COLUMBUS, Ohio (WCMH) – A man’s plea to prevent eyewitness evidence from being used at his murder trial has been dismissed.

After 16 months, 50 pages of arguments and two hearings, a Franklin County judge denied a motion to suppress eyewitness identifications used to arrest the man accused of murdering an Ohio State University student in 2020. His attorneys argued the two identifications – one in person shortly after the shooting and one during a photo line-up the next afternoon – violated Columbus police protocol, were suggestively obtained and too unreliable to be admissible in court.

In a five-page decision, the judge sided with the prosecution, ruling the suspect identifications will be given to a jury to decide whether they are reliable enough to convict him. Neither the defense nor prosecutor’s office responded to requests for comment.

Kintie Mitchell was arrested about an hour after 23-year-old Chase Meola was fatally shot outside a party near Ohio State’s campus on Oct. 11, 2020. Mitchell, then 18, and three other men were detained a few blocks from the shooting, minutes after police gathered initial descriptions of the shooter as a Black man of average height with shoulder-length dreadlocks.

Mitchell and the other men were individually shown in a “show-up” identification to four witnesses, one of whom identified Mitchell as the shooter and the others as with him during the shooting. About 12 hours later, a second witness identified Mitchell out of a six-photo array. On Oct. 21, a grand jury indicted him on two counts of murder and one weapons violation.

Previous rulings on evidence admissibility require courts to use a two-pronged test: First, the court must decide whether the way the identification was obtained was so “impermissibly suggestive” that there is a “substantial likelihood of misidentification.” If that prong is satisfied, the court must evaluate the evidence’s reliability using several factors, including the witness’s ability to view the offender at the time of the crime, the accuracy of their prior suspect descriptions and their level of certainty.

‘The scene was chaotic:’ The defense’s case against police procedure

The defense argued that officers failed to follow protocol during the show-up identification. Describing in court documents the “chaotic” scene – the area still teeming with party-goers and potential witnesses yards away from Meola’s body – Mitchell’s attorneys noted that police weren’t able to separate witnesses before interviewing them. 

When police did separate four witnesses, the one who positively identified Mitchell was first allowed to sit with his girlfriend in a police car and make phone calls to family, both in contrast to normal protocol.

Prosecutors argued that the witness’s girlfriend was allowed to sit with him because she “could not identify the suspect and would not be a witness.” Footage from inside the police cruiser showed that the couple didn’t discuss the shooting before the witness was taken to see potential suspects.

Further Columbus police policy violations include the witness not being given instructions before the show-up, not being told he didn’t have to identify one as the suspect; suspects being presented in handcuffs with no apparent attempt to conceal the handcuffs, and officers failing to document the order the detained men were presented to the witness. And the defense argued the witness may not have discussed the shooting with his girlfriend in the car, but when asked to initially describe the shooter, his girlfriend interrupted him on multiple occasions to correct his statement.

The second identification from a photo line-up should also be thrown out, the defense argued, as the witness offered differing, vague descriptions of the shooter – he wore a white shirt, then a dark shirt and had “dreads, like an afro,” for instance – and was not asked whether he talked with any witnesses or saw social media posts about the shooting. The defense also contended that Mitchell’s headshot was slightly smaller than the others, naturally drawing the eye toward him, and three of the six men had short hair, despite multiple witnesses describing the shooter as having shoulder-length hair.

The prosecution conceded to most of the defense’s accusations but argued the policy violations weren’t egregious enough to contaminate the witnesses’ identifications. Mitchell matched the witnesses’ basic descriptions of the shooter, and court precedent specifies that identifications made off basic suspect descriptions, so long as they include things like race, gender, age and hairstyle, are admissible in court. Both witnesses were also fairly certain of their identifications.

Prosecutors dismissed the defense’s claim about photo size, finding any difference negligible and not inherently suggestive. The state pointed to legal precedent allowing for variability in photo arrays, including photos of people who don’t exactly match a suspect description.

Judge Jeffrey Brown offered no explanation as to why he found the state’s argument about the first witness more compelling, ruling only that, “The court does not find that any of these items when considered alone or together, make the identification process impermissibly suggestive.” His opinion makes no mention of the photo line-up identification.

The unreliability of eyewitness identifications

Because the first prong of the legal test wasn’t satisfied, the judge did not have to consider the reliability of the identifications — and he didn’t. In court filings, the defense acknowledged such procedural restrictions but argued it matters nonetheless.

Mitchell’s attorneys described the scene of the shooting: It was 2 a.m. outside a party where witnesses admitted to drinking alcohol. A fight began shortly before the shooting – and it happened within seconds. The shooter pointed his gun at someone else before ultimately shooting Meola in the head.

It was dark and suspect descriptions were vague. The witnesses saw an incredibly traumatic event, the defense argued, citing multiple research studies indicating lower witness reliability during high-stress situations. Most significantly to the defense, the majority of witnesses to Meola’s death were white, and the shooter was Black.

The reliability of eyewitness identification has been tempered by scientific evidence showing its faults. The defense cited the Innocence Project, which has found that in more than two-thirds of DNA-based exonerations, people were convicted using incorrect eyewitness identifications. More than half of those eyewitness identifications were made using a photo array. In other studies, a witness’s certainty in identification was shown to have no impact on their accuracy.

Cross-racial identifications have even lower accuracy rates; the defense cites several studies indicating such inaccuracy, including one showing that people made four times as many mistakes trying to identify a suspect of another race versus identifying a suspect of their own.

To consider either identification’s reliability, as both the defense and prosecution noted, the judge would have had to abandon court precedent. Prosecutors argued any question of reliability should be left to a jury to consider, with cross-examination a sufficient safeguard against unreliable testimony. Yet the defense made its case that the judge should weigh the science more heavily than legal precedent established before eyewitness identification was thoroughly studied.

“There have been many developments in the science of eyewitness identification since the landmark U.S. Supreme Court cases mentioned above were decided. These scientific studies show that eyewitness identification is not as reliable as has been previously thought,” the defense wrote in their motion to suppress evidence. “Despite these advancements, Courts are still reluctant to suppress unreliable eyewitness identification.”