Hantu Laut
Forensic science and DNA analysis are not yet a perfect science, there are many grey areas and many of the so called experts in Australia are not truly competent to give foolproof analysis of DNA samples let alone our Malaysian expert who had been criticised by the foreign experts.A mistake or carelessness would have sent an innocent man to prison or worse the gallows.
With the ongoing Anwar's sodomy trail the debate on DNA and forensic science is heating up in court with foreign and local experts and solicitors from both sides slugging out in the court room.
Below is a report which incidentally mentioned one of the Australian experts in the Anwar's sodomy trail, the well known and hypercritical Dr Brian McDonald.
Forensic science in the dock
The Australian
BRIAN McDonald knows he has few admirers among the forensic scientists at the nation's DNA testing laboratories. "I am the antichrist as far as forensic science is concerned," he says. The reason is simple. McDonald, an independent DNA consultant and molecular geneticist, has built a career by finding flaws in the test results some of these laboratories have provided for the justice system.
Of the hundreds of tests for DNA (basic genetic cell material) that have been referred to him by defence lawyers, he says, he has found problems in 30 per cent to 50 per cent of them.
Such a high rate of contested test results is one reason there is growing unease among lawyers that, in some cases, DNA evidence alone is sending people to prison.
DNA analysis, when applied correctly, is widely considered the most reliable form of forensic science. But if the problems picked up by McDonald have also been present in even a small proportion of other DNA tests used in court, there is a risk some people in jail are victims of scientific error.
The growing doubts about the evidence of scientific experts is not confined to DNA analysis.
A debate is raging in academic circles about what some lawyers believe is a lax approach by the courts that is exposing juries to "junk science", opinions presented by so-called experts who are unable to explain the scientific methods that helped them reach their conclusions.
Courts in Queensland and the Northern Territory have drawn the line at accepting evidence from people who claim to be experts in forensic odontology, or matching bite marks.
But a recent article in the Criminal Law Journal has unleashed a barrage of criticism at far more established fields of forensic science as well as new areas such as barefoot morphology, or the identification of people from the weight-bearing patterns left by their feet. The authors of that article, who include David Field, director of Bond University's Centre for Forensic Excellence, call for a new approach from the courts to weed out unreliable forensic evidence.
They have urged the courts, before accepting expert evidence, to insist on being told about the known and potential error rates of each field of forensic science.
To bolster their argument, they cite research last year in the US showing the subjective nature of fingerprint matches.
Years after fingerprint experts had made decisions on certain prints, researchers presented them with the same prints and were startled by what happened.
"Two-thirds of the experts made inconsistent decisions; [that is,] they disagreed with themselves," the journal article says.
The same concerns about lack of rigour have been voiced by University of NSW legal academic Gary Edmond, who is a member of the council of the Australian Academy of Forensic Sciences.
He says fingerprint analysis, like DNA analysis, is at the the better end of forensic science. But he says in some areas forensic science is "over-reaching and sometimes just poking around in the dark".
When asked to nominate disciplines at the worst end of the spectrum, Edmond names bite-mark matching, photograph comparisons and footprint matching.
Similar concerns arose last year in the US when a report for the National Academy of Sciences recommended that court testimony by experts should be grounded in science and should acknowledge areas of uncertainty.Read more.
