DNA Evidence
In October 1975, 11-year-old Lesley Molseed was found stabbed to death on the Yorkshire moors in the north of England. The following year, Stefan Kiszko was jailed for her murder. He spent 16 years in prison, before he was finally proved innocent and released in 1992. Kiszko died the following year. His mother, who had doggedly campaigned to clear his name, died a few months later. Another man, Ronald Castree, was this week convicted of the murder of Lesley Molseed. The evidence consists mainly of a DNA match with traces of semen found at the crime scene 32 years ago. How reliable is this conviction?
Risk of Error
The subject of DNA evidence is often discussed by people interested in risk, and especially risk communication. There is a common misperception that DNA matching is 100% reliable. In reality, the process is prone to technical and human errors. Incorrect perception of DNA evidence is common not just among the general public (which includes members of juries) but also among police, lawyers and judges. This issue was a feature of the notorious murder trial of the American footballer, O. J. Simpson in 1995.
The misperception with DNA evidence comes from confusing the very low probability of two unrelated people sharing the same DNA profile with the much higher probability of a false positive match due to error. The first probability, of a random match between unrelated samples, was stated in court during the trial of Ronald Castree as one in a billion. The second probability, of a false positive due to error, seems not to have been mentioned in court. In practice, the probability of error is of the order of 1 in 100. Cellmark Diagnostics was one of the labs that found matches between O. J. Simpson's DNA and a blood stain found at the crime scene. Cellmark reported its own false positive rate as about 1 in 200. The impressive one-in-a-billion random-match figure is irrelevant in the context of a 1% error rate.
And these 1% error rates presumably relate to routine cases. What do we have in the Castree case? The murder was committed 32 years ago. According to reports in the Manchester Evening News (see below), Lesley Molseed's body was fully clothed when it was found. Someone, presumably the killer, had ejaculated over the girl's clothing. The police forensic team later recovered samples from the clothing using sticky tape. Some of the evidence mysteriously went missing during the Kiszko trial in 1976 and the clothes themselves were destroyed in the eighties. In 1999, two pieces of old sticky tape in half a brown envelope were found at the Forensic Science Service in Wetherby, Yorkshire. A tiny handwritten reference number linked the tape to the Lesley Molseed case. Genetic material was recovered from the tape using a new process and a number of DNA profiles were generated. One of these was matched with that of Ronald Castree in 2005.
Other Evidence
In order to be confident of a conviction, DNA evidence has to be balanced with other evidence. I am not in a position to judge, of course. It's never easy to get the full story from newspaper reports and this is particularly true for a difficult case that happened a long time ago. However, a couple of things struck me from the press reports.
Castree was clearly a womanizer. His wives and girlfriends do not have fond memories about him. Castree told police that he had been visiting his wife in hospital at the time of Molseed's abduction, which was in the middle of the day. However, in court, his former wife said that she could not remember him ever coming to see her in the daytime. Castree asked a former girlfriend, Joanne Gartside, to be a character reference at his trial, but she refused. However, after Castree's conviction, Gartside told the Manchester Evening News:
"No matter how aggressive his character was, he was never violent towards me. He was dead against men who hit women."
Castree has always protested his innocence and continues to insist that he did not kill Lesley Molseed.
But if Castree didn't do it, who did? Well, this has been a puzzle for over 30 years. Several people have pointed out that the Molseed murder was committed during the days of The Yorkshire Ripper. There was much criticism of the police investigation of the Ripper cases. An offical inquiry resulted in The Byford Report, which was confidential when it was completed in 1981 but most of the report was published online last year as a result of the Freedom of Information Act. Peter Sutcliffe was jailed in 1981 for 13 Ripper murders and 7 attempted murders. In the Report, Lord Byford says:
"I came to the conclusion that consideration of any other crimes which might have been committed by Sutcliffe and any of his associates was a matter for the West Yorkshire Metropolitan Police... it is highly improbable that the crimes in respect of which Sutcliffe has been charged and convicted are the only ones attributable to him. This feeling is reinforced by examining the details of a number of assaults on women since 1969 which, in some ways, clearly fall into the established pattern of Sutcliffe's overall modus-operandi."
But, of course, the police at the time were not looking for the killer of Lesley Molseed because Stefan Kiszko was already in jail for her murder.
Maybe Castree killed Lesley Molseed. Maybe it was Sutcliffe. Maybe it was someone else. After 32 years, can it really be possible to be certain, especially considering the tenuous evidence of the old sticky tape and the background of official criticism of police procedures at the time?
Manchester Evening News, 24 Oct 2007
Manchester Evening News, 12 Nov 2007
- Login to post comments
