Of Dogs, Drugs, Drones, and Cocaine Science Fairs

Yesterday, the Supreme Court had its first day of oral arguments on the legality of same-sex marriage. But you know what else the high court-sters was discussing in the moments when they weren't puzzling over all those equals signs in their News Feeds?

Dog training. Specifically, training dogs to sit.

Florida v. Jardines is a case about dogs and marijuana. I know what you're thinking.

At issue is the Fourth Amendment, which guards against unreasonable searches. In 2006, Miami Detective Douglas Bartelt used his drug sniffing dog Franky to patrol the outside of the home of Joelis Jardines in Miami. The dog sat at the door, indicating the presence of drugs. The Florida courts said that the evidence against Jardines wasn't permissible because his privacy had been violated by the search. The evidence was suppressed; the Supreme Court upheld the decision 5 to 4. Some objected on the basis of narrow views of privacy, some on broader ones. But most of the Court agreed that a cop couldn't just walk up to your front door with a drug-sniffing dog.

Because — it turns out! — dogs are much, much better than humans are at sniffing out a bag of cocaine. To Justice Antonin Scalia, that would amount to the same thing as, for instance, the police using infrared cameras on an airplane (or a drone!) to peer through your roof and get a good look at your hydroponic grow lamps. That scenario was exactly what Scalia and a majority of the court rejected as unconstitutional in 2001's Kylo v. U.S, in a case involving an Oregon man who was arrested for growing marijuana. Using a dog to smell inside your house is also an unlawful tresspass. So let a thousand weed gardens bloom in the privacy of our homes; the police can't simply peek in without a warrant.

Beyond what it says about the right to privacy at home, the case points to some common miscomprehensions about dog training. For example, we think that a dog who works for the police would be well trained in many ways. Not so. Sometimes, dogs who are taught specific tasks are not taught basic manners. In court documents, for instance, Franky is described as "wild," even on leash. If Franky could learn to distinguish heroin from MDMA, I bet he could also learn to control himself around his handler. We certainly expect more than that from dogs who can hardly tell their water bowl from the toilet. But we have funny standards, I guess.

Also, we think of trained dogs as being infallible. But if a hot dog vendor had rolled up right next to Jardines' house when Franky arrived, I'm not sure where his focus would've been. There are just so many variables at play. The kind of container the drugs are in, and how long the drug has been in its container, will affect how much the substance has permeated so as to be sniffable. And even  in controlled environments, there are factors that we can't smell. For instance, the Customs Department had to stop using detergent to wash the towels it used in training because the smell was confusing the pups. Then there's the issue of the handler. A dog might smell the marijuana or whatever, and he might be giving the appropriate signals to the handler, but the handler isn't reading them properly. (Head over to Cannabis Culture Magazine for one former drug dog trainer's explination of how and why ninety percent of drug dogs and their handlers "suck at their job").

In her ruling, Judge Elana Kagan wrote that "drug-detection dogs are highly trained tools of law enforcement, geared to respond in distinctive ways to specific scents so as to convey clear and reliable information to their human partners."

Indeedy. But "highly trained" is subjective.

Drug sniffing dogs are generally trained using the same steps as one would use to train a dog to, say, sit at a curb. If you ask a dog to sit at every curb, and then reward that behavior, the very sight of the curb will eventually become a cue for "sit." You won't have to ask for it. Curb will just mean sit. It becomes a visual cue.

Average Joe dog says: If I do that butt-on-the-ground thing when she says that word, I usually get something good.

Police dog says: If I do that butt-on-the-ground thing when I smell drugs, I usually get something good.

The special thing about police dogs like Franky is that they are trained to respond to scent cues, not the aural and visual cues we usually teach to our pets. Obviously, police dogs are trained a lot more extensively than your average pet—and, to be sure, with a lot more cocaine, weed and other non-traditional training tools. But the process is essentially the same. Judge Kagan's ruling doesn't present the matter in that way, however. She writes that the drug-sniffing dog is "to the poodle down the street as high-powered binoculars are to a piece of plain glass."

I don't completely agree. Drug-sniffing dogs are just dogs. Unlike seeing eye dogs who are usually bred and trained from birth, most drug sniffing dogs start out as shelter dogs and have unknown origins. Given enough training and time, I think many a poodle could be a drug sniffing dog. (And they'd look so cute doing it). Indeed, we've trained dogs to do many things that are helpful to us -- from flushing a fox out of its hole during a hunt to leading blind people to detecting land mines. I'm fairly sure that they don't see any difference between the stuff we think of as recreational and what we think of as useful. To them, it's all just behaviors that need to be performed in order to get some kind of reward from the human.

Perhaps the most interesting part of this case is the fact that it is largely without precedent. Writing for the majority, Judge Antonin Scalia pointed out that dogs have been used as detection tools since at least the 1300s; they were in use in America at the time that the Fourth Amendment was written. "If bringing a tracking dog to the front door of a home constituted a trespass," he writes, "one would expect at least one case to have arisen during the past 800 years. But the Court has found none."

There are previous privacy cases of course. Famously, last year's U.S. vs. Jones ruled that GPS tracking a suspect's car was unconstitutional because it constituted his "property." But that reflected a narrower view of privacy than the Court used in 1967, in Katz vs. U.S., when it threw out evidence police obtained by attaching a recording device to a phone booth and established the well-known “expectation of privacy.” This new decision doesn't support that wider view of privacy, and it doesn't necessarily help us address the kind of surveillance that police departments might want to do with drones and other exciting new tracking devices that can sense things in ways that no human or dog has before—cases that courts across the country will be addressing soon. And we didn't even talk about robot dogs yet!

Something else without precedent? The detective who belongs to Franky the Famous Drug Dog also donated cocaine to his daughter's school experiment. You see, earlier this year, Detective Barteltt's fourth grade daughter Emma needed a good science fair project. The obvious choice: an experiment that involved timing the speed at which three of the family's retired dogs were alerted to the presence of an ounce of cocaine. He provided the drugs for the trial, which was conducted at his training studio under his watchful eye.

Just to summarize: despite the ruling, police dogs can smell weed through your door while the kids can play with cocaine for school. Also: not all people who are in love can legally marry.

Clearly everything is going to the dogs.

Anna Jane Grossman edits TheDogs, a blog about animal behavior, and is the co-owner of School For The Dogs in Manhattan. 

Connections