We live in age of scientific crime detection. Nevertheless, there are two forms of evidence that still exert profound influence in the judicial system – confessions and eyewitness testimony. In I confess – a guide to police interrogation I looked at how police interrogation methods can sometimes lead to suspects confessing to crimes they did not commit. This week I want to look at the fallibility of eyewitness evidence.
I came across one shocking example of this the case of the rape of Jennifer Thompson. We need to clarify that this is not like the invented charges that I looked at in False rape accusations. Jennifer Thompson was clearly a victim. In July 1984 Thompson a 22-year-old college student living in Burlington, North Carolina, was sleeping in her apartment after coming home from a date. What she did not know that a man had broken into her apartment and was still there. She went to sleep but she woke up in the middle of the night. Realising that there was a man in her room, she tried to scream but her attacker had a knife. She offered him her money, credit cards, and even her car if he would just leave her alone. But he wasn’t interested.
During the attack the college student had one focus: to survive and identify her rapist. She had a strategy. She had heard that when you were raped, the best thing to do was not to fight. That would be exactly what a rapist would want. They need to control and humiliate the woman. She was at a physical disadvantage with respect to her attacker. So she would have to outsmart him. Her first plan was to try to get information. She would remember everything: his face, his voice, his hair his eyes and any other distinguishing features, such as scars or tattoos. She would memorise his clothing, try to identify his height by comparing it to her own.
The second part of her plan was to just try to get out. She claimed that she had a knife phobia and got him to put down the knife. She wanted to get to the kitchen because the rapist had told her that he had forced the door there. So she said that was feeling really thirsty and needed to get something to drink. Once in the kitchen she turned the light on, knowing that if it was on, he would not come too close to me, for fear of being identified. He did indeed stay around the corner. Here is how she describes what happened next:
“And I started to run the water and throw ice cubes in my sink, and open up cabinets and draws, making as much noise as I could, and getting up enough courage to run. And that’s when I opened up the door and I just took off out the door. I didn’t know where I was going to go, but I was leaving.”
She ran for her life, pursued by the rapist. Fortunately she was able to get to the house of a couple. The woman, who worked at the university, recognised her as a student.
She first arrived at the hospital at about four o’clock in the morning, where she was given the forensic rape test. Unfortunately, the doctor seemed annoyed at having been woken up and was rather unsympathetic. What’s more he did an incomplete examination; she was not given penicillin for any possible sexually transmitted diseases, nor was given the morning-after pill. She would later have to go through the whole humiliating process again. She also learned there appeared to be a second victim, Elizabeth Warren, who would prove important later in the story.
After the rape kit, she went to the police station, where she was asked to use one of those facial composite kits to create an accurate photofit of the perpetrator. And of course she had to describe what had happened several times and to give as many statements as she could.
Thanks to an anonymous tip Ronald Cotton, an African-American restaurant worker became a “person of interest”, as they all say in the police dramas I watch. He did have a previous sexual assault conviction and a dodgy alibi. So two days after the crime Thompson was shown his photo and those of five others. She identified Cotton. She thought he looked exactly like the man who had raped her. Not a lot of time had elapsed since the crime so her memory was still very fresh. There are a couple of things I need to say about the photo identification. First she was shown all the photographs at the same time. This is a seriously flawed methodology, which I will come back to later. Secondly, she actually took five minutes to decide between two of the photos, which is actually quite a long time.
And just over a week later came the identity parade, which was not like you see on the television. The normal venue was being renovated, so they had to use another room. She was in the same room with Cotton and the other men from the line-up. She now saw Cotton in the flesh. His demeanour and his posture further convinced her that Ronald Cotton was the man. He looked exactly like the man from the sketch that she had given to the police. His mannerisms, his voice, his height, his weight all appeared to fit. the rapist. And as time went by, she became more convinced that Cotton was the man. Whenever she relived the trauma it was Ronald Cotton’s face she would see
During the trial, Thompson once again identified Cotton as her rapist. When the defence attorney attempted to introduce evidence of the second, and seemingly related rape, of which Cotton had been excluded, the Judge refused to allow this evidence to be heard. Thus, Ronald Cotton was convicted and sentenced to life in prison based on what, apart from the identification by Thompson, was rather flimsy circumstantial evidence.
