On Monday, November 22, the Florida Innocence Commission met to dig into the issue of eyewitness misidentification. The Commission heard from a number of witnesses, including Professor Gary Wells, the godfather of scientific research into witness memory and eyewitness misidentification. The Orlando Sentinel reports:
To prove how vulnerable eyewitnesses are to making errors, Wells showed the panel a video of a man dropping a bomb down a building’s air shaft. He then showed them a lineup of six suspects and asked panel members to pick the bomber.
About one-quarter of them picked a suspect, and each was wrong. That’s because the bomber wasn’t among them.
Wells has studied eyewitness misidentification for more than 30 years and has worked with law-enforcement agencies across the nation to minimize those mistakes. His conclusion: Eyewitnesses get it wrong 20 percent of the time.
Obviously, where people who have an intent to subvert justice taint witness identification procedures, we cannot have any confidence in the identifications made in those procedures and there just isn’t much we can do to stop such bad faith. But the vast majority of the time, you have both law enforcement personnel and witnesses who are acting in good faith to try and identify the true perpetrator of the crime, yet they still get it wrong despite their best efforts. Dr. Wells’ work has tried to figure out why and the answer has as much to do with the fallibility of human memory as how lineups and photo arrays are prepared and administered to witnesses. The St. Pete Times notes:
The problem with eyewitness identification comes down to this, said Iowa State University psychology professor Gary Wells: Memory, like physical evidence, can be tainted. One way that happens is through the standard police lineup. Traditionally, a witness will look at half a dozen mug shots at the same time and try to identify a suspect. People tend to pick the person who most resembles the culprit, Wells said, even if the actual criminal isn’t depicted. He said witnesses should get just one photo at a time. Studies show that forces people to dig deeper into their memories. People are also influenced by cues they get from those showing the photos, he said. Detectives may unknowingly give nonverbal cues. Or they’ll say things.
For example, “Now, take your time” might mean the witness is looking at the wrong face. In one case, a witness said detectives applauded when she picked their suspect. That kind of affirmation can create an inflated sense of certainty in the mind of a witness, who will take it to trial.
The solution? Don’t let detectives conduct the lineup, Wells said. Give the photos to someone who doesn’t know which depicts the suspect.
Many law enforcement agencies, large and small, urban and rural, around the country are performing witness identification procedures with the new, suggestion-proof methods proposed by Dr. Wells, with wonderful success. A representative from the Hillsborough County Sheriff’s Office even testified at the meeting that his agency is performing sequential, double blind lineups and photo arrays and seeing more accurate outcomes leading to swifter justice.
So what is not to like? Well, for some on the Commission, whether Florida will adopt a policy that conforms with the national trend and the available science comes down to economics. Some say that smaller agencies would have to use already limited manpower to do lineups and would have to make both the preparer of the lineup and the administrator of it available for depositions, court hearings, and trial, which could be an expensive proposition. But there are ways to prevent this such as drafting a statue that requires sequential double blind administration with exceptions for impracticability that allows the agency to simply blind the administrator through other means, such as use of laptop computer technology or the folder-shuffle method. Ohio has enacted a statutory mandate that does just this.
Others, like the Florida Police Chiefs and the Florida Department of Law Enforcement, favor just letting law enforcement agencies individually figure out what is best. This is obviously a bad idea as it would create a lack of uniformity and, in turn, a lack or justice from jurisdiction to jurisdiction. Plus agencies have already had the opportunity to create policies with examples in North Carolina, Ohio, Wisconsin, New Jersey, and Massachusetts as good guides. Yet, in response to two public record requests sent by IPF to over 350 law enforcement agencies in Florida, only 37 agencies that responded even had a written policy on eyewitness ID procedures, and an even smaller fraction of those were even close to what Dr. Wells and the rest of the scientific community have suggested as a best practice.
Lastly, we have the obligatory idiotic statement from an ignorant prosecutor (from The St. Pete Times):
Bruce Bartlett, chief assistant state attorney for the Pinellas-Pasco State Attorney’s Office, said he is troubled by what he sees as an unfair assumption.
“It suggests that your law enforcement is dishonest. And I’m sorry, but I’m not ready to accept that,” he said. “I’m not saying mistakes don’t happen, because they do. But I’m not ready to believe your cops are all crooked.”
This is the typical straw man. No one has even suggested law enforcement is dishonest. That was not the tenor of the Commission meeting at all. Bartlett should call up the Sheriff’s office in neighboring Tampa and ask them whether they proactively implemented the widely-accepted best practices because they thought their own sergeants and deputies were dishonest. Or maybe they did it because the best practices diminish unintentional suggestiveness that seeps into ID procedures even when everyone is well-intentioned. Considering Bartlett was not even at the Commission meeting to hear the testimony, he probably should have not even opened his mouth to save himself the embarrassment.
In any event, it appears that the Commission is poised to do something positive in this area and we will keep you informed as the result comes more into focus.