Quantcast

«

»

Jul 03 2012

Fed. Court: 25% Alert Rate Good Enough for Drug Dogs

I’ve written before about the many problems with the use of drug-sniffing dogs, the most obvious of which is that they are usually wrong when they “alert” to the presence of drugs. But a federal court has ruled that it doesn’t matter if they are wildly inaccurate, an alert is still probable cause.

The nose of a drug-sniffing police dog is not so sharp, but it’s good enough to support cocaine charges against Herbert Green.

That was the opinion of federal Judge Glen Conrad, who denied a motion this week to suppress the drugs found in Green’s sport utility vehicle with the help of a police dog named Bono.

Green’s lawyer had argued that Bono’s track record — drugs were found just 22 times out of 85 “alerts” by the dog — was so poor that police lacked probable cause to search Green’s SUV…

Bono “may not be a model of canine accuracy,” Conrad wrote in an opinion filed Thursday in U.S. District Court in Roanoke.

However, the judge ruled that other factors, including the dog’s training and flawless performance during re-certification sessions, were enough to overcome a challenge raised by Green’s attorney, public defender Randy Cargill.

This is simply nuts. The dog has a certificate — who cares if he’s wrong 75% of the time!

27 comments

1 ping

Skip to comment form

  1. 1
    theschwa

    The certificate has a blue ribbon attached to it! How can you question its legitimacy??

  2. 2
    DaveL

    I think the court here is using a novel definition of the word “probable”.

  3. 3
    unbound

    About on par for polygraphs that are still used today. Most government agencies still use those and think the results are okay too…

  4. 4
    theschwa

    I bet if the dog “alerted” in way that was meant to indicate “no drugs here, move to next car” the results would be different.

    False positive? OK.
    False negative? Euthanize.

  5. 5
    Blondin

    I wonder how they do the certification. Is it like a dowsing test where the handler knows where the drugs are or is it properly blinded? It might be interesting to see a Mythbuster episode on this. I know they did one on tracking dogs but that might be quite different.

  6. 6
    Lowcifur

    In the court’s defense, the dog is adorable.

  7. 7
    JustaTech

    I’m all for using dogs (and their noses) to find things. But I do tend to think that they are going to be most useful (and have a higher success rate) at finding things that make sense to them. I.e. whole animals. Dogs tend to be good at tracking animals because it is a natural instince that humans have bred to enhance for millenia because it is useful. (Tracking a pig you are going to eat and tracking a lost child have the same basic principle.) Finding plant matter is just going to be a lot harder because they don’t care the same way.

    Not to mention, those drugs aren’t going to move themselves, so there’s less urgency, so there’s no excuse for “fast and slobbery but gets the job done”.

  8. 8
    augustpamplona

    However, the judge ruled that other factors, including the dog’s training and flawless performance during re-certification sessions, were enough to overcome a challenge raised by Green’s attorney, public defender Randy Cargill.

    I’m going to guess that a “flawless performance” during re-certification sessions is about as meaningful as a dowser’s flawless performance during non-blided testing.

    In some cases where nothing was found after an alert by Bono, police later determined that drugs had been in the vehicle earlier, likely leaving an odor the dog was trained to detect, Neese said.

    Taking those cases into account, Conrad found that Bono’s accuracy rate was at least 50 percent.

    I’m going to guess that “police later determined that drugs had been in the vehicle later” means that the floor mats were dirty and they arbitrarily called the dirt “drug residue”.

    The problem here is that if they had thrown this one out it would have become a precedent (as it should) and would eventually serve to question every single future drug search based solely on probable cause provided by a sniffing dog (unless it serves as a wake-up call to all of the trainers & handlers about how to not train a drug dog to respond to cues which I find highly unlikely). The judge probably just doesn’t wan to do that.

  9. 9
    JustaTech

    @Blondin: Actually, they did do a “how to evade the drug sniffing dog” episode. (It was combined with a follow-up to the evade-the-tracking-dog episode.)

    Basically, as long as the dog isn’t distracted (bitch in heat) and finds the drugs quickly, they’re pretty good (in a very stylized testing situtation) at finding their training smell.

    But as with all dogs, it’s all about the handler.

  10. 10
    Abby Normal

    I’d have to agree with the judge. If the dog has been through retraining and been recertified since those inaccurate positives, then I can see that as sufficient reason for probable cause. If it had been a breathalyzer giving false positives and it went back to the manufacturer for repair and recertification then I wouldn’t count its old score against it. Unless there’s additional information showing that retraining is ineffective, either in general or for this unit specifically, or the certification process is unreliable, then I think the judge made the right call.

  11. 11
    Robert B.

    Um. Is there any standard on how probable “probable cause” has to be? Is a policeman fired if his “probable cause” turns up empty 3/4 of the time?

    25% accuracy sounds low, but if the prior probability of someone carrying drugs is small enough, the dog might still be strong evidence. If you’re looking for something sufficiently rare, even a good test will give more false positives than true positives, just because you do so many more tests on negative subjects.

    (What percent of people actually carry illegal drugs in their cars? Do we know? AFAIK, it could actually be 25%, in which case the dog is just picking cars at random.)