On Jan. 17, 1985, the day Cotton was convicted by a jury of one charge of rape and one of burglary and sentenced to life in prison, Thompson toasted her victory with champagne. “It was the happiest day of my life,” she said. A year into his prison term Cotton came across a prisoner with uncanny similarity to him. Indeed inmates and guards would often mix them up. The inmate was Bobby Poole. He had apparently been bragging about committing the rapes and had allegedly joked that Cotton was doing his time for him
In 1987 the North Carolina Supreme Court reversed the conviction and ordered a retrial based on the second rape evidence that the judge had not allowed. But the second trial proved to be a disaster for Cotton. Elizabeth Watson now became convinced that Ronald Cotton was her rapist as well, even though she had been unable to pick Cotton out of both a photo and live line-up before the first trial. She had actually picked a police “filler” in the original line-up.
Bobby Poole’s was also called to testify but he would not confess to both the rapes, and Jennifer Thompson denied ever having seen him before. Indeed she was outraged that the defence could even suggest that Poole had been the rapist. She was emphatic: “Bobby Poole did not rape me. Ronald Cotton raped me.” Ronald was again convicted, but this time for two rapes. He was sentenced to two life sentences plus 54 years, confirmed on appeal in 1990 and the North Carolina Supreme Court in 1991.
Luckily a lawyer Richard Rosen agreed to take up the case. Now new forensic techniques were coming to the fore. The attorney filed a motion for DNA testing. The Thompson crime evidence was too deteriorated to test. However, the Watson materials completely excluded Ronald Cotton as the perpetrator. The defence then requested that the results be sent to check against the DNA of convicts. The DNA positively identified Bobby Poole. Detectives went to see Poole, who confessed. On June 30, 1995, ten and a half years into his sentence, Cotton was officially cleared of all charges and released from prison. The following month he was officially pardoned by the Governor of North Carolina. He was initially offered just $5000 compensation, but thanks partly to the efforts of Thompson, still pitifully low for all that this man was put through.
There is a terrible irony in all this. Jennifer Thompson, who had been so determined to identify her man, had, in the end, got it all wrong. She felt a terrible sense of guilt over what she had done. But I do not think she ever acted in bad faith. Memory is not like a video recording. Traumatic situations can affect what you remember. And each time you remember the memory can become subtly different. Thompson, who had initially taken five minutes to choose the right photograph, became more convinced that Cotton had raped her. She thought that people would think she was a racist. What is true is that witnesses do seem to have problems accurately identifying a suspect if another race. They can pick out the broad stereotypical features but not the finer details.
There are many uplifting aspects to this story. Two years after his release Jennifer Thompson and Ronald Cotton met for the first time at a church not far from where she had been raped. Thompson felt terrible; she struggled to stand up and began to sob. She looked at him and said: “If I spent every minute of every hour of every day for the rest of my life telling you that I’m sorry, could you ever forgive me?” Thompson didn’t expect Cotton’s response:
“I’m not mad at you. I’ve never been mad at you. I just want you to have a good life.”
So Thompson and Cotton have become friends. They have also become tireless campaigners for reform of eyewitness identification procedures. There is a lot wrong with the system. According to the Innocence Project Eyewitness misidentification is the greatest contributing factor to wrongful convictions proven by DNA testing, playing a role in 75% of convictions overturned through DNA testing nationwide. In 40% of those cases more than one witness identified the suspect. When you send the wrong person to prison for a crime all society is affected – the real criminals are free to go on committing crimes.
I have seen a number of interesting proposals to give the identification process a more scientific foundation. We need double-blind presentation; the photos or line-up members should be presented by an administrator who does not know who the suspect is. The fillers should resemble the eyewitness’s description of the perpetrator and the suspect should not stand out. Only one suspect should be included in the line-up. A sequential presentation of the line-up is preferable. Otherwise it becomes like a multiple choice test with the victim narrowing the options. That is not the kind of identification you want at the time of the identification, the eyewitnesses should provide a statement in her own words indicating her level of confidence in the identification. Finally these procedures should be videotaped to increase the transparency of the process. In the end I agree with the Innocence Project: memory is evidence and must be handled as carefully as the crime scene itself to avoid forever altering it.