Reasonable Doubt: Six reasons eyewitness testimony may convict the innocent

    1 of 1 2 of 1

      Eyewitness evidence may be crucial to convicting criminals. However, according to past and recent studies, misidentification is the leading cause of wrongful convictions. In cases where people were exonerated through DNA evidence, 70 percent were wrongful convictions based on misidentification. Information regarding these studies may be found on the Innocence Project’s website, an organization in the United States aimed at freeing the wrongfully convicted and reforming the criminal-justice system.

      Canada has its own organization, The Association in Defence of the Wrongfully Convicted, with a mandate similar to that of the Innocence Project .

      Perhaps the most well-known wrongful conviction in Canada is that of Thomas Sophonow, who was wrongfully convicted of murdering 16-year-old Barbara Stoppel. Sophonow had an alibi and was not officially exonerated until 2000. Sophonow spent almost four years in jail. An Inquiry was held, with a report released on November 5, 2001.

      In Sophonow’s case, a witness, Mr. Doerksen, had run after the murderer in an attempt to confront him. It was dark, and the lighting conditions were very poor. Doerksen was asked to view a photo lineup, containing pictures of men that fit the description, including Sophonow. However, Doerksen did not identify anyone as the culprit.

      Unfortunately, subsequent to that, he saw Sophonow in a public building after he had read a newspaper article with Sophonow’s picture. Unwittingly, and subconsciously, Doerksen began to believe that Sophonow was actually the man he had chased. This example highlights the malleability of human recollection: people not only begin to forget information but they are also susceptible to forming false memories based on new information learned since a particular event has occurred.

      Further complicating the issue was the fact that pictures used in the photo lineup included a picture of Sophonow that stood out dramatically from photographs of other men, making it “startlingly unfair”. Further still, witnesses had been unable to identify Sophonow in a live lineup, noting that he only looked more similar to the perpetrator than the other men.

      The inquiry report contained numerous recommendations to address concerns with eyewitness identification and the use of photo lineups. Guidelines similar to the Sophonow recommendations have been adopted as law in 14 states. This is not the case in Canada, where failure to follow the recommendations may not affect the judge’s view of trial fairness. This is most unfortunate, given the great risk of wrongful convictions.

      Here are six reasons eyewitness evidence may convict the innocent, and how those risks can be minimized.

      1. Estimator variables

      Estimator variables include factors of the observation that cannot be controlled, such as lighting, distance, obstructions,a  witness's eyesight, whether the witness was stationary or distracted when viewing the suspect, whether the witness has any familiarity with the suspect, and any other factors related to the ability of the witness to see clearly. It also includes the malleability of witness recollection.

      A thorough inquiry into how great a role these factors play in each case is necessary, and a witness’s testimony should be weighed accordingly. Even credible witnesses may be mistaken, and, unfortunately, confident witnesses may unfairly influence the jury. The safest way to guard against this danger is to ensure there is reliable corroborating evidence, including forensic evidence such as DNA.

      The remaining reasons relate to system variables: those variables that can and should be controlled by police agencies attempting to bolster eyewitness evidence.

      2. Whether or not the administrator of the photo lineup knows which photo is the suspect’s

      The administrator of a photo lineup should not know which photo is the suspect’s. This is extremely important because the administrator can give away subtle clues, causing the witness to figure out which photo is the “right one”. Witnesses feel pressured to “get it right”, and they may feel their job is to pick out the right person. This may cause them to be hypersensitive to cues from the administrator, whether they are consciously aware of it or not. 

      In the alternative, an envelop-shuffling method may be used, where individual photographs are placed in envelops and then shuffled, so that the administrator does not know which photo is being viewed at any given moment.

      It is also important that the photographs are shown sequentially, not all at once, so the witness is not simply comparing them to one another to try to pick out the “best match”, when the actual perpetrator’s photograph may not be included. Seeing the photographs sequentially also helps prevent the witness from confusing specific details in the witness’s mind. Memory can be influenced by suggestions in other photographs such as distinguishing facial marks or features. For example, in one case a witness thought she had seen a scar at the time of the incident. It turned out she had not noticed the scar until after the photo lineup. This mistake tainted the usefulness of her evidence.

      3. The composition of the photo lineup

      Perhaps the most important aspect to the fairness of the procedure is the composition of the lineup: the quality and characteristics of the suspect’s photo compared to the recommended nine other photos used as “fillers”. How well matched are the other photos to the suspect, or to the description of the suspect? Does the photograph of the suspect stand out in some way, like in Sophonow’s case? For example, in one case the suspect’s face in his photograph was larger than the faces in the other photographs, making the suspect’s picture stand out compared to the rest.

      In another case, an officer noted that a software system was used to compile the lineup. He testified that he was not sure whether race was a descriptor the program took into account. He explained using the descriptors of similar hairstyle, hair colour, and eyes, which resulted in limited results. He testified that he had to expand the number of photographs available to him, and he ended up using some photographs of men that had a lighter skin colour than the witness’s key identifying trait of “olive-coloured skin”.  The result was that the witness was able to quickly rule out most of the photographs.

      Another case highlights the difficulty with using older photographs if more recent ones are not available. In this case, an older photo of the suspect showed him with a goatee and dreadlocks in his hair, and the witness was not able to identify him as the perpetrator.  However, when the photo lineup was repeated with a more recent photo showing the suspect without facial hair and an almost shaved head, she was able to pick him out.

      4. The Instructions given when implementing the photo lineup

      To reduce pressure on the witness to pick out “the right guy”, the following instructions should be conveyed: that the suspect’s photograph may or may not be present, that the administrator of the lineup does not know which photo, if any, is the suspect’s, and that the investigation will continue regardless of whether the witness identifies the suspect.

      5. Confidence statements.

      Confidence statements have two aspects to them. The typical confidence statement is when the administrator obtains a comment from the witness regarding his or her confidence level that he or she has selected the perpetrator. The other side of the coin includes any confirmatory statement provided by officers that the witness picked out the right person. Confidence statements from witnesses are necessary and useful to assessing the reliability of the eyewitness evidence, as well as what weight the jury should place on a correct identification of the suspect. Confidence statements where officers confirm to the witness they chose the suspect’s photo should be completely avoided because of the risk of reinforcing a misidentification.

      6. Whether or not administration of the photo lineup is recorded.

      This is perhaps the most practically useful way to protect against misidentifications. Having the administration of the lineup audio- and video-recorded is the most effective way at trial to analyze the efficacy of the procedure and determine what weight to apply to a witness’s identification of a suspect.

      A word of caution: you should not act or rely on the information provided in this column. It is not legal advice. To ensure your interests are protected, retain or formally seek advice from a lawyer.