a bird and a bottle

The Trouble With Eyewitness ID’s
March 1, 2007, 8:38 am
Filed under: criminal justice, law

In a mind-boggling 75% of the convictions overturned by DNA evidence, eyewitness IDs were the basis for the original conviction. Witnesses often get it wrong, either because so much time has passed since the alleged offense and the trial, or because the defendant may look like the person the eyewitness saw (this is paritucularly a problem across racial groups). This unreliability seems to work only against the defendant; if the eyewitness description doesn’t quite match the defendant, the jury may assume that because of the speed of the events, it’s ok that the eyewitness description was a little off. Often, the jury takes an ID at face value. When witnesses err, the ramifications can be immense — juries often find eyewitness IDs very persuasive, especially in the face of a shaky alibi, or if the defendant does not take the stand.

Happily, states are starting to address the issue. Eyewitness ID Reform Blog reports that ID reform bills have passed in West Virginia and Maryland, and that a bill is pending in New Mexico. While the bills do not bar the use of eyewitness IDs (not a realistic solution), they do instruct the police and prosecutors on best practices for identifications, preferring sequential lineups and implementing DOJ guidelines.

These bills won’t solve the problem, but they’re a good start.


4 Comments so far
Leave a comment

London, UK, 1987: three muggers assaulted me on the street while I was returning home from an evening class at LSE. It was a slash attack using box cutters that landed me in hospital. Six months later, I returned to NYC and put the incident out of mind until I received a call one day from Metropolitan Police: Would I be willing to return to London to identify a suspect?

I remembered the night of attack much less vividly than the day of the line-up. More ambush than mugging, it had occurred suddenly on a dimly lit street. A passing police car interrupted the orgy of violence; the muggers scattered into the shadows. Deus ex machina, it was a lucky moment that probably saved my life.

In contrast, the line-up took place in a well-lit open room with full-height windows, no glass separating me from the line of suspects. I looked at each face to face but could not identify my attacker with certainty. A detective sergeant paced behind me and whispered, “Second from the right” in passing. Still, I could not say for sure.

On the return flight to New York, I felt utterly defeated and humiliated and deemed myself a failure.

Comment by Swampcracker

The most insidious aspect to identification evidence is that, while it is often the weakest evidence in a case, it is most often the linchpin in the jury’s evaluation of a case, as jurors tend to place tremendous (and undue) weight on identification testimony.

In an important ruling six years ago, the New York Court of Appeals decided, for the first time, that where identification is an issue, a trial court would have discretion to allow the defense to call an expert on psychological studies that show how certain factors can impact the reliability of an identification. For example, stress factor, weapon focus, cross-race ID, post-event assimilation of information, deteriorating retention time, and the lack of any correlation between confidence and accuracy can all render an identification less reliable than it might otherwise appear. This last factor is particularly noteworthy — when a seemingly credible witness points to a defendant and says “That’s him. I will never forget his face,” a jury would be hard-pressed not to convict. Yet, counterintuitively, the witness’s confidence in the accuracy of the identification has no relationship whatsoever to its accuracy.

These factors have been studied extensively for more than 20 years. Yet, recently, the D.A.’s office in New York have presented charlatan “experts” to try to debunk these well-established criteria for evaluating the reliability of IDs. Some trial courts have bought into this cynical approach, and have ruled that, despite the voluminous studies, the proffered defense expert evidence is not “generally accepted in the scientific community” and is therefore inadmissible. In a case argued February 14 before the Court of Appeals, the court will decide whether such identification expert evidence passes the test for admissibility.

Only when the jury can fully understand the inherent weaknesses in identification evidence, and only when identification procedures are conducted fairly, can we hope to reduce significantly the incidents of wrongful convictions based on mistaken identifications.

Comment by RickyVermont

Swampcracker – your comment exposes something important that I failed to mention: the pressure on witnesses to ID someone at a show-up or lineup. As you discuss, people may feel defeated if they cannot ID someone, even in very different conditions than that of the offense. I think it’s unusual though that people, under the pressure you experience, stand their ground….

Ricky – I’m disappointed to hear that things seem to be getting worse though some would argue they have gotten better. And I agree with you that things won’t really change until there is jury buy-in. My question is: can jury instructions do that (or at least help)? Or does it require something more?

Comment by bean

Poor Swampcracker. Sounds like a horrible experience, both at the time of the incident and, later during the ID procedure. While I have seen many cases in which the police procedures are suggestive, even highly suggestive, it is the rare to see evidence of the kind of blatant misconduct as occurred in your case.

But your experience highlights the difficulties in ever making a reliable identification, unless it is virtually immediately after the incident. You, at least, did not succumb to the natural inclination to want to make an ID, and to select the one of the group that most resembled the perpetrator. The two most sought-after reforms in ID procedures are sequential lineups (where you don’t see all the individuals togther, but only one at a time) and double-blind lineups (where the officer who conducts the lineup with the witness does not which of the individuals is the suspect). The latter reform would have prevented the outrageous attempt by the officer to “assist” you.

Bean, in NY judges do often give “expanded” identification instructions, stressing some of the factors to consider — like lighting conditions, amount of time to observe, etc. — but they do not give instructions on the kinds of factors that I mentioned earlier — the psychological factors like stress and cross-race. For example, studies have shown that people of all races have a much harder time making an accurate identification of an individual of a different race, without regard to the witness’s bias or lack of bias. Jurors just don’t get these concepts without an expert to explain them, especially since many of these factors are counter-intuitive. Hopefully, the use of ID experts (as well as false confession experts — but that’s for a different article one of these days!) will be permitted in greater numbers of cases in the days ahead.

Comment by RickyVermont

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: