Texas sentencing is out of date: No one can predict future crimes, even on death row
The state of Texas plans to execute Ramiro Gonzales on June 26. Gonzales was sentenced to death in 2006 for kidnapping, sexually assaulting and fatally shooting 18-year-old Bridget Townsend five years earlier. At the time, Gonzales was also 18.
One of the key factors that led the jury to sentence Gonzales to death was testimony offered by psychiatrist Edward Gripon claiming, with certainty, that Gonzales would be a future danger. Gonzales’s case offers a striking illustration of the unreliability of predicting future dangerousness in capital cases, something even Gripon has come to recognize.
He testified pursuant to a 1973 Texas law that requires juries “to determine whether a defendant presents a future danger to society before imposing a death sentence.” Texas is the only state in the country with such a law.
This law has spawned an industry of experts who, like Gripon, make themselves available to testify for the state about future dangerousness. Gripson estimates he “testified in roughly 25 death penalty cases.” But he is far from the most infamous.
That distinction belongs to psychiatrist James Grigson. As a post from Vanderbilt Law School explains, “Grigson, nicknamed ‘Dr. Death,’ testified in over 150 capital murder trials for the prosecution in Texas and often stated that there was a 100 percent chance that capital murder defendants would kill again, despite never having personally examined them.”
Journalist Abbie VanSickle, writing in The Atlantic, notes that people like Grigson and Gripon had to tell the jury “whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society.” VanSickle rightly labels this question “convoluted.”
In the Gonzales case, Gripon told the jury that the defendant “certainly” had antisocial personality disorder, and potentially “some type of significant underlying psychosexual disorder.”
He testified that “sexual assault has the highest continuum of recidivism,” and “there is lots of data out there” indicating that recidivism rates “are way up in the eighty percentile or better.” Gripon concluded that Gonzales “would pose a risk to continue to commit threats or acts of violence” “wherever he goes,” even in a carceral setting.
There are different approaches that can be taken to predict the danger a defendant poses. The AMA Journal of Ethics identifies common approaches while warning of the unreliability of each one.
First, there are what the journal calls “actuarial approaches.” They “attempt to assess individual risk using information derived from group data rather than from an individualized assessment of dangerousness.” The accuracy of such assessments in predicting rare events is “low because its prediction is limited to those who are similar to the population from which the actuarial data were drawn.”
In contrast, what the journal calls “clinical approaches” depend on “individualized and contextualized assessments based upon nothing more than a psychiatrist’s intuition, experience, and clinical orientation.”
Identification of risk factors in the clinical approach is, the Journal of Ethics concludes “vulnerable to individual bias.” In fact, we know that race plays a key role in predictions of future dangerousness, with people of color much more likely to be labelled as a future danger, especially in capital cases.
Another problem in predicting future dangerousness is that experts, like Gripon in the Gonzales case, erroneously associate dangerousness with mental illness. As the Journal of Ethics notes, “The absolute risk of violence among the mentally ill as a group is...very small.”
In fact, the Gonzales case itself shows how hard it is to get predictions of future dangerousness right.
At trial, as The Independent stated, “Gripon relied on facts that were later proven false. He told the jury that people who commit sexual assault are especially likely to reoffend, but the basis for that claim was later discredited and numerous studies have shown they aren’t.”
Gripon also based his testimony on Gonzalez's cellmate’s claim that he heard the defendant “admit to torturing Townsend and return[ing] to the crime scene to sexually defile her body. But the cellmate later recanted, explaining that officers were threatening a more severe sentence in his own case if he didn't help him paint a nightmarish picture of Gonzalez.”
And during his time on death row, Gonzales changed. As The Texas Tribune reported in 2022, he “turned to faith.” Prison officers testify to his “generosity” and genuine “desire to do good after the crimes of his youth.”
He has also shown, as the Marshall Project says, “incredibly good character in his years in prison — he has never committed an act of violence or threat, is a positive influence on those around him and a faith leader for fellow death row prisoners.” Evidence also suggests that he is “fully rehabilitated and could not possibly pose a threat of future danger to anyone, and thus cannot be executed under Texas law.”
Even Dr. Gripon, who evaluated Gonzales again two years ago, has changed his mind. Talking about his 2006 testimony, Gripon now says “I don’t think that diagnosis would now be accurate, particularly in retrospect” — and most importantly, that Gonzales does not “present a danger to others.”
“At the time of the commission of this offense,” Gripon explained, “Gonzales was barely 18 years old. With the passage of time and significant maturity he is now a significantly different person both mentally and emotionally. This represents a very positive change for the better.”
“If this man’s sentence were changed to life without parole, I don't think he'd be a problem.”
Texas should listen to its own expert and spare Gonzalez's life. In addition, it should pay attention to the position of the American Psychiatric Association that the question of future dangerousness should no longer play a role in jury decisions in capital cases. As the APA notes, “[t]he unreliability of psychiatric predictions of long-term future dangerousness is by now an established fact within the profession.”
It is long past time for Texas to get out of the business of asking juries to use those future predictions when making life and death decisions.
Austin Sarat is the William Nelson Cromwell Professor of Jurisprudence and Political Science at Amherst College.