Drug Dogs Don't Even Have To Be Right Half The Time To Be Considered 'Reliable' By The Courts

from the good-news,-K9s:-even-if-you-suck-at-your-job,-you-can-keep-your-job dept

Thu, Feb 11th 2016 3:27am — Tim Cushing

All in all, this motion to suppress evidence worked out for the plaintiff, but it does little to address concerns that drug dogs are basically blank permission slips for inquisitive cops.

The defendant -- Emile Martin -- was in a vehicle driven by another person (simply referred to as "Montgomery" in the opinion). This vehicle crossed the centerline multiple times and was pulled over by Deputy Brandon Williams. The driver could not produce registration or proof of insurance, which led to the issuance of a citation… eventually. But the citation process was unnecessarily prolonged to provide the deputy with a chance to have a K9 unit brought in to sniff the car for drugs.

Based on its findings of fact, the court agrees that the stop was unduly prolonged in order to allow time for the canine and its handler to reach the scene. Prior to the point that the dog alerted, at 3:37 a.m., there was merely a hunch, but neither probable cause nor reasonable articulable suspicion, that criminal conduct was afoot. The lapse of 33 minutes from 3:04 a.m. to 3:37 a.m. for the stop in this case constituted a plainly unjustifiable seizure for that length of time under the Fourth Amendment. As noted above, when Deputy Williams returned to his cruiser with Montgomery’s driver’s license and the Grand Prix title at or shortly after 3:11 a.m., he had everything he needed to begin writing the traffic citations.

However, Williams did not begin writing the citations until 3:21 a.m., and had not completed them when Dul alerted on the vehicle following the open-air sniff at 3:37 a.m. While Deputy Williams spent some time awaiting confirmation from dispatch of the license’s validity and the results of the warrant search, that does not excuse his failure to even begin writing the citations until ten minutes after he could have done so. The stop here was unduly prolonged far beyond the time reasonably required to complete the stop’s mission.

Under the Supreme Court's Rodriguez decision, officers cannot artificially prolong traffic stops in hopes of stumbling across something "better" than a traffic violation. Once the stop's "mission" has reached its conclusion, drivers are free to go, no matter how many more questions -- or dog sniffs -- the officer might wish to pursue.

Still, a drug dog was brought in and it did alert during its "search" of the vehicle. This alert was also challenged, presumably in case the defendant's citation of Rodriguez failed to result in suppression. Data was obtained on the dog's ("Dul") "hit" rate. The data wasn't exactly a confirmation of Dul's superlative skills.

The defendant has not presented any evidence challenging the adequacy of Dul’s training and certification regimen. However, he questions Dul’s reliability based on a review of the dog’s performance record, both in training sessions and in the field. The defendant argues that Dul’s training and field performance records suggest a failure rate of up to 25%. The evidence offered on this phase of the motion is generally undisputed.

Considering law enforcement officers "ask" dogs for permission to effect warrantless searches, one would hope 75% wouldn't be an acceptable success rate. Of course, many arguments were presented by the government as to why being right only three-fourths of the time is nigh unto infallibility. According to law enforcement testimony, there are any number of reasons why a drug sniff might result in a false positive, but none of those are reasons to doubt a dog's assertions.

This is the case because officers are unable to confirm false negatives in the field (as no search is conducted), may fail to find drugs where a dog correctly alerts, and may not realize a dog has alerted based on a residual odor of drugs no longer present.

This would be one thing if law enforcement was alone in finding this acceptable. Unfortunately, the court also finds this lack of accuracy to be of little import when discussing the justification of a search. Dul may only be right 75% of the time, but the bar has been set so low by previous decisions that drug dogs whose intuition is worse than a coin flip are considered to be trustworthy generators of probable cause. (h/t Brad Heath)

Notwithstanding the dispute regarding Dul’s failure rate, the court is satisfied that in conjunction with his training and certification, his performance record amply supports the officers’ reliance on his alert to support probable cause to conduct a search. Dul’s performance record is superior to that of dogs which have been found to be reliable by other courts. See Green, 740 F.3d at 283-284 (affirming district court’s finding that dog with 43% success rate was reliable); United States v. Bentley, 795 F.3d 630, 636 (7th Cir. 2015) (accepting field detection rate of 59.5%); United States v. Holleman, 743 F.3d 1152, 1157 (8th Cir.) (57%).

