Judge Doesn't Buy CBP's Argument That Dog Can 'Smell' The Difference Between Concealed And Unconcealed Humans

from the can-also-tell-if-weed-purchased-without-a-prescription-just-by-sniffing dept

If there's an unreasonable, warrantless search happening, there's a good chance Deputy (literal) Dog is on the case. Cops love their K9 buddies, mainly because nearly any motion or noise a police dog makes can be construed as an "indication" or an "alert." It's a blank permission slip, signed with a paw print.

In reality, though, contraband sniffing dogs aren't preternaturally smart. They are still man's best friend and they love pleasing their handlers. The reward that comes after every "alert" further impresses on the animal the value of "alerting."

This is one of law enforcement's major blind spots. Because contraband-sniffing dogs locate contraband at least part of the time, they're presumed to be infallible/miraculous. No one seems to be more amazed by these animals' skills than law enforcement officers. (Well, them and far too many judges…) But it's not often that you see a law enforcement officer claim his canine partner can do something anyone else would consider impossible, including an expert on law enforcement dogs.

On the way to having the fruits of his search suppressed, CBP Agent Aaron Miranda stated out loud -- and on the record -- that his dog could "smell" the difference between CONCEALED and UNCONCEALED humans.

Agent Miranda testified that Boeli is trained to find things by their odor, including concealed humans, marijuana, heroin, methamphetamine, and all derivatives of those drugs. Agent Miranda testified that when Boeli comes across an odor that he is trained to find “he changes posture, his demeanor. His breathing changes. He tenses up, starts breathing out his nose, closes his mouth. His whole body changes up.”
Now, it's one thing to sniff out concealed humans when no other visible humans are in the vehicle. That I can believe. But Miranda says Boeli can do this even when other, unconcealed people are in the car… or even when that car is in motion.
Agent Miranda testified that Boeli is trained to detect concealed people in different scenarios including inside a moving vehicle. Agent Miranda testified that Boeli can “distinguish cars with people in them from cars with concealed people in them.”
Do tell.
Agent Miranda testified that “I am not an instructor, so I don’t really know how they do the training or how he does it. It amazes me too, but it works.”
PLEASE HOLD YOUR LAUGHTER UNTIL THE END OF THE PRESENTATION.

Miranda doesn't know how this happens or how it works. IT JUST DOES. Here's his definition of "works."
Agent Miranda testified that on a typical day thousands of cars come through the checkpoint and that Boeli could alert “anywhere from two to six or seven” times. Id. at 46. Agent Miranda testified that Boeli has found “five to seven human beings and twelve narcotics busts” over their five-year career. Id. at 105.
Math time:

Five years at ~200 working days per year multiplied by… oh, let's just call it 1,000 cars per day. That's 1,000,000 vehicles he and Boeli have stared/sniffed at. And Boeli's superior sniffing skills have netted the team "5-7 humans" and "12 drug busts." Generously assuming each illicit human was discovered in a separate vehicle, that's a hit rate of .002%.

Now, most vehicles traveling this stretch of California highway will contain nothing illegal, so the hit rate can't really be determined with any accuracy. But five years with 19 busts in an area (Highway 86 outside of Westmorland, California) where one would expect a higher rate of drug/human trafficking than further inland, this still seems incredibly low. And yet, Agent Miranda can't say enough good things about Boeli's ability to sniff out concealed humans in moving vehicles driven by unconcealed humans -- even when an expert on contraband-sniffing dogs contradicts his claims.
At the evidentiary hearing, Defendant presented the testimony of an expert witness who operates a business training dogs for law enforcement and security organizations.

[...]

