US court puts reliability of eyewitnesses under scrutiny
Witness testimony has been central to countless criminal trials, and the “dock identification” by a witness of the alleged offender has always been a moment of high drama. Dock identification has long been abolished – of course, the witness would point out the defendant, in the dock, flanked by security – and video ID parades are now the preferred method of testing eyewitness reliability, as set out in the Police and Criminal Evidence Act code D.
However, the New Jersey court has become the latest institution to question this kind of evidence, noting a “troubling lack of reliability” in eyewitness identifications. The unanimous judgment by the state court says that the tests for eyewitness reliability, set out by the US supreme court 34 years ago, should be revised.
The judgement’s wider significance is that it draws upon the latest scientific studies. The Innocence Project, which submitted evidence to the New Jersey court, suggests as many as a third of the 75,000 witness identifications that take place in the US every year may be incorrect.
The New Jersey court judgment stems from the case of Larry Henderson’s 2004 conviction for reckless manslaughter and weapons possession.
Where a defendant questions police practice in identification, it sets out guidelines on the many factors that could affect reliability and which the judge must outline to jurors. These factors include:
• Whether the lineup procedure was administered “double blind”.
• Whether the witness was told that the suspect may not be in the lineup.
• Whether the police avoided providing the witness with feedback that would cause the witness to believe he or she selected the correct suspect.
• Whether the witness was under a high level of stress.
• Whether a weapon was used.
• How much time the witness had to observe the event.
• How far the witness was from the offender, and what the lighting conditions were.
• The length of time between the crime and identification.
• Whether the case involved cross-racial identification.
The issue will be examined by the US supreme court in November. It could have serious implications for the way the many US law enforcement agencies handle eyewitnesses.
In the UK we have seen that eyewitness identification is not always as reliable as we might like to believe. In the case of Barry George, convicted of the 1999 murder of Jill Dando, two witnesses who had failed to pick him from a lineup were given a lift home with a third who had. The two subsequently made statements identifying him.
Witnesses to the events shortly before police shot dead Jean Charles de Menezes described seeing a man in a bulky jacket vaulting a ticket barrier. In fact he did not jump the gate, was wearing a light shirt which could not have concealed any bomb, and only ran when he saw his train pulling into the station.
Perhaps our own judges in the UK need to make an effort to unpick the tricks memory plays on us.
David Banks is a media law consultant and co-author of McNae’s Essential Law for Journalists
guardian.co.uk © Guardian News & Media Limited 2010