The only upside here is that the Rodriguez decision will provide a remedy for those whose stops have been artificially extended to bring in drug dogs whose "alert" means nothing more than ¯\_(ツ)_/¯.

In this case, the extension of the stop resulted in suppressed evidence, not the drug dog's questionable reliability. At some point, drug dogs may start being mentioned in the same breath as other law enforcement pseudoscience -- like bite mark evidence or hair comparison. But until then, dogs that can't even manage a 50% hit rate will still be allowed to give officers permission to perform warrantless searches.

Dul-the-Dog (PDF)Dul-the-Dog (Text)

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: 4th amendment, accuracy, drug dogs, searches

37 Comments

If you liked this post, you may also be interested in...

Reader Comments

Subscribe: RSS

View by: Time | Thread

That One Guy (profile), 11 Feb 2016 @ 1:39am

Heads legal search, Tails illegal search
Given a four-legged probable cause can be the difference from a legal search and an illegal one, I'd say anything below 80-90% accuracy rate should be grounds for suppression of evidence and 'retirement' into non-police work related duties for the dog in question.

If flipping a coin wouldn't be acceptable grounds to search a vehicle, then a system, dog or otherwise with worse odds of being correct shouldn't be either.
[ link to this | view in thread ]
Ninja (profile), 11 Feb 2016 @ 4:25am

I'd be promptly fired if I had an accuracy rate like that. The problem is that the police and the courts suffer absolutely no loss from the dogs being very unreliable. With that in mind it's easy to say that a 10% accuracy rate is enough...
[ link to this | view in thread ]
That One Guy (profile), 11 Feb 2016 @ 4:36am

It's all a matter of priorities
Accuracy only matters if you think that the rights of the people against unreasonable searches is more important than the 'rights' of police to perform any searches they want. If you prioritize the latter more than the former any level of accuracy is acceptable, because it allows more searches that would otherwise be prohibited.

If a particular dog was shown to have a 99% false positive rate over the course of 100 stops, they would still say it was acceptable and valid because that's 100 searches they wouldn't have otherwise been able to perform.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 5:00am

Another proud day for the goose stepping stazi
[ link to this | view in thread ]
DannyB (profile), 11 Feb 2016 @ 5:20am

Slippery slope
If drug dogs are to be required to be capable of doing their jobs, this could lead to bad things. The next thing you know, police officers will be expected to be capable of doing their jobs.

This will not end well. Eventually it will lead to putting donut farmers out of business.
[ link to this | view in thread ]
TechDescartes (profile), 11 Feb 2016 @ 5:42am

“Dul”
A dog by any other name would smell as…dull?
[ link to this | view in thread ]
me@me.net, 11 Feb 2016 @ 6:18am

abuse of assett forefeiture
All this is, TENNESEE, is security theatre to abuse assett laws, it's nothing more than road piracy.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 6:37am

Re: It's all a matter of priorities
Why waste time and money on the pretense, just come out and say it ... little people have no rights.
[ link to this | view in thread ]
JonC (profile), 11 Feb 2016 @ 6:43am

I have a sock puppet that had completed a training regimen for drug detection, and has a very nice looking certificate indicating as much. It sounds like my sock puppet should be ready to pursue a career in drug detection.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 7:06am

Too sensitive?
I've seen drug dogs a few times signal on people that while they weren't currently carrying anything... the dog wasn't wrong that they made frequent use of illegal drugs.