Defendant’s expert was asked, “In your opinion is it possible to train a dog to alert to concealed human smell but ignore or differentiate other human smell in the same vehicle?” The expert answered, “Not inside the same vehicle. If there is somebody inside the vehicle and somebody in the trunk, or you take the person out of the trunk, it doesn’t matter. It is all – it is the same to the dog.”
The court found Miranda's testimony similarly unconvincing. When the burden is on the government to prove that its search was reasonable and justified by a trained dog's "alert," it has to offer something better than ¯\_(ツ)_/¯
In this case, Defendant has challenged the reliability of the training program and the circumstances surrounding the alert behavior. Defendant challenged Boeli’s ability to distinguish cars with people inside from cars with people concealed inside during the cross examination of Agent Miranda. Agent Miranda was not able to provide any evidence to support Boeli’s training to detect concealed people in moving vehicles. Agent Miranda testified “I don’t really know how they do the training or how [Boeli] does it. It amazes me too, but it works.” Agent Miranda stated, “I trained with my dog on these odors, but the actual training that goes into him knowing the difference is something that the... instructors do that and stuff. I don’t know how the dog would know the difference, but they just do. It is pretty amazing.”
Well, a trainer was asked and he stated that the "amazing" behavior witnessed by Agent Miranda simply does not exist. The most amazing act here was performed by the CBP and Agent Miranda, who managed to make all of the evidence they obtained with their warrantless search vanish into the ether.

Hide this

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: aaron miranda, border patrol, contraband, dogs, smelling


Reader Comments

Subscribe: RSS

View by: Time | Thread


  • identicon
    Anonymous Coward, 6 Jan 2016 @ 1:03pm

    a better analysis:
    Five years, 200 days per year, 2 alerts per day = 2000 alerts
    7 humans + 12 drug busts = 19 actual hits.

    Now, that's an accuracy rate of ~1%.

    That does not meet the burden of "probable cause".

    link to this | view in chronology ]

    • icon
      TechDescartes (profile), 6 Jan 2016 @ 1:19pm

      Re: The Power of Imagination

      Close. It's actually √-1.

      link to this | view in chronology ]

    • identicon
      Anonymous Coward, 6 Jan 2016 @ 1:37pm

      Re:

      Yeah, that's what I was thinking. I don't care so much about what percentage of cars he alerts to. I care about what's IN the cars he alerts to, and apparently, literally 99% of the time the answer is nothing (and that's using the two alerts per day figure, when the officer testified it's "anywhere from two to six or seven".) A 1% chance (of people OR drugs - they wouldn't even know what exactly they were looking for based on the alert) is not anywhere near probable cause.

      link to this | view in chronology ]

  • icon
    JonC (profile), 6 Jan 2016 @ 1:20pm

    You've got it all wrong. These dogs aren't trained to find concealed humans by smell, they're trained to have x-ray vision.

    link to this | view in chronology ]

  • icon
    Roger Strong (profile), 6 Jan 2016 @ 1:23pm

    "Months later, with new doubts cast on claims that the dog could detect and 'alert' to a 'backdoored encryption system that's actually secure', Agent Miranda and his dog Boeli were hired by Donald Trump to detect and block terrorist data packets on the internet."

    link to this | view in chronology ]

  • identicon
    Anonymous Coward, 6 Jan 2016 @ 1:38pm

    We're regressing back to the age of superstition and old wives tales.

    I bet this officer's knee aches every time someone with drugs crosses his path.

    link to this | view in chronology ]

    • identicon
      Anonymous Coward, 6 Jan 2016 @ 3:46pm

      Re:

      Assuming the officer's knees are in reasonably good health, that would probably be more accurate.

      link to this | view in chronology ]

    • identicon
      Anonymous Coward, 6 Jan 2016 @ 4:37pm

      Re: superstition etc

      until he took an arrow to the knee

      link to this | view in chronology ]

  • identicon
    Anonymous Coward, 6 Jan 2016 @ 1:39pm

    >>Five years at ~200 working days per year multiplied by… oh, let's just call it 1,000 cars per day. That's 1,000,000 vehicles he and Boeli have stared/sniffed at. And Boeli's superior sniffing skills have netted the team "5-7 humans" and "12 drug busts." Generously assuming each illicit human was discovered in a separate vehicle, that's a hit rate of .002%.

    According to government estimates, there should have been about 100,000 drug cars that the cop and dog stopped. I distrust that number because of its source, but let's run with it for a moment. Let's further assume that every car they tagged with hidden people also had drugs.

    That implies the dog missed in excess of 99% of all drug traffickers while they were on duty. Why are this person and dog even still employed??????

    link to this | view in chronology ]

  • identicon
    Anonymous Anonymous Coward, 6 Jan 2016 @ 1:42pm

    Dogs Smell?

    If you don't think dogs can smell, just get one wet...

    link to this | view in chronology ]

  • identicon
    Personanongrata, 6 Jan 2016 @ 2:09pm

    Can You Detect This?

    Miranda doesn't know how this happens or how it works. IT JUST DOES.

