Leading Cause of Wrongful Convictions: Mistaken Identification by Eyewitnesses

By: Houston Criminal Attorney John Floyd and Paralegal Billy Sinclair


There have been 251 innocent people exonerated in this country by DNA evidence over the last two decades. The most disturbing aspect of this phenomenon of “convicting the innocent” is that more than 75 percent of those convictions involved mistaken identifications (according to the New York-based Innocence Project)—one or more witnesses pointing a finger of guilt at the wrong person. What is even more disturbing is that at least one-third of these mistaken identification cases involved two or more witnesses.


The lesson in these shocking figures is that what people see, or believe they saw, is not always reliable. This is especially true when the witness identification procedure is corrupted by rogue cops deliberately trying to frame innocent individuals. That’s what happened to Donald Wayne Good who, on June 18, 1983, was arrested, charged, convicted, and sentenced to life imprisonment in Irving, Texas for an aggravated rape, aggravated robbery, and burglary of a habitation for which he did not commit.


Beyond a doubt someone did break into the home of “Jane Doe,” bound her and her eight-year-old daughter, and forced Doe into a bedroom in the home where he raped the mother. The local Irving Police Department arrived after and processed the crime scene. The rape victim was taken to a local hospital where a “rape kit” examination was performed. After this examination was conducted, the victim met with investigators at which time she described her attacker as a white male in his mid-20s, six feet tall, weighing 190 pounds, clean shaven, with a dark tanned medium or large build, and blondish-brown hair. Based on this description, a police sketched artist prepared a “composite sketch” which was distributed throughout the Irving Police Department.


This is where Irving police detective Fred Curtis came into the picture. One of the detectives assigned to the Doe rape investigation, Curtis had been investigating a number of other “unsolved daytime burglaries” in the area. Curtis believed Good, who had been arrested three days (and subsequently released) before the Doe rape for bond forfeiture of a previous DWI arrest, was involved in the daytime burglaries. The detective called Good into his office to interview him about the string of burglaries. The interrogation didn’t go well for Curtis because Good refused to cooperate with the investigation. At this point in the interrogation the detective snatched up the composite sketch of the Doe rapist he had just received and told Good he looked “somewhat similar” to the rapist. And the detective then threatened Good by telling the suspect that he could “fix it” to make sure Good looked just like the Doe rapist if he didn’t cooperate. Good still refused to cooperate.



The rogue detective was true to his threat when Good continued to refuse to cooperate with the burglary investigation. At that point Curtis terminated the investigation and told Good he (the detective) was going to get Good charged with the Doe rape. The detective took several photographs of Good at the police headquarters, repeatedly altering the “light settings on the camera” to make sure Good had a “dark tan” in the final photograph. The detective then added the doctored photograph of Good in a photo lineup he presented to Doe and her daughter. The daughter was unable to make an identification but Doe “immediately” picked Good’s photo from the lineup array. Good was immediately re-arrested for the rape of Jane Doe.


The Dallas County Sheriff’s Department then conducted a live lineup that featured Donald Dwayne Good. Each lineup participant was asked to state specific phrases the Doe rapist had used during the home invasion. Jane Doe once again identified Good as her assailant and this time the daughter also identified Good as the home invader. Good was put on trial but the jury could not reach a verdict. The state tried Good a second time and he was convicted on all charges, although this conviction was reversed for “prosecutorial misconduct” unrelated to the lineup misconduct. Good was tried a third time and convicted of burglary of a habitation with intent to commit rape. At all three trials the mother and daughter “positively” identified Good as their assailant.


In 2001 a Dallas County criminal court judge granted a motion filed by Good for post-conviction DNA testing of biological evidence recovered at the crime scene. The DNA testing exonerated Good as the assailant, and in November 2004 the Texas Court of Criminal Appeals ordered all charges against him dismissed. One month later a Dallas County criminal court judge dismissed all the charges against Good and ordered his release from custody. He had spent 19 years in prison for the frame up. Good promptly sued Detective Curtis and a number of other officials connected with his wrongful conviction. While the Fifth Circuit Court of Appeals recently turned back efforts to hold the county officials liable in the case, the appeals court did rule Good’s claim against Curtis could go forward.


Like a significant number of other mistaken identifications in the DNA exoneration cases, the mistaken identification of Donald Dwayne Good was caused by police misconduct in the lineup process. These are not rare incidents. Law enforcement officials routinely manipulate lineups, particularly those involving the “photo array,” to secure an identification of a suspect upon whom they have “focused” their investigation. In the Good case, Detective Curtis, according to the Fifth Circuit, doctored Good’s photo by repeatedly changing the camera’s light settings so Good would have the “dark tan” which had stuck out in Jane Doe’s description of her assailant.


But there are other ways the police can manipulate witnesses into mistaken identifications by “reconstructing” false memories about the crime. In a recent article (Jan. 6, 2010), Scientific American reported that most Americans have a “popular misconception” of how memory works. Hal Arkowitz and Scott Lilienfeld, authors of the article titled “Why Science Tells Us Not to Rely on Eyewitness Accounts,” said that “many people believe that human memory works like a video recorder: the mind records events and then, on cue, plays back an exact replica of them. On the contrary, psychologists have found that memories are reconstructed rather than played back each time we recall them.”