For that scenario, wouldn't the dog still be right, just with poor timing?
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 8:09am

Re: Too sensitive?
The dog would be wrong because it is supposed to be alerting to the presence of drugs. These dogs have more rights that we the citizens do. If you harm a police dog, even without intent you are facing assault charges. If they strip away your rights and allow an illegal search to magically be legal with no penalty for being wrong, that is an injustice. The entire premise of search dogs nullifying rights without repercussion is so crazy that I didn't believe it when I first heard of the practice. If a dog alerts on my car, I am going to demand it be retired after the search of my vehicle turns up nothing. I might as well carry my own rights nullifying animal and do the same search on the cops car while mine is being searched. Odds are the cops car will have plenty to give me the right to search their vehicle.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 8:18am

If a search is still valid...
> ... This is the case because officers are unable to confirm false negatives in the field ...

Are they arguing that if the dog does NOT alert, the search is still valid?
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 8:19am

Considering how well the TSA did their last breach test, drug dogs are awesome. Maybe the court was keeping that in mind.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 9:11am

"He barked at a house and that was evidence enough."
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 9:12am

seems like doggie was 100% right
seeing as there was heroin found in the back seat
[ link to this | view in thread ]
Whatever (profile), 11 Feb 2016 @ 9:30am

I think you need to consider that a drug dog is providing probable cause, and not passing absolute judgement. 75% success rate (dog signals and drugs are found) is actually a pretty good result.

Remember, the dog may react to the scent of drugs that the officers are unabled to find. That could be because the drugs are well hidden (trap door or similar), it could be because the drugs were recently removed from the vehicle (like they made a delivery), or that a certain amount of drugs may be present in the car in a quantity that the police are unable to spot.

It seems entirely reasonable.
[ link to this | view in thread ]
That One Guy (profile), 11 Feb 2016 @ 9:44am

Re:
43% success rate... detection rate of 59.5%...

75% success rate is passable comparatively, in that it's at least slightly better than the 'you have more accuracy with a coin toss' of other dogs, but even then one-in-four searches are allowed that would otherwise be illegal, and that is far too high.

If dogs are going to be allowed to act as legal, on the spot 'probable cause' issuers, then their accuracy needs to be as perfect as possible, in the range of a 5-10% error rate at most before their nose is no longer given legal weight.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 10:54am

I remember watching a mythbuster episode about the various ways that people could hide something from one of those dogs. They got a police department to cherry pick their best dog and the dog was incredibly able to find whatever it was the mythbusters hid no matter how they hid it and what they did to hide it. It was indeed impressive.

The problem is that dogs are like people, different dogs are different. There are some people that are very intelligent and others that can't spell their names. I've seen some dogs that have excellent hearing and I've seen other dogs that couldn't hear a bomb exploding right next to them. To assume that just because a police department can cherry pick their best dog and it performs great that every dog they have performs the same is silly.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 11:00am

Re:
How many of the dog positive reactions resulted in no drugs found and yet the vehicle occupants were still treated as criminals?

In the eyes of LEOs everywhere, all proles, plebes and other cretins (basically everyone not rich or a cop) are criminals and they shouldn't expect their rights being respected - because they won't.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 11:02am

Re:
The dog can not hear because you exploded something right next to their ear - why would you do that?

No animals were harmed in the making of this comment.
[ link to this | view in thread ]
Rekrul, 11 Feb 2016 @ 11:34am

I wonder if a way to stop giving drug dogs so much credit wouldn't be to get an experienced handler to appear as a witness for the defense, and have them cause the dog to "alert" on both the judge, prosecutor and arresting officer.

Assuming it's a false positive, it will clearly demonstrate that the dog can be made to give a false alert whenever the handler wishes. Even if it's a true positive, I doubt the judge, prosecutor and cop would want to admit to having illegal drugs on them, so they would be forced to label it a false positive.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 12:08pm

Re: Too sensitive?
Under that logic everyone should be arrested by the first cop that sees them because everyone breaks little laws all the time
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 12:14pm

Re: If a search is still valid...
and lose the power they hold over their fellow citizens by admitting they break laws that protect the people they target.
[ link to this | view in thread ]
Anonymous Coward, 11 Feb 2016 @ 4:42pm

Re:
Because fuck the 25% that got arrested for no reason at all, right? After all, criminals will suddenly start holding up babies if police aren't allowed to shoot them, right?
[ link to this | view in thread ]
Whatever (profile), 11 Feb 2016 @ 5:31pm

