"Will a Texas law that overturns convictions based on bad science save this death row inmate?" was first published by The Texas Tribune, a nonprofit, nonpartisan media organization that informs Texans — and engages with them — about public policy, politics, government and statewide issues.
PALESTINE — Robert Roberson shuffled into a courtroom this week wearing a striped gray jumpsuit and handcuffs, his life once again hanging in the balance.
After 15 years on death row, his face has grown gaunt, and patches of dark hair shoot up from his balding head. But he has maintained during his time in prison that he didn’t kill his sickly two-year-old daughter, Nikki Curtis, though he was convicted of that crime.
He says that Nikki fell from the bed where they were sleeping in their home in this small East Texas town, and he awoke hours later to find her unresponsive. But as doctors and nurses struggled to revive her blue, limp body in the emergency room that morning, suspicions of child abuse quickly arose — they said a short fall wouldn’t have caused such damage.
At his trial, doctors testified that her injuries were consistent with what is often referred to as “shaken baby syndrome,” a now-questionably diagnosed condition his attorneys say helped jurors opt for the death penalty.
But instead of facing the state’s death chamber — which he narrowly avoided two years ago — Roberson was back in court Tuesday, again fighting against his conviction and for his life, thanks largely to a relatively new state law that allows courts to overturn a conviction when the scientific evidence that originally led to the verdict has since changed or been discredited.
“The only reason Mr. Roberson is still alive today is because the state of Texas passed a rather trailblazing statute,” his attorney, Gretchen Sween, declared to the courtroom.
The law, often referred to as the "junk science law," was the first of its kind in the nation and passed with scant opposition in 2013. In pushing for its passage, state Sen. John Whitmire, a Houston Democrat, listed infant trauma as one of several examples of faulty science the bill was meant to target.
Since its passage, multiple death penalty cases have been sent back to court for further review, and it has been cited in cases like that of the “San Antonio Four,” where women convicted on faulty sexual assault evidence were exonerated after nearly 15 years in prison.
In June 2016, Roberson became one of the first death row inmates to have his conviction set for reviewed under the law — a decision the Texas Court of Criminal Appeals made just days before his scheduled execution. The court was instructed to decide whether Roberson would have been convicted if new scientific evidence — like new views on fatal short-distance falls and shaken baby syndrome injuries — were available at his original trial.
In the last decade, experts have become divided on shaken baby syndrome, where an infant is killed from being violently shaken back and forth. Many doctors strongly stand by the diagnoses, but others, including the doctor who is first credited with observing the condition, think it is used too liberally in criminal cases — that deaths are labeled as murder without considering other possibilities and medical histories. The Washington Post reported in 2015 that 16 shaken baby syndrome convictions had been overturned since 2001.
Roberson’s attorneys argue in part that new scientific evidence has suggested it is impossible to shake a toddler to death without causing serious neck injuries, which Nikki did not have, and has linked the symptoms used to diagnose shaken baby syndrome to other conditions as well, including short-distance falls.
“There has been a tremendous amount of new scientific evidence,” said Gary Udashen, board president of the Innocence Project of Texas. “Biomechanical engineering studies have shown that you can generate enough force from a short-distance fall to cause serious head injuries.”
"A watershed moment"
Over time, science has become an increasingly present part of criminal proceedings, but it’s always evolving — what may have been thought of as irrefutable scientific evidence 15 years ago could be balked at today. And scientific “experts” who testify in cases sometimes lack any true expertise.
The origins of the junk science law, and Roberson’s renewed chance at an appeal, began to take form as the state’s skepticism of forensic science grew with the science's prevalence — think DNA — and after several high-profile cases involving “bad science” emerged.
Perhaps the most well-known is the case of Cameron Todd Willingham, a Corsicana man executed in 2004 for the deaths of his three young daughters, despite scientists discrediting the earlier fire examination and finding no reason to call the house fire that killed his children arson.
