The evidence does not lie – or does it?

Will exposes signal end to blind reliance on "science"?

CSI trumpets the notion that evidence does not lie.

But critical news stories may be signaling the end of uncritical evidence of this dubious tenet.

Take this introduction to a 2007 Denver Post series, “Trashing the Truth”:

Virtually every night on prime time, TV detectives pluck tiny samples of DNA from clothes, carpets, and even car tires, test it and nail the bad guy, all in one episode. But in real life, DNA samples … get mishandled with impunity…. Law enforcers have won passage of laws letting them off the hook for perjuring evidence on which people's lives and liberties hinge. The result: Killers walk…. These are the stories you won’t see on CSI. In cases around the country, the truth is being trashed.

As the Post series meticulously documents, contrary to the portrayals on fictional crime dramas and by expert witnesses for the prosecution, evidence rooms are characterized by "darkness and disorder" and the accidental and intentional destruction of tens of thousands of potentially important DNA samples.

Forbes magazine recently echoed the alarm, in an opinion piece "What's wrong with CSI: Forensic evidence doesn’t always tell the truth" by Roger Koppl, an economics professor and director of the Institute for Forensic Science Administration:

Forensic evidence is foolproof, right? It's how those clever cops on CSI always catch the killer. DNA evidence springs innocent men from prison. Fingerprints nab the bad guys.

If only forensics were that reliable. Instead, to judge by the most comprehensive study on the reliability of forensic evidence to date, the error rate is more than 10% in five categories of analysis, including fiber, paint and body fluids. (Meaning: When the expert says specimen X matches source Y, there's a 10% probability he's wrong.)

Even Government Technology, hardly a muckracking journal, is calling for reform. GT's July 9 story, "Police Crime Labs Struggle with Funding, Training and Bias Issues," focuses on the Houston crime lab, where an investigation found "hundreds of cases where incompetence, inadequate training and resources, lack of guidance and even intentional bias on the part of a crime lab - which is not independent from the HPD - contributed to mistakes."

The problems "may be inherent in crime labs across the country," the GT article concludes, citing reports of DNA testing errors nationwide - in Washington, North Carolina, California, Minnesota, Pennsylvania and Nevada.

Problems obvious

When a TV station in Houston looked into that city's crime lab operations back in 2002, the problems were obvious to an independent forensic expert:

"They weren't running proper scientific controls. They were giving misleading testimony. They were computing their statistics incorrectly - in a way that was biased against the accused in many cases,” said forensic expert William Thompson of UC Irvine.

Errors favor prosecution

Most troublingly, the errors are not random - they almost invariably favor the prosecution. Thompson identified a "team culture" mentality in the crime lab, a mentality that may lead technicians to bend the evidence against defendants in court.

Journalist Scott Henson at Grits for Breakfast has been keeping up with this issue for several years, publicizing not only DNA evidence scandals but also problems with other supposedly neutral scientific technologies in the criminal justice system. These include false-positive breathalyzer tests for drunk drivers and urinalyses that routinely send probationers and parolees back to jail.

Indeed, Henson says it was the 25% rate of false positives in breathalyzer tests that first turned his attention toward "the reality that accuracy appears optional in many forensic science endeavors, with error rates of 10% or more routinely accepted in a variety of forensic fields."

What’s the solution?

Most outsiders agree that a first step toward improving the abysmal state of scientific evidence collection and analysis is outside oversight.

Beyond that, Roger Koppl, the economics professor writing for Forbes, has some other interesting ideas, primary among them opening the labs to free market forces:

The core problem with the forensic system is monopoly. Once evidence goes to one lab, it is rarely examined by any other. That needs to change. Each jurisdiction should include several competing labs. Occasionally the same DNA evidence, for instance, could be sent to three different labs for analysis.

