On his first, full day officially on the job, the new Pittsburgh public safety director Stephen A. Bucar issued a statement Wednesday saying that the city police bureau will not adopt practices on eyewitness identification demanded by Allegheny County District Attorney Stephen A. Zappala Jr.
“Although the model is endorsed by certain academic and research facilities, there is dissenting opinion in those same communities that disputes the conclusion that the best practice for eyewitness identification procedure lies with a sequential process rather than a simultaneous process,” Mr. Bucar said.
The issue revolves around how eyewitnesses are shown suspect photographs. In a simultaneous process, typically a set of six or eight photographs of similarly looking suspects is placed on a single page shown to the witness. In a sequential process, the witness is shown each individual suspect’s photograph one at a time.
Earlier this year, the Allegheny County Chiefs of Police Association adopted a new model policy for identification using sequential photo lineups, and last week Mr. Zappala sent a letter to city officials telling them that his office would only prosecute suspects where the evidence against them is such an ID. This district attorney said his office would not prosecute cases in which the identification came solely from a simultaneous photo array.
Mr. Zappala's letter referenced a Pittsburgh Post-Gazette story outlining a federal lawsuit filed earlier this month by a man wrongfully arrested for two armed robberies in 2012 based on faulty identification. That suit was the second in four months questioning arrests by the Pittsburgh robbery squad, which used identification as the main — or only — source of evidence.
Although the district attorney noted in that letter that a mistaken identification does not indicate police misconduct, stories like that one “erode confidence in our police and in the quality and confidence in police investigative processes. Obviously, this cannot occur.”
In his statement in response, Mr. Bucar, a former FBI special agent, said the U.S. Department of Justice’s Bureau of Justice Assistance has not concluded that the sequential procedure is more reliable, and that there is debate in the academic community about whether a sequential process is better.
“Although Mr. Zappala appears to have concluded that sequential is superior to simultaneous procedure, as long as there is credible dissent and disagreement, the Pittsburgh Department of Public Safety is not prepared to endorse one over the other. In the interim, we are evaluating aspects of both, and are utilizing those that are not in dispute in the spirit of ensuring that innocent people are not wrongfully accused while the guilty are properly identified,” Mr. Bucar said.
“While I can appreciate Mr. Zappala’s concern, I am compelled to proceed with caution on this matter since my experience dealing with professional associations and with the Department of Justice makes me hesitant to endorse an ill-informed position.”
Mike Manko a spokesman with the DA’s office, said that while Mr. Zappala is “interested in the opinion of the public safety director,” he “has concluded that the procedures being advocated by our office are the best way to ensure confidence in the law enforcement investigative process and the best way to protect victims and protect the rights of the accused.
”For reasons that are obvious, a particular police agency cannot be permitted to tell a prosecutor what the evidence is that they will use to determine somebody’s liberty.“
While Mr. Bucar cited debate in the scientific community, Nancy Steblay, a leading eyewitness scientist from Augsburg College in Minneapolis, said that’s not the case anymore.
A 2011 review she co-authored reported that the sequential process is superior.
“We have over 70 studies we’ve reviewed and verified, and you see the pattern over and over again for sequential,” Ms. Steblay said. “In an enormous set of studies, you’re always going to find some outliers. But that’s what science does — look for the pattern.
“Although it’s not perfect, it’s far superior — significantly superior — to the simultaneous procedure.”
Using a sequential process pushes a witness toward a more absolute judgment by comparing the memory of the suspect to a single photo at a time, she said.
“It is a more conservative procedure, which means the witness is less likely to make an error,” she said. “If the witness makes an identification, this is more likely to be of a guilty suspect versus an innocent suspect.”
In a 2012 national survey completed by the Police Executive Research Forum, about 32 percent of those law enforcement agencies who responded use the sequential process.
Ms. Steblay attributes the 68 percent to jurisdictions not being in touch with the problems associated with the simultaneous process.
“I suspect a number of them are not even familiar with the issue,” she said.
But many jurisdictions — including entire states like North Carolina, New Jersey and Connecticut — have mandated the use of the sequential procedure.
According to the Innocence Project, eyewitness misidentification was a factor in 73 percent of post-conviction DNA exoneration cases.
Ms. Steblay did say that the National Academy of Science is working on a broader report regarding eyewitness evidence research, including memory fallibility and policing procedures, that is due out in coming months.
In his letter to city officials, Mr. Zappala said that, as of Aug. 16, if the new model procedures aren't used, “the identification may be considered as establishing a suspect, but may not be considered in and of itself for charging purposes.”
As part of his statement, Mr. Bucar also questioned the process by which Mr. Zappala relayed his concerns.
“I would have expected a phone call from the District Attorney’s Office prior to learning of his position in a newspaper article. My office did not receive his letter in the U.S. mail until Friday, July 25th, after the existence of the letter was reported and the content paraphrased in the media on Thursday, July 24th.”
Paula Reed Ward: email@example.com, 412-263-2620 or on Twitter: @PaulaReedWard.