But while state lawmakers passed other bills related to forensic science, like creating a statewide investigatory commission, it took several tries for the junk science law to make it out of the Capitol in Austin. Whitmire filed similar bills in 2009 and 2011, but neither piqued much interest, and some prosecutors said an already well-established appeals process made the bill unnecessary.
That changed with Neal Robbins and Cathy Henderson. Both were convicted in infant deaths after the medical examiners who testified at their trials later recanted their certainty that the deaths were homicides. But their appeals based on those recantations produced opposite results.
In 2011, the Texas Court of Criminal Appeals denied Robbins’ appeal on the claim that his conviction and life sentence were based on bad scientific testimony — the medical examiner had changed her original conclusion that said his daughter’s death could only be murder. Though she originally thought the child’s injuries couldn’t have been caused by the reported attempts at CPR, she later said her increased knowledge made her unable to determine a cause of death.
The next year, the same court tossed out Henderson’s conviction and death sentence after scientific advancements conflicted with the medical examiner’s original testimony that dropping a baby from her arms onto the floor couldn’t have caused the fatal brain injury.
“In Robbins, this Court chose finality over accuracy; in Henderson we did the opposite, and in 2013, the Texas Legislature also chose accuracy over finality,” wrote then-Judge Cathy Cochran in 2014 after the court, under the new law, overturned Robbins' conviction. He was released from prison in 2016.
Robbins’ appellate attorney, Brian Wice, said the blowback from the court’s first ruling created the support to eventually pass the bill.
“It was a watershed moment, not just in Texas but nationally, because Texas was really the first state to enact a provision that was able to unlock the courthouse door to defendants like Neal Robbins,” he said.
Another delay of justice
For Roberson, the junk science law has extended his life, but it hasn’t saved him yet. His court-ordered examination finally began in what was assumed to be at least a week-long hearing Tuesday. But it was again delayed after several hours when it was announced that long-lost evidence, including Nikki’s brain scans, was suddenly found in the basement of the local district clerk’s office.
Sween, his attorney, said in court that they ultimately want a new trial for Roberson, one that would include the changes in science on shaken baby syndrome and short-distance falls, as well as a deeper look into Nikki’s previous illnesses and history of breathing problems. The child had previously suffered spells where she'd stop breathing and turn blue, and she had been at the hospital only days before with a violent illness that included a 104.5-degree temperature, according to court documents. The review is also looking at whether the original accusation that Roberson sexually assaulted Nikki, which was dropped mid-trial when the evidence couldn't be corroborated, affected his conviction. She said she’s optimistic Roberson can be given a fair shot.
“All we have to do is show that the science that was presented at trial and the science that was available now ... has changed in ways that more likely than not, might have changed the jury’s verdict,” she said in court.
But the prosecution said it isn’t conceding anything, saying in court that the evidence was “clear and convincing” that Roberson killed his daughter.
“The science has not changed as much as [Roberson’s attorneys] say,” said Anderson County Assistant District Attorney Scott Holden in his opening statement. “There was no explanation except for the intentional death of Nikki Curtis.”
For now, Roberson’s case is up in the air. With the discovery of the lost evidence Tuesday, the judge put it up to the county and Roberson’s attorneys to let her know after they’ve properly evaluated it. After the court does reconvene again, the judge will make a recommendation to the Court of Criminal Appeals on whether a new trial should be granted, which the higher court can either order or reject.
Likely, it will be a lengthy delay, and the wait isn't welcome news for Nikki’s older half-brother, Matthew Bowman. He still carries the weight of his sister’s death and has no doubt that Roberson killed her. He was 4 when she died, and the 20-year-old sitting in the courtroom Tuesday was in visible anguish listening to Sween proclaim Roberson’s innocence, often shaking his head with a clenched jaw. He shook and wiped tears from his eyes when an emergency room nurse on the witness stand described her lifeless body.
“I do believe that all the science that’s coming up is B.S., and we shouldn’t even be here right now,” he said during a break outside the courtroom. “I think everything should have been done and taken care of a long time ago.”