Houston chronicle


‘Scent lineups' may be failing smell test
Prosecutors, suspects skeptical of using dogs to get convictions
By MIKE TOLSON
HOUSTON CHRONICLE
Sept. 27, 2009

Dogs have occupied a special place in law enforcement for decades thanks to their heightened sense of smell, and their role has only grown in recent years because of use in explosives detection and the newly popular practice of “scent lineups.”

But forgive Curvis Bickham if he remains less than impressed. Eight months in jail will harden your opinion about a lot of things.

Bickham does not know for sure whether the dogs used to connect him to a triple murder in southeast Houston last year made a mistake, produced a false “hit” because their handler encouraged them or responded correctly to scent that was deliberately placed on items from the murder scene. All he knows is that he was accused of capital murder, a crime that could have cost him his life.

“I lost my home, I lost my business, I lost my reputation,” Bickham said. “I have three little boys depending on me — ages 6, 8 and 9 — and they charge me with the most heinous thing they can charge a man with.”

Bickham, 49, gets emotional in the retelling of the October 2008 arrest and the painful months that followed before charges were dropped for lack of evidence.

The evidence against Bickham may have been slight, little more than a positive reaction from Keith Pikett's bloodhound. But it was enough to charge him.

Now Bickham has added his voice to the growing chorus of critics who call Pikett a fraud and the use of so-called scent lineups “voodoo science.”

“This has got to stop,” Bickham said.

The Innocence Project of Texas agrees. A group that works on behalf of the wrongly accused, the project issued a scathing report last week on both Pikett and the practice of getting courtroom evidence from scent lineups, in which a specially trained dog matches a suspect's scent to items from a crime scene.
Useful in investigations

The group does not suggest that dogs not be used by police, conceding they can be a useful investigative tool. It's the next step that bothers them — the eagerness of police to use scent results as the basis of a criminal charge and prosecutors to accept them as evidence. It has called on the state Attorney General's Office to investigate other cases in which Pikett's dogs were crucial to getting a conviction.

“It is junk science that is being used to convict and charge people,” said Jeff Blackburn, the group's chief counsel.

The Innocence Project's report cited not only Bickham's case but several others who turned out to be wrongly accused based on scent matches from Pikett's dogs. Calvin Miller was charged with rape and robbery in Yoakum County before DNA evidence cleared him. Michael Buchanek was named a prime suspect in a Victoria murder based on a Pikett scent lineup, before another man confessed and pleaded guilty.

Miller is suing Pikett, a deputy with the Fort Bend County Sheriff's Department. Bickham is considering doing the same.

Pikett declined an interview request based on the advice of his attorney, Randall Morse, who did not return calls from the Chronicle.

Other dog trainers have attracted scrutiny. In Florida, John Preston occupied a niche similar to Pikett's in the 1980s. Three men he helped imprison have since been released after other evidence showed the convictions were wrong.

Pikett got into police dog work by accident in the 1990s, when his pet bloodhound Samantha proved exceptional at tracking contests in dog shows, he told the Chronicle in a 2004 interview. In 1998, he became a Fort Bend deputy, and he and his dogs have become celebrities of sorts because of their success in scent detection and work in high-profile cases.

As publicity spread about Pikett's work, more and more police agencies and prosecutors began to use him. He has said he has no idea of how many cases his dogs have been involved in, but it numbers in the thousands. He also has testified that his dogs have almost never been wrong. One of them, “Clue,” erred once in 1,659, he has testified, while “James Bond” made one mistake in 2,266 tries.

Well-known police dog trainer Bob Eden, who has written two books on the subject and helped develop minimum training standards, calls such numbers incredible. Literally. Although dogs can perform with amazing consistency, Eden said, handlers are all but certain to make blunders and inadvertently tip the dog to the desired match, or to an incorrect match if the handler has no knowledge of the correct one.

Lack of records troubling

Of equal concern, Eden said, is the absence of carefully recorded training records to demonstrate the dog's proficiency and error rate. Pikett has acknowledged that he does not keep such records.

“Prosecutors should never put him on the stand without training records,” Eden said. “Everything that is done with that dog should have a training record.”

Three experts hired by the Innocence Project criticized Pikett's work. The former head of police dog training in the United Kingdom, Bob Coote, reviewed a video of one of Pikett's scent lineups and was appalled.

“This is the most primitive evidential police procedure I have ever witnessed,” Coote stated. “If it was not for the fact that this is a serious matter, I could have been watching a comedy.”

Pikett has plenty of supporters. Many prosecutors swear by him. In a lengthy article on the murders of Patricia and Kent Whitaker for The Prosecutor, a publication of the Texas District and County Attorneys Association, Fort Bend prosecutor Jeff Strange told how Pikett's dogs were able to link evidence from the Whitaker home to the killers.

‘Pervasive problem'

The Whitaker case, however, contained a substantial amount of evidence against Bart Whitaker and the others who took part. What concerns Blackburn are cases that have little additional evidence, meaning the dog-scent results could be key to conviction. Texas, like most states, allows dog scent results into evidence despite an absence of scientific norms and standards.

“In some counties, police are abandoning normal police work in favor of Pikett and his dogs,” Blackburn said. “We believe it is becoming … a widespread and pervasive problem.”

Former Harris County prosecutor Vic Wisner, an assistant district attorney for more than 24 years, came to view Pikett's work skeptically while working on a case involving a series of pharmacy burglaries. The dogs connected items from the break-ins to the scent of a suspect provided by two detectives, who subsequently filed felony charges. A task force working on the same case, however, had a different suspect, who finally was apprehended during a burglary.

Wisner dismissed the charges against the uninvolved suspect and sent an e-mail to fellow prosecutors to make them aware of what happened in the event any of their cases involved Pikett's dogs. His one-word summation of the scent evidence: “ludicrous.”

mike.tolson@chron.com


Junk Science
Truth in Justice