U. S. Supreme Court to Hear Case on Eyewitness Identification in November

Comments and Observations – Gary L. Wells

Wednesday August 24, 2011


In November, the U.S. Supreme Court will finally return to the issue of eyewitness identification (Perry v. New Hampshire, No. 10-8974). It will be the first time the Court has considered eyewitness identification issues since 1977.

Considering the fact that almost all the scientific research on eyewitness identification has been conducted since 1977 (Wells, 1977), the lapse of time in revisiting the issue is an enormous one. Furthermore, the development of forensic DNA testing for claims of innocence in the last 20 years has shown that 75% of all DNA exonerations are cases involving mistaken eyewitness identification (Innocence Project).

The 1977 case, Manson v. Braithwaite, is in dire need of fixing. Scientific studies show that suggestion is more powerful than the 1977 Court could have anticipated, that three of the five reliability factors listed in the Manson ruling are corrupted by the very existence of suggestion, and that the two-pronged approach of Manson has no significant ability to deter suggestive police conduct  in eyewitness identification (see Wells & Quinlivan, 2009).

“Manson was a reasonable proposition in 1977, but we know much more today. Manson lacks the architecture to serve two functions intended by the court, namely the safeguard against wrongful convictions function and the incentive to avoid suggestive procedures function. Both biological science (via DNA) and social science (via eyewitness identification experiments) have shed new light on eyewitness identification errors and have revealed these errors to be much more prevalent than the 1977 Court could have surmised.” (Wells & Quinlivan, 2009), page 21).


Perhaps most interesting is that the incentive to avoid suggestive procedures is absent from Manson because the use of suggestive procedures actually enhances the witness’s ability to survive the second prong in Manson. The second prong, which trumps any concerns about suggestiveness, asks whether the witness is nevertheless certain, had a good view, and was paying attention. But, research experiments consistently show that suggestiveness actually increases witness certainty, leads them to exaggerate how good their view was, and leads them to claim to have paid more attention (Wells & Bradfield, 1998). Accordingly, when there is suggestiveness, the witness passes the Manson test and the suggestiveness is dismissed as a concern.

Because a “Manson hearing is not going to result in the exclusion of the identification anyway, then why not use suggestive procedures to make sure that the witness not only picks the suspect but also expresses high certainty, reports an exaggeratedly good view, and claims to have paid close attention?” (Wells & Quinlivan, 2009) page 17)


Whereas I applaud the Supreme Court for taking up the issue of eyewitness identification in the upcoming case of Perry v. New Hampshire (No. 10-8974), the Perry case is the wrong case and the issues in Perry are the wrong issues. In Perry, the issue is whether a defendant has a right to a hearing on the reliability of an eyewitness identification when the government itself did not engage in behaviors that caused suggestiveness or unreliability. The answer for the Court should be an easy one based on the Court’s own prior rulings, namely “yes.” After all, the overriding principle guiding Manson v. Braithwaite (and previous rulings by the Court, such as Neil v. Biggers, 1973) is that the ultimate issue is reliability – not whether the source of any unreliability was due to actions of the government.

The Perry case is unfortunate if it is seen by the Court as the “response” to one of the great tragedies of the criminal justice system, namely the conviction of innocent people based on faulty eyewitness identifications. A ruling in favor of the defendant in this case would simply allow for the defendant to have a hearing on the reliability of the identification using the Manson approach. The problem is that the Manson v. Braithwaite (1977) criteria would continue to be used to make that reliability judgment. The Manson criteria and the general approach underlying Manson is fundamentally flawed.

None of the mistaken identifications that have been discovered with DNA testing ever had a chance of being detected as mistaken identifications in pre-trial hearings. Unless the Court goes beyond the narrow issue in Perry, that will remain the case.

The ruling today (August 24, 2011) by the New Jersey Supreme Court is likely to be much more significant than the Perry case. The Henderson ruling is arguably the most sophisticated and detailed reasoning and ruling in the history of any U.S. court on the issue of eyewitness identification.

Nevertheless, I will attend the Court on November 2 for the oral arguments in Perry just in case something important actually happens.