It is in this process of “reconstruction” that unscrupulous law enforcement officials can manipulate the witness with misleading suggestions or false information about either the crime or any alleged perpetrator of the crime. Scientific American quoted the eminent memory researcher and psychologist Elizabeth F. Loftus of the University of California who told the magazine that the act of remembering is “more akin to putting puzzle pieces together than retrieving a video recording.” And Arkowitz/Lilienfeld pointed out that questions from a third party, even an attorney, “can alter the witness’s testimony because fragments of the memory may unknowingly be combined with information provided by the questioner, leading to inaccurate recall.”


Seasoned law enforcement officials instinctively understand this “reconstruction” phenomenon, even though they may have never heard of Ms. Loftus’ theories. They know certain kinds of suggestions and the planting of misleading information can lead to the identification they seek in a case which is necessary to secure an indictment and place the defendant on trial as was done with Donald Dwayne Good. Arkowitz/Lilienfeld explained how this false “reconstruction” process can work:


“Many researchers have created false memories in normal individuals; what is more, many of these subjects are certain that the memories are real. In one well-known study, Loftus and her colleague Jacqueline Pickrell gave subjects written accounts of four events, three of which they had actually experienced. The fourth story was fiction; it centered on the subject being lost in a mall or another public place when he or she was between four and six years old. A relative provided realistic details for the false story, such as a description of the mall at which the subject’s parents shopped. After reading each story, subjects were asked to write down what else they remembered about the incident or to indicate that they did not remember it all. Remarkably about one third of the subjects reported partially or fully remembering the false event. In two follow-up interviews, 25 percent still claimed that they remembered the untrue story, a figure consistent with the findings of similar studies.”


Clearly these studies demonstrate that crime victims and crime witnesses can believe they actually saw or experienced events that did not happen. What do these studies mean for criminal defense attorneys? First, it places a greater burden on defense attorneys to more closely question law enforcement officials connected with interviewing witnesses and those involved in any lineup procedures about every minute aspect of the interview/identification process to determine if any improper influences were injected into that process.


Secondly, defense attorneys must question witnesses about their memory recall processes to determine exactly how the witness “reconstructed” the event. It is essential that the attorney make sure the witness’ reconstruction was not tainted by any undue influence from an outside source. For example, Jane Doe should have been asked whether her identification of Donald Dwayne Good’s photo was based on the “dark tan” he had in that photo—and if that was the basis for her identification, then she should have been questioned about whether Good had the same “dark tan” when she viewed him in the live lineup.


Finally, criminal defense attorneys should attempt to introduce expert witness testimony about the possibility of false eyewitness identification. Arkowitz/Lilienfeld explain why:


“In addition, allowing experts on eyewitness identification to testify in court could educate juries and perhaps lead to more measured evaluation of the testimony. Most U.S. jurisdictions disallow such experts in courtrooms on grounds that laboratory-based eyewitness research does not apply to the courtroom and that, in any case, its conclusions are mostly common sense and therefore not very enlightening. Yet psychologist Gary Wells of Iowa State University and his colleague Lisa Hasel have amassed considerable evidence showing that the experimental findings do apply to courtroom testimony and that they are often counterintuitive.”


We submit that testimony from experts on eyewitness identification, such as Dr. Wells or Ms. Loftus, is far more credible, reliable, and scientific than allowing a dog trainer to testify as an “expert” about “dog scent” identifications which are currently allowed in some Texas courtrooms. Furthermore, memory reconstruction studies have more scientific validation than do most forensic evidence now allowed in criminal trials—evidence which has garnered a national reputation as “junk science.” As Arkowitz/Lilienfeld pointed out: “Science can and should inform judicial processes to improve the accuracy and assessment of eyewitness accounts. We are seeing some small steps in this direction, but our courts will have a long way to go to better ensure that innocent people are not punished because of flaws in this very influential type of evidence.”


These “small steps” by the courts in the right direction are long overdue. The Innocence Project points out that as far back at the late 1800s, experts were trying to inform the nation’s judicial system that eyewitness identifications are highly susceptible to error. Hugo Munsterberg’s book “On the Witness Stand” was published in 1907 which questioned the reliability of eyewitness identification; and Munsterberg’s work was followed by Edwin Borchard’s 1932 book “Convicting the Innocent” in which the Yale law professor said the leading cause of wrongful convictions was mistaken identification by eyewitnesses.


Criminal defense attorneys through their work in the courtrooms, their blogs, and in their law journal articles have an obligation to press the courts to allow testimony in criminal trials of experts on eyewitness identification. Juries have a right to hear flaws in what prosecutors feel is their best evidence: the pointed finger of guilt by witnesses under oath on the witness stand. We know better. It is not the best evidence. It is a flawed body of evidence that has already sent too many innocent people to prison and most probably some to state death chambers. This travesty must end.




Good v. Curtis, 2010 U.S. App. LEXIS 3207 (5th Cir. Feb. 18, 2010)

By: Houston Criminal Lawyer John Floyd and Paralegal Billy Sinclair