The Colorado Supreme Court appeared largely comfortable on Thursday with the recently revised definition of “reasonable doubt” that is now part of the template jury instructions for criminal cases.

To date, multiple Court of Appeals decisions have endorsed the constitutionality of the new phrasing, which resembles the instruction given in Colorado’s federal trial court. U.S. Supreme Court Justice Ruth Bader Ginsburg endorsed the wording in 1994 as a “clear, straightforward, and accurate” definition of reasonable doubt for jurors.

In evaluating the revised language, the state Supreme Court chose to review the case of Alexis Teran Sanchez, who Adams County jurors found guilty of three hit-and-run-related offenses. On appeal, he argued the new reasonable doubt instruction violated his rights to due process and a fair trial.

However, his lawyers employed an unusual method of challenging the language. In significant part, they relied on a study that they funded through the University of Denver’s Department of Psychology. The simulated jury verdicts suggested that the current language may slightly increase conviction rates over the previous instruction.

During oral arguments, Justice Richard L. Gabriel said it gave him “a little bit of heartburn” that Teran Sanchez’s lawyers commissioned the study in connection with their case.

“I don’t mean to impugn the integrity of these professors at all, or your firm,” he said. “I suppose what I’m more concerned about was the fairness of that to the district attorney. We see this study for the first time” on appeal.

Justice Susan Blanco also noted that the study participants were seemingly given a definition of reasonable doubt, but without the full battery of instructions that criminal jurors receive in a real trial.

“You’re studying this one instruction by itself and there is no context for how does that translate to the courtroom and what’s going on in an actual trial,” she said.

Prior to 2023, Colorado’s template jury instructions for criminal cases contained a longstanding explanation of reasonable doubt: A doubt based upon reason and common sense, which was not vague or speculative, that would “cause reasonable people to hesitate to act in matters of importance to themselves.”

Then, in January 2023, the judicial branch announced that a small group of judges had voted to change the language of the model instruction. The committee included two members of the Supreme Court — Justice Carlos A. Samour Jr., who chairs the panel, and Blanco, who was then a Larimer County trial judge. The vote was not unanimous, although the committee did not disclose who voted to revise. Both Samour and Blanco are participating in Teran Sanchez’s appeal.

The instruction now advises jurors to vote to convict if they are “firmly convinced of the defendant’s guilt.” The prosecution has not proven guilt, the instruction continues, “if you think there is a real possibility that the defendant is not guilty.”

Defense attorneys were concerned the change lessened the prosecution’s burden to prove a defendant guilty. Although the state’s Court of Appeals issued multiple decisions upholding the new wording as constitutional, two appellate judges have recently suggested that the “real possibility” phrasing unconstitutionally assumes the defendant’s guilt, among other problems.

To the Supreme Court, Teran Sanchez’s attorneys leaned on the DU study they arranged, which featured 896 U.S. citizen participants recruited online. The participants read about a crime and received five variations of the reasonable doubt instruction. Those included the version from Teran Sanchez’s trial, the old instruction, and two versions that the U.S. or state supreme courts have found unconstitutional.

The authors, Assistant Prof. E. Paige Lloyd and research consultant Abigail J. Langeberg, found that participants were slightly more likely to convict using the new language compared to the old, and more likely to convict than with the two unconstitutional instructions. However, not all of the results were statistically significant.

“Empirical data is the type of thing this court regularly looks to,” argued NoahLani Litwinsella, an attorney for Teran Sanchez. “The empirical data demonstrate at least a reasonable likelihood that the jury understood it was able to convict on proof of less than reasonable doubt.”

“We’re talking about people sitting at home on the Internet, on their couches, presented with a particular fact patten,” countered Todd Bluth for the prosecution, and “only one jury instruction. We present one jury instruction, this reasonable doubt instruction, without any of the context. And that is in direct contradiction to the manner in which we evaluate jury instructions in Colorado, which is holistically in context.”

Teran Sanchez’s attorneys also argued the “modern usage” of key phrases in the instruction would lead ordinary jurors to interpret “real possibility” and “firmly convinced” to be less favorable to a defendant’s rights than they are intended to be.

“If we were to look at your approach, how often would we have to go through this exercise?” asked Justice Maria E. Berkenkotter. “Because language, quite frankly, always changes, and it has probably never changed faster than it does now with social media, the Internet.”

The justices asked substantially fewer questions of the prosecution, but Gabriel wondered whether trial judges are using the new instruction in large numbers.

“I have heard some judges say, ‘I’m not quite sure. I know this prior instruction has been definitely approved by the court, so I’m going the safe road there,'” said Bluth, adding that the Court of Appeals’ decisions upholding the new language have encouraged trial judges to switch.

The oral arguments occurred at Holyoke Junior/Senior High School in Phillips County as part of the Supreme Court’s “Courts in the Community” program.

The case is Teran Sanchez v. People.