    It works with PFM.

    The acronym PFM is a highly technical term used in various technical/science/engineering disciplines. It stands for:

    Pure Fucking Magic.

    http://www.urbandictionary.com/define.php?term=PFM

    PS It reads as if Agent Miranda would have trouble detecting his own ass if he was sitting on both of his hands.

    link to this | view in chronology ]

  • identicon
    DCL, 6 Jan 2016 @ 2:48pm

    this is ruff issue...

    I can see that there may be a small possibility that a clever dog can detect some difference between a concealed person in certain situations.

    People hiding in cars generally are in confined spaces for long periods and that could lead to some scenarios where a specific odor combos leak out... maybe BO, urine, halitosis, or maybe extra CO2 (who knows what a specific dog can really detect).

    But overall it still doesn't pass the sniff test for probable cause.

    link to this | view in chronology ]

    • identicon
      David, 6 Jan 2016 @ 3:05pm

      Re: this is ruff issue...

      If you squeeze 40 people into a truck, full of fear, on the run for several weeks without possibility of changing, partly with urine in corners of the floor or just running down people's legs, possibly with corpses...

      I can imagine that a dog can smell such a vehicle of pure terror even when driving by slowly. And humans, if they just trusted their nose, would probably also smell something when standing at the side of the truck.

      A single courier or high-up person changing state? Most likely not.

      link to this | view in chronology ]

      • identicon
        Anonymous Coward, 7 Jan 2016 @ 7:00am

        Re: Re: this is ruff issue...

        I smell something, having read your post.

        link to this | view in chronology ]

    • icon
      Roger Strong (profile), 6 Jan 2016 @ 3:17pm

      Re: this is ruff issue...

      Keep in mind that the claim is that the dog can detect concealed humans - and distinguish them from non-concealed humans - in cars driving past.

      link to this | view in chronology ]

  • icon
    Andrew (profile), 6 Jan 2016 @ 5:29pm

    Until about 18 months ago, I lived in the rural Georgia area of Jasper County. It's pretty unremarkable except for its Tv/movie references, specifically its main town and surroundings where the location where the 91 film "My Counsin Vinny" was filmed (a film where two city boys were arrested for murdering a clerk at a store, based on the testimony of eyewitnesses, and facing the death penalty. Marissa Tomei won an oscar for her role, also the last role of Fred 'Munsters' Gwinn)

    For those wondering 'why should I care' is that much of the film consists of defence attorney Joe Pesci cross examining/testing the various witnesses on the stand, to accurately assess their ability to tell the truth.
    We need to start doing this for drug dogs. A basic competency test. 5 test cars, some, all or none of them may have one or more pieces of contraband on them. The drug dog and its handler do a 'sweep' on them, and note their suspicions down in specifics. Missed alerts (say rover alerting on the trunk of a car with no contraband) will go significantly against the credibility of the dog, as would missing packages. Miss more than half of them, toss any case involving the dogs alerts, as they're essentially useless.

    Needless to say, the tests would have to be run by outsiders, no pre-arranging the stashes, or tipping the wink when they're close by 'friends' of the dog's handler.

    It's time sniffer dogs were actually made to prove their competency in blind testing, rather than treated like 'psychics' (where their hits are based as much on reading people as competency, and misses are always forgotten/glossed over)

    link to this | view in chronology ]

    • identicon
      Anonymous Coward, 6 Jan 2016 @ 6:39pm

      Re:

      Maybe even better is a dogs ACTUAL alert to hit rate per year in the field. If the actual percent is below 80% then the dog goes back for re-training; but can only be re-trained 3 times in the dogs lifetime.

      link to this | view in chronology ]

    • identicon
      Anonymous Coward, 6 Jan 2016 @ 6:54pm

      Re:

      Whatever happened to the Amazing Randi? Or does he only deal with spoon-related bullshit?

      link to this | view in chronology ]

  • identicon
    Anonymous Coward, 6 Jan 2016 @ 5:32pm

    In my many years on this earth, I've observed that uneducated people tend to believe something simply because they really want it to be true, and they'll ignore or even argue against all logic that proves otherwise, all because they can't understand said logic in the first place. Basically, they're clueless as to how clueless they are. This ignoramus Miranda seems to be one of those people.

    link to this | view in chronology ]

  • icon
    That Anonymous Coward (profile), 6 Jan 2016 @ 11:16pm

    This rock not only can repel tigers, but it will find you bad people.
    Please send your checks made payable to cash to TAC c/o TechDirt.

