Dogs and the Scent of a Crime: Science or Shaky Evidence?By Hilary Hylton / Austin Monday, Aug. 03, 2009
In detective dramas, a dog's powerful sense of smell has become a predictable crime solver: the trusty canine takes a sniff of a suspect object and follows the scent, eventually arriving at the perpetrator of the evil deed. But in real life, is this reliable evidence — or is it junk science that has helped put away innocent people?
In mid-August, the Innocence Project of Texas plans to unveil a detailed study focusing primarily on the extensive work of one Texas dog handler whose use of scent-ID techniques is under fire in the federal courts. At the heart of the study is the work of Deputy Keith Pikett, a canine officer with the Fort Bend County Sheriff's Office, just southwest of Houston. The first case studied involves Calvin Lee Miller, who was charged with robbery and sexual assault after Pikett's bloodhounds alerted police to a scent on sheets that Pikett said matched a scent swipe from Miller's cheek. DNA evidence later cleared Miller, but only after he served 62 days in jail. In a second case, former Victoria County Sheriff's Department Captain Michael Buchanek was named as a "person of interest" in a murder case after Pikett's bloodhounds sped 5.5 miles from a crime scene, tracking a scent to Buchanek's home. Another man later confessed to the murder. (See pictures of puppies behind bars.)
Both cases have resulted in lawsuits seeking damages from the municipalities and law-enforcement agencies that used Pikett's work. They could be costly. In a separate case, a California man, Jeffrey Allen Grant, served three months in jail in 1999 after TinkerBelle, a bloodhound, mistakenly identified him in a rape case. He won $1.7 million in damages. (See pictures from the 2009 Westminster Kennel Club dog show.)
Jeff Blackburn, head of the Innocence Project of Texas, has labeled the dog-scent evidence as "junk, not even junk science." He adds, "We are working on a very intense, independent investigation of Pikett's activities." Pikett, who through his lawyer has declined public comment, is being sued for civil-rights violations in federal court by Miller and Buchanek. Blackburn says the innocence team is combing Texas public records to assess Pikett's impact on other cases. In the meantime, the Innocence Project of Texas study is being supported by canine-law-enforcement experts who, while not going so far as to call dog-scent evidence junk, fear that misapplication of the undisputed canine talent for recognizing smells will discredit good cases along with the bad. (See pictures of a real-life hotel for dogs.)
The notion of a crime-busting dog can be appealing, not to mention a break for jurors from mind-numbing expert-forensic-witness testimony. But experts caution that it is not the dog who testifies but rather the handler. "The animal knows what he is smelling, and everyone else has theories of what he's smelling," says Russ Hess, executive director of the U.S. Police Canine Association. For hundreds of years, humans have relied on the ability of dogs to distinguish scents to track prey, whether in the hunt for food or the search for a prison escapee. Bloodhounds are the recognized experts in supersensitivity to odors (some states allow scent evidence only from bloodhounds to be admitted). But even the best-trained scent dog — and Hess says the dogs require constant training — can make mistakes. "They are fallible, just like a person," says Charles Mesloh, a former canine officer and criminologist at Florida Gulf Coast University.
Dogs, he adds, respond to handlers, perhaps for reward or praise, or simply because of emotional connections, wanting to please their human partner. "Dogs aren't stupid — they cheat," Mesloh says. "What goes down the leash, comes up the leash." In the Netherlands, where tough evidence protocols are in place, a suspect scent is taken to a lab, where the dog's reactions are tested without a handler present.
Dogs have proved their value to both the military and law enforcement, Hess says, detecting explosives, working with narcotics officers and locating missing persons and bodies. But the alleged misuse of dog-scent evidence could cast a shadow over its value to law enforcement. In the 1980s, polygraph tests came into fashion and were hailed as an important forensic tool, but their misuse and overuse prompted a negative public reaction; Mesloh fears the same could befall the use of scent evidence. "The hammer fell on polygraphy, and it never really recovered," Mesloh says. "Now, [for dog scent], the blood is in the water."