prejudicial and “junk science at its worst.” They wrote that Allen relied on accusations that Roberson sexually assaulted his daughter and another woman. These accusations were never brought to court due to a lack of evidence; the woman asked prosecutors to dismiss the charge against Roberson in her case.
Hare himself read parts of the petition recently and called Allen’s testimony a “gross misuse of the instrument and completely out of line with the conditions for the use of the PCL-R, which have been amply and widely articulated over the past 30 years.”
In a Marshall Project interview, Allen defended his use of the checklist as careful and ethical. He said it was not his job to independently investigate Roberson’s guilt, and that his testimony probably had little effect: “Frankly, when you’ve got someone who kills a baby, was found guilty of that, it doesn’t matter what I say.”
But Allen and Hare agree with critics of the checklist on one point: Allen’s testimony may have been less persuasive if Roberson’s own lawyers had asked sharper questions about the diagnosis at his trial. Psychologist Henry Richards, who has used the checklist to evaluate people in other cases, read an excerpt of the trial transcript for The Marshall Project, and called the cross-examination cursory and lightweight. The lawyer responsible, Stephen Evans, did not respond to a request for comment.
Roberson’s new lawyers weren’t so sure that a fiercer cross-examination would have been enough. They wrote in their 2010 petition that “It is impossible to remove the harsh stain from the jury box once a psychologist throws the label ‘psychopath’ at the defendant.”
