Editorial: Screen out junk science

The Dallas area remembers few crimes that equal the ghastliness of 7-year-old Ashley Estell’s murder. Authorities thought they had the right suspect, the man who plucked her from a Plano city park and strangled her.

Michael Blair spent 14 years on death row, but the case against him came apart, and prosecutors declined to try him again last year.

The weak link: forensic work. Supposed expert testimony based on hair-fiber analysis produced damning trial evidence, but the findings were later discredited by newly developed DNA technology.

The theme of shaky forensics also played out this year in a South Texas case based on the bizarre-sounding but somehow established practice of using dogs to identify suspects through a so-called “scent lineup.”

Perhaps most bizarre of all is that dog evidence gets into courtrooms in Texas and across the nation absent regulation or scientifically accepted protocols.

In the Texas case, it was DNA, once again, that cleared a wrongly accused suspect. Calvin Lee Miller, 43, was picked up based on a tip in a string of attacks on elderly women in the Victoria area.

Enter the bloodhounds. Dog handler Keith Pikett of the Fort Bend County sheriff’s office takes a swab of a suspect’s body and observes his dogs sniff it out along with swabs from anonymous people. The dogs supposedly provide conclusive cues about which swab is the bad guy’s. Only Miller spent 62 days in jail before DNA proved he wasn’t the bad guy. The bloodhounds also mistakenly led authorities to a Victoria County sheriff’s captain in a high-profile murder case.

The Innocence Project of Texas is investigating how many of Pikett’s hundreds of criminal cases should be reviewed for possible miscarriage of justice. That’s an important initiative, but the battle against shaky or junk forensics needs to be fought on many fronts.

The National Academy of Sciences released an exhaustive report this year after an 18-month, congressionally mandated study on forensics. The conclusions would disillusion anyone who thinks television’s CSI reflects the reality of the nation’s crime labs.

The report says about 50 percent of DNA exonerations reveal outmoded, invalidated or improperly used forensic techniques. Poor training and lack of resources harm police investigations despite strong interest from prosecutors and juries expecting CSI-like results.

The report also suggests a clearinghouse for peer-reviewed guidance, something that would surely benefit the forensics community and the nation’s courtrooms. Currently, judges have to make decisions about what is admissible based on conflicting expert opinions on the reliability of forensic techniques. There needs to be a better way, and Congress should pursue options for making it happen on a national scale.

Texas, meanwhile, has already established its Forensic Science Commission, an investigatory body that’s still in its infancy, with a staff of one. Using paid consultants, the panel examines complaints on a restricted list of forensic disciplines but has yet to issue an opinion on the cases it has agreed to review.

State lawmakers should pay close attention to the outcome of these cases and the value of the commission’s recommendations. The panel may well deserve expanded authority and bulked-up resources.

With this state’s nation-leading list of exonerations — 38 so far — Texas has a heightened responsibility to make sure jurors aren’t hearing hogwash from the witness stand.

No comments: