Legal
The Racial Bias of Drug-Sniffing Police Dogs
A new report from NPR is detailing the steps that some police departments are taking — and not taking — after studies in 2011 found evidence of racial bias in drug-sniffing dogs.
Drug-sniffing dogs are standard-issue equipment in American police departments, and even more so after 9/11 released a flood of extra federal cash to purchase the canine cops, which can cost upwards of $15,000 each.
But just like their human police counterparts, who studies have shown display “robust racial bias” that makes them more predisposed to arrest or shoot black people than white people, drug-sniffing dogs are susceptible to racist behavior. The dogs have some plausible deniability: It’s their handlers’ fault.
A notable 2011 study found that dogs trained to detect bombs or drugs would raise an alert even if there were no drugs present — recording more than 200 “erroneous alerts” during a review — most probably because their handler decided there were drugs present for some other reason, and “told” the innocent dog to alert with a nod or other subtle cue. Essentially, the study found that often if the handler thought there should be drugs, the dog would “obey” and make the false positive.
This is a problem. An alert from a canine unit — nicknamed “probable cause on four legs,” as NPR reported — is cause enough for an arrest, a seizure, and quite possibly the first step in the civil asset forfeiture process for the unlucky owner of the house, car, or suitcase the dog “found” to be drug-tainted.
According to some defense attorneys, this obvious flaw in the justice system is a feature, not a bug: The offices of police departments and prosecutors thrive on the cash seized in asset forfeitures, cash they can take without a criminal conviction. Police departments are often short on funds — and “[i]f they want your money, they get Fido to alert to it. And Fido alerts to everything,” a defense attorney told SF Weekly.
As NPR noted, the 2011 study set off a furor among K9 units, who defended their honor with vehement denials of any bias — and later refused to participate in any follow-up studies.
But at least some dog-handlers admit that there’s a problem and are trying to root it out. This is good, as drug-sniffing dogs aren’t going away even in the era of marijuana legalization. Instead, they’re merely being retrained to detect cocaine, methamphetamine and heroin — and traces of cocaine are on as much as 80 percent of U.S. currency.
To be better assured that dogs are reacting to drugs and not to instructions — subconscious or otherwise — from their handlers, the Pacific Northwest Detection Dog Association is retraining canine units to “scientific levels of impartiality,” NPR reported.
Retraining units are given a room with multiple potential hiding spots to suss out. But there aren’t always drugs hidden in these spots — whether there is or not is determined by a dice roll prior to the search. Sometimes there won’t be any drugs for multiple searches in a row.
Fred Helfers, the retired K9 cop who now runs the tests, explained to NPR how a drugless room would sometimes nevertheless trigger an alert. Phantom drugs? No — bias, an insistence that there just has to be drugs.
As he said:
“There were some new teams that failed that sequence,” Helfers says. “Because they didn’t trust their dog.”
He says those handlers couldn’t get past their expectation that drugs should be there. “I think they ‘overworked’ the car. Instead of going around once or twice and trusting their dog and watching their dog work, maybe they’d seen something that wasn’t there,” Helfers says.
Not every drug-dog team in the country is held to such high standards. Nor is every team retrained with snuffing out bias in mind. It’s also unclear what ramifications handlers who do alert to nothing at all risk.
Biased arrests, shootings or other high-profile instances of problem policing are just that: High profile. Public outrage over racial bias in policing has manifested in political pressure, with police chiefs losing their jobs and officers going to trial. There doesn’t yet appear to be any such reckoning for unreliable drug-dog teams.
And that, says Andy Falco, a retired K9 handler who now works as a re-trainer and as an expert witness for defense attorneys, is part of the problem.
“I think in the beginning [the bias and tendency to alert to nothing] is subconscious,” he told NPR, “but at some point, you know, how hard do you want to work, right?”
TELL US, do you think cops should use drug-sniffing dogs?