Re: Re:
Again, it's not 25% getting arrested for nothing. It's 25% of the time that the dog alerts but no drugs are found during a search. Nothing more than that.
[ link to this | view in thread ]
Whatever (profile), 11 Feb 2016 @ 5:33pm

Re:
Except of course that isn't how the dogs are trained. If they trained them to generate false positives, the dog would always do it, because it's the shortest route to getting it's reward.
[ link to this | view in thread ]
BroylesJB (profile), 11 Feb 2016 @ 8:43pm

Pseudoscience
Don't forget the widely discredited fire "science" associated with arson investigations that place many innocent people behind bars.
[ link to this | view in thread ]
Anonymous Coward, 12 Feb 2016 @ 7:06am

Re: Re:
And you do not think that is happening right now.
[ link to this | view in thread ]
Anonymous Coward, 12 Feb 2016 @ 7:07am

Re: Re: Re:
Oh - there is plenty more than that if you look.
[ link to this | view in thread ]
Anonymous Coward, 12 Feb 2016 @ 10:23am

Re:
Remember, the dog may react to the scent of drugs that the officers are unabled to find.

It took me a while to figure this sentence out. So what you're arguing here is that if a dog reacts to drugs that are not found, that's a great excuse? Lemme ask you - if the dog reacts to drugs that are not found, what exactly would the officer present as evidence, other than the dog? Do you think they should put the dog on the stand to testify against the suspect?

You do realize that evidence is a real nice thing to have when you're going to treat a citizen like a criminal, don't you? Otherwise, it makes the cops look like assholes. Or more accurately, it makes them look exactly like what a growing number of the public perceives them.

And before you use your other gems as a reason to continue making excuses for the dog, please re-read the comment, specifically the part about evidence.

If drugs were removed from the vehicle, again, what do you intend to charge the suspect with? And how do you intend to collect what is no longer there?

It seems entirely reasonable.

Only if you apply cop logic as opposed to common sense.
[ link to this | view in thread ]
tqk (profile), 12 Feb 2016 @ 12:51pm

Re:
With that in mind it's easy to say that a 10% accuracy rate is enough...

Not perfect, but helpful in some situations? That sounds like a lot of things we continue using all the time.

I'm more annoyed that sneaky dog handlers can teach their animal to alert on cue, not that dogs are imperfect.
[ link to this | view in thread ]
Michael Price, 12 Feb 2016 @ 2:23pm

Re: Re: Re:
Nothing more than a search based on something that the dog alerted to but nobody else could find. In other words as far as anyone knew, they had nothing. And it can be over 50%.
[ link to this | view in thread ]
Reverend Draco, 13 Feb 2016 @ 2:33pm

4th Amendment
"I realize that the 4th Amendment requires that I produce a warrant prior to searching you or your property. . . but my dog said I can search you without one."

It's an excuse as childish - and as inexcusable - as "my dog ate my homework."
[ link to this | view in thread ]
Rekrul, 13 Feb 2016 @ 5:03pm

Re: Re:
Except of course that isn't how the dogs are trained. If they trained them to generate false positives, the dog would always do it, because it's the shortest route to getting it's reward.

Various experts have said that it's possible for an experienced handler to cause a dog to give an alert by the tone of their voice and the commands that they give them. Not to mention that there really isn't any objective standard to how the dogs are trained.

https://www.youtube.com/watch?v=RjRzOXgweVU
[ link to this | view in thread ]
Anonymous Coward, 15 Feb 2016 @ 6:31am

Re: Re: It's all a matter of priorities
Why waste time and money on the pretense, just come out and say it ... little people have no rights.

They don't say it because they hate the truth.
[ link to this | view in thread ]
Anonymous Coward, 15 Feb 2016 @ 6:34am

Re: Re:
If they trained them to generate false positives, the dog would always do it,...

There you have it from Whatever, folks, dogs are 100% reliable.

What a load.
[ link to this | view in thread ]
AnonCow, 17 Feb 2016 @ 2:57pm

I think that dogs are profiling. We just don't know it.
[ link to this | view in thread ]