  12. 12
    d cwilson

    I’m going to guess that “police later determined that drugs had been in the vehicle later” means that the floor mats were dirty and they arbitrarily called the dirt “drug residue”.

    Probably not even that. Most paper money in circulation has drug residue on it. Some of it sticks our hands every time we handle money. It’s entirely possible that thousands of people have detectable amounts of drug residue built up in their cars after a few years of driving.

  13. 13
    namelesscynic

    Couple of questions come up here.

    1. Green had $7K in cash, and 1.5 kilos of cocaine. So he was just out for a Sunday drive?

    2. Ruling out cases where the dog found residue that a standard search couldn’t see, dog’s accuracy was 50%. Is that good enough? (And even if it isn’t, say that the dog has a 25% accuracy rate. I’m thinking that the number of cars with drugs in them is significantly less that 25%, so he’s still scoring ahead of the curve.)

    3. What would you like the police to use? What weapons, exactly, should they deploy in the “war on drugs”?

  14. 14
    Bronze Dog

    I’m thinking Clever Hans quite a bit. The dogs would be motivated to do what their handlers want, which would likely include false positives. Do they double-blind the tests? Can they train the dogs toward honesty over pleasing their handler? I doubt it.

  15. 15
    Infophile

    @13 namelesscynic:

    In response to point 3, we really shouldn’t be expected to present an alternative when we point out that method X doesn’t work well. Most people around here are opposed to the war on drugs. I see it as doing far more harm than good (and to be honest, I can’t even think of an example of some good it’s actually doing). Allowing dogs alerting to be considered probable cause just gives police another tool to harass people they don’t like – there have been multiple cases reported where police would claim a dog alerted when it was conveniently out of view of their dashboard camera. Even if the search doesn’t find anything, they still get to inconvenience the car’s owner. (And in some cases, it’s even been proven that they plant drugs.)

  16. 16
    Modusoperandi

    “He’s a great mechanic.”
    “But your car keeps breaking down right after you leave his shop.”
    “Yes, but it runs just fine while it’s there.”

  17. 17
    namelesscynic

    @Infophile,
    Doesn’t matter if you’re opposed to the War on Drugs. It’s happening, whether you like it or not. And they aren’t going to end it just because you happen to think their methods are less than effective.

    Sandia Labs has a prototype chemical sniffer that’s prohibitively expensive, drug-sniffing wasps are just silly, and no matter what the method, if you happen to be the one-in-a-million who ends up with a cop willing to plant drugs, it doesn’t matter if there’s a dog involved or not. (And yes, the cop that crooked is extremely rare, regardless of what your truthiness wants you to believe.)

    Fine, dogs aren’t 100% effective. But they have their uses, and they’re better at finding drugs than not having any tools at all.

  18. 18
    augustpamplona

    namelesscynic wrote:

    Fine, dogs aren’t 100% effective. But they have their uses, and they’re better at finding drugs than not having any tools at all.

    Roadblocks and searching everyone is effective. It’ll catch close to 100% of the drug carriers in the population of people intercepted. That doesn’t mean that we should do this in a free society. We have to draw the line somewhere and I do not think that the dogs getting it right 25% of the time is the right place to draw the line.

    It’s even possible (maybe even rather likely) that all we have going on here, most of the time, is a gut feeling. Even if we cannot always point to the source of a gut feeling, gut feelings are formed from real, subtle cues and not in a vacuum. Nevertheless, gut feelings, without any articulable reasons to back them up, are not considered probable cause. Certainly, the police officer may observe the suspect seemed nervous and use that as the pretext to call a K-9 but the original observation of nervousness in itself is not sufficient for probable cause (if it was, they would not be calling the dog). If you are using the drug dogs to “confirm” the gut feeling (i.e. suspect seemed nervous because they have something to hide) it makes a huge difference whether the dogs are truly acting as a detection mechanism or as a way of responding to the (conscious or unconscious) cueing of the handler. It might make sense to use the former as probable cause whereas the latter should never be used as justification for probable cause since it is adding nothing to the hunch.

    It is apparent that what we may be seeing here is not a demonstration of effectiveness of these dogs but rather a demonstration of the effectiveness of the hunches of police officers. Your reasoning, thus, does not argue for the admittance of dog sniff evidence as probable cause but rather for the admittance of human hunches as probable cause (under the pretext that the dog agreed*). Are you sure you want to head down that path?

    * Because he’s been trained to agree –intentionally or not.

  19. 19
    kenreilly

    All of you seem to be oblivious to the loss of your 4th amendment rights. The knowledge of the police that anytime the wish to search someone they have a 75% chance of doing so makes a complete mockery of the concept of reasonable search

  20. 20
    F [i'm not here, i'm gone]

    75% wrong. I wonder if this same rationalization explains police weapons qualification and other procedure. (Well, they did absolutely smashing jobs on tests!)

  21. 21
    Infophile

    @19 kenreilly:

    Probability doesn’t work that way. I can’t say what the exact percent should be without knowing the overall prevalence of people having drugs in their cars and the false negative rate here, but it’s not a simple 75%.

    But add to that the fact that the police can simply claim a dog alerted when it didn’t