New Rules Being Drawn for Eyewitness Identification


	New Rules Being Drawn for Eyewitness Identification Sept. 23, 2005 Journal Inquirer (CT) By Alex Wood In an effort to avoid wrongful criminal convictions, leaders of Connecticut law-enforcement agencies plan changes in eyewitness-identification procedures more far-reaching than those that the state Supreme Court urged in a decision this week, Chief State's Attorney Christopher L. Morano said Thursday. But Gary L. Wells, an Iowa State University psychology professor who is among the leading researchers on eyewitness identification, says the new procedures don't go far enough. He was especially critical of their failure to include a "double-blind" procedure in which the police officer presenting a photo array to a witness doesn't know which photo shows the suspect. Morano said the development of the new procedures has been going on for months and involves the 13 regional state's attorneys, the Connecticut Police Chiefs Association, and officials in charge of the state-police and municipal-police academies. Even before the Supreme Court issued its decision this week in a case involving the knifepoint robbery of a pedestrian on an East Hartford street, the law-enforcement group had reached a consensus on a set of changes that includes the one urged by the high court, Morano said. That consensus "was not easy to get," he added. While refusing to overturn the robbery conviction of former Hartford resident Laquan Ledbetter, 22, the Supreme Court strongly suggested that police tell future witnesses who are about to view a suspect or a photo array that the perpetrator of the crime "may or may not" be present. If police fail to do so, the court ruled, the trial judge must instruct the jury that the approach taken "tends to increase the probability of misidentification." Psychological research has shown that witnesses tend to pick the person in a photo array who looks most like the criminal. The warning that the criminal "may or may not" be present is designed to reduce that tendency. Morano said the "may or may not" warning was one of the changes the law-enforcement group had decided on before the Supreme Court decision. If police follow the new procedure consistently, he said, there will be no need for any trial judge to give the cautionary jury instruction set out in the decision. Wells described the "may or may not" instruction as "good" but added in an e-mail to a reporter that "it is almost shocking that this minimal effort is just now being taken. "We have known about the importance of this instruction for over 25 years in the published, empirical literature," he added. Standard instructions Under the new procedure, Morano said, a police officer showing a suspect or photo array to an eyewitness will be expected to read a set of standard statements from a form, just as police read the famous Miranda warnings to suspects before interrogating them. He said those statements will include reminders that it is as important to clear innocent people as to identify the guilty and that people's appearance can change due to changes such as hair styles -- and the statement that the perpetrator may or may not be present. In addition, the officer will say that police will continue to investigate the incident "whether you identify someone or not." The form also will include a space for the officer to write down any statements the witness makes while viewing the suspect or the photo array. The witness will subsequently be asked to sign the form to confirm what was said. In addition, the form will include instructions to the officer: not to use words, gestures, or expressions that indicate who the suspect is; to stand outside the field of view of the witness, if practical, in order to avoid giving even subtle, nonverbal feedback; and to avoid making any comment on an identification made by the witness. The last instruction is designed to avoid the following problem, which has been uncovered by psychological research: Witnesses who are told they have picked the suspect from a photo array tend to become more confident in the identification. Judges and juries, in turn, give great weight to the confidence of the witness in deciding whether to believe the identification. Some ideas nixed The law-enforcement group hasn't adopted all the changes advocated by psychological researchers, such as the "double-blind" system, modeled on procedures for scientific research. Such a system is designed to avoid giving even the subtlest cues to the witness as to which choice the investigators think is "right." Morano said Connecticut officials rejected this idea on practical grounds. Because many of the state's police departments are small, it may not always be feasible to bring in a detective who isn't involved in the investigation to handle an identification, he said. Nor is it always practical to ask a witness to come back and make the identification later, the top prosecutor said, adding that it is often necessary to "strike when the iron is hot" when dealing with witnesses. Some have suggested that clerical personnel could be enlisted to handle identifications, but Morano said he would be concerned about having anyone other than sworn police officers involved in a criminal investigation. But Wells found such arguments unconvincing, saying New Jersey has had "absolutely no difficulty" using the double-blind procedure even though it has police departments ranging from two to about 2,000 officers. He said the double-blind procedure has also been adopted in Boston, Minneapolis, North Carolina, and Wisconsin. "There is no possible argument that non-blind lineups are better than blind lineups," he wrote. "Mistaken identification is the primary cause of the conviction of the innocent. The legal system has an obligation to absolutely ensure against the inadvertent influence of the lineup administrator." Another proposal rejected by the Connecticut group was replacing the traditional photo array with a system in which an eyewitness would be shown photos one at a time. Morano said Wells has acknowledged that the scientific studies have produced unclear evidence on this system. Wells wrote that he believes "the sequential lineup has advantages over the simultaneous lineup" but added, "I think it ought to remain a policy choice at this time." Richard R. Brown, a Hartford criminal defense lawyer, said photo identification procedures have improved over the years, but still include at least one inherent source of bias. Because photo arrays consist of police mug shots, he said, the witness knows that everyone shown has been in trouble with the law. That may increase the tendency of witnesses to pick the person who looks most like the perpetrator, he added.
	National Association of Criminal Defense Lawyers (NACDL) 1150 18th St., NW, Suite 950, Washington, DC 20036 (202) 872-8600 • Fax (202) 872-8690 • assist@nacdl.org