    Perhaps, maybe just maybe, we need to demand more from the justice system. I dunno how the dog does it but it works! This is not a legit basis for anything, and that officer should be fired for believing in magic over actual evidence.

    We really are seeing more of the, well they were bad people so how we got there doesn't matter and that is a bad way to let them proceed. Everyone can make someone elses bad person list, and we have to have better standards than the ends justify the means.

    link to this | view in chronology ]

    • icon
      Ninja (profile), 7 Jan 2016 @ 3:06am

      Re:

      Right? I have never been in the middle or even near some terrorist attack or anything related. I always have a rabbit foot with me which means rabbit foots are good against terrorism.

      Never mind the fact I've never been to any point of interest for terrorists in my life, that's a non issue.

      link to this | view in chronology ]

  • icon
    Éibhear (profile), 7 Jan 2016 @ 3:31am

    Training

    '... Boeli comes across an odor that he is trained to find "he changes posture, his demeanor. His breathing changes. He tenses up, starts breathing out his nose, closes his mouth. His whole body changes up."'

    So the dog is trained to identify many specific odours, but *not* trained to react in a specific way to a particular smell? How is that?

    - Officer Miranda, what odours are Boeli trained to identify?

    - Humans, concealed and not, pot, heroin, amphetamine, and so on.

    - Not Cologne?

    - Yes Cologne, but he's trained not to alert when he smells it, as it's not a crime to wear Cologne [at least, most Colognes].

    - Your honour?

    - The witness will stick to answering the questions.

    - Yes, your honour.

    - Officer, how would Boeli alert to marijuana?

    - He would sit still, look straight at the area where the odour's coming from and whine twice.

    - And amphetamines?

    - He would do the same, except instead of whining, he would lift and drop his left-front paw twice.

    - OK. And a concealed human?

    - Boeli is trained to bark twice and then sit still and look straight at where the odour's coming from.

    - Officer Miranda, is this how Boeli alerted you to the fact that the defendant was in the vehicle?

    - No. On that occasion, Boeli changed his posture, his demeanor. His breathing changed. He tensed up, started breathing out his nose, closed his mouth. His whole body changed up.

    - Is that what Boeli is trained to do when he identifies a "concealed human"?

    - No.

    - Then, how did you know that he was alerting you to something?

    - Ehhm....

    - He didn't give you an alert, did he?

    - Ehhmmm....

    link to this | view in chronology ]

  • identicon
    DanD, 7 Jan 2016 @ 3:47am

    Counting Cars

    So, there are five people on the front and back seats, and one midget in the trunk. That dog can tell that there's a midget in the trunk by smell? This means that the dog can count, at least to six. I wonder where that canine got his CPA?

    DanD

    link to this | view in chronology ]

  • identicon
    Lisboeta, 7 Jan 2016 @ 5:01am

    This would be funny, if it weren't so damned scary. I think the dog-trainer must've had a harder job 'training' Miranda to work with dogs than he had in training Boeli.

    link to this | view in chronology ]

    • icon
      Ninja (profile), 7 Jan 2016 @ 6:25am

      Re:

      Oh this would be fun to watch. I definitely would like to see somebody trying to train this cop. And maybe release some lousy comedic movie about it afterwards with big bright sayings: "BASED ON A REAL STORY".

      link to this | view in chronology ]

  • icon
    Jeremy Lyman (profile), 7 Jan 2016 @ 5:14am

    Think of how much time and money they could save if they'd give up training dogs and just use police issue Dowsing Rods.

    link to this | view in chronology ]

  • icon
    andrew_duane (profile), 7 Jan 2016 @ 6:19am

    link to this | view in chronology ]

  • identicon
    Anonymous Coward, 7 Jan 2016 @ 8:51am

    Let me guess he does whatever the agent tells him to do after the fact.

    link to this | view in chronology ]


Follow Techdirt
Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Discord

The latest chatter on the Techdirt Insider Discord channel...

Loading...
Recent Stories

This site, like most other sites on the web, uses cookies. For more information, see our privacy policy. Got it
Close

Email This

This feature is only available to registered users. Register or sign in to use it.