This procedure may seem like a waste. But such checks would save taxpayer money. Extra tests are inexpensive compared to the cost of error, including the cost of incarcerating the wrongfully convicted. A forthcoming study I wrote for the Independent Institute (a government-reform think tank) shows that independent triplicate fingerprint examinations in felony cases would not only eliminate most false convictions that result from fingerprint errors but also would reduce the cost of criminal justice if the false-positive error rate is more than 0.115%, or about one in a thousand.

Other reforms suggested by Koppl and others include making crime labs independent of law enforcement, requiring blind testing, and giving the defense the right to its own forensic experts:

When crime labs are part of the police department, some forensic experts make mistakes out of an unconscious desire to help their "clients," the police and prosecution. Independence and blind testing prevent that. Creating the right to a forensic expert for the defense would help restore the imbalance in scientific firepower that too often exists between prosecution and defense.

The Denver Post series, Trashing the Truth, includes the following segments:

1. Bad faith difficult to prove: Through carelessness or by design, tiny biological samples holding crucial DNA fingerprints often disappear on authorities' watch. Innocent people languish in prison, and criminals walk free.
2. Room for error in evidence vaults: In some evidence rooms, chaos and disorganization make searches futile. Others are purged of valuable DNA samples, leaving cases unsolvable.
3. Missing rape kits foil justice: Rape kits routinely vanish, unfuriating victims and prosecutors alike. Even when evidence is intact, laws can keep suspects like William Harold Johnson walking free in our midst.
4. 14 years later - Tell my story: Floyd Brown has an IQ in the 50s. Its authenticity in doubt, his confession to a 1993 murder has him locked up indefinitely in a North Carolina mental hospital. A bloodstained stick that could settle his innocence or guilt has vanished.

Ripple effects of psychologist's porn addiction

Sex offenders to get new trials

A state psychologist's own sexual deviancy will lead to new trials for two Iowa sex offenders whom he opined should be locked up as sexually violent predators, the Des Moines Register reported today.

The evaluator is Joseph Belanger of North Dakota, whom I blogged about last December after he acknowledged an adiction to online child pornography.

Belanger's deviancy is causing ripple effects including a review of more than 100 cases and an upcoming appeal before the North Dakota Supreme Court.

Belanger resigned from North Dakota State Hospital but his license is still active during a continuing investigation, the Register reported.

Although many state evaluators testify that it is difficult for sexual deviants to benefit from treatment, Belanger "stressed that he has worked with a Zen teacher and that 'I believe recovery is possible,' " according to the Register report.

The Des Moines Register story is here. My Dec. 9, 2007 post is here.

Canada: Restorative justice touted for hate crimes

Citizens of peaceful and tolerant New Brunswick, Canada, have been shocked by a recent outbreak of racist and anti-Semitic vandalism of churches and synagogues.

The answer?

Restorative justice, says criminology professor Elizabeth Elliott of Simon Fraser University in Vancouver. Elliottt is a leading Canadian expert on restorative justice and author of the book, New Directions In Restorative Justice: Issues, Practice, Evaluation.

If religious leaders and other victims are willing to meet with the offenders and if the offenders agree to participate, "there is an excellent learning opportunity here," said Elliott.

New Brunswick already has restorative justice programs in place both for juvenile and adult offenders, as do other Canadian metropolises such as Nova Scotia and British Columbia.

Of course, the offenders have to get caught first, no small problem in a vandalism case.

New Brunswick's Telegraph Journal has the story.

Hat tip: Understanding Crime

Who Killed Chandra Levy?

A new DNA technique has exonerated the father, but the passage of time makes the mystery of who killed JonBenet Ramsey in 1996 unlikely to be solved. Likewise for the 2001 murder of Chandra Levy, Washington's most famous unsolved crime. In both cases, premature certainty about one suspect (in the latter case a congressman) led police to ignore critical leads and commit a chain of errors that spiraled out of control.

The Levy case became overshadowed by September 11. But now, six years later, Washington Post reporters Sari Horwitz and Scott Higham have conducted an in-depth exploration of the case. The 12-part reconstruction illustrates how far off track a police investigation can go when it is overwhelmed by "white-hot media coverage," in this case fueled by the possible involvement of California congressman Gary Condit. (As part of their extensive interviews, the reporters got Condit to talk for the first time about the case.)

The Post series provides a rare look at an unsolved homicide case from the inside, following the twists and turns of an investigation that was filled with false hopes, false leads and false suspects. It would tarnish a police department and wreck a reputation. It would move with tremendous energy and purpose in one direction and end up in another. It would be marked by an enormous effort by police - and a chain of mistakes that got longer and longer.

The 12-part series starts here.

"Misfeasance not malfeasance"

Detective won't face charges in Tim Masters case

A special prosecutor has decided not to file criminal charges for perjury or illegal eavesdropping against the Colorado detective who spearheaded the investigation of 15-year-old Timothy Masters for the murder of Peggy Hettrick, a case about which I have blogged extensively (click here for my past posts).

You will recall that Lt. James Broderick was convinced of the boy's guilt despite the absence of any physical evidence linking young Masters to the crime. He continued to pursue him for years, finally hiring prominent forensic psychologist Reid Meloy to render an opinion based on Masters' personal sketches. That opinion helped garner a conviction; after a decade in prison, Masters was recently freed based on DNA evidence.

Prosecutor Ken Buck said that although he uncovered "several flaws" during his "limited investigation," he did not believe that Broderick engaged in deliberate criminal conduct, nor was there a "reasonable likelihood" that a jury would convict the detective at trial.

A separate investigation into whether prosecutors in the case violated professional standards is due to conclude soon. That investigation is by the Colorado Supreme Court's Office of Attorney Regulation. The former prosecutors, Terry Gilmore and Jolene Blair, are both now judges.

The Colorodoan quotes one former police investigator in the case, Linda Wheeler-Holloway, as saying that the prosecutor's decision is no surprise.

"People didn't play fair. By not telling the whole story, leaving things incomplete, that kind of skewed things in their favor…. There was a lot of faults committed in a lot of arenas that led to the wrongful conviction of Tim Masters."

Writer Pat Hartman, who has an extensive blog on the case entitled "Free Tim Masters Because," has a scathing denunciation of the prosecutor’s decision.

"This wrapup of Broderick's involvement is inadequate and unsatisfactory. It's like watching an elephant be pregnant for months and then give birth to a mouse. Now there's supposed to be an internal [police] investigation…. With this tepid whitewash as precedent, it’s not difficult to foresee the results of that investigation.”

The prosecutor's 11-page report is here. The Coloradoan and the Denver Post have news coverage.

New Scientist reports low sex offender recidivism

Peter Aldhous over at New Scientist is reporting on the declining rates of sex offending in California, which I blogged about on June 23 (click here), as well as similar reported declines in Minnesota. The article, "Sex offenders unlikely to commit second crime," begins like this:

Sex crime statistics tend to make depressing reading, but now there is some good news from the most populous state in the US. Just 3.2 per cent of more than 4,000 sex offenders released on parole in 2002 were re-imprisoned for another sex offence in the subsequent 5 years, according to new figures from California.

While experts know that sex offenders are less likely to reoffend than most other criminals (New Scientist, 24 February 2007, p 3), the very low rate of re-imprisonment in the new study will challenge public perceptions about the risks these criminals pose.

The figures are broadly consistent with a 2007 Minnesotan study, which found that 3.2 per cent of sex offenders released from 1990 to 2002 had been re-imprisoned for a further sex crime within 3 years of their release.

What's more, sex offenders in Minnesota are even less likely to reoffend....

Unfortunately, you have to subscribe to read the remainder of the article, as well as prior coverage of this topic by Mr. Aldhous. However, my June 23 blog post on the new California data is here, and the Minnesota recidivism study is online here. A comprehensive, 225-page report by researchers on behalf of the California Sex Offender Management Board is online here. The data on 5- and 10-year recidivism are a bit hidden at the CSOMB website, but you can get them HERE and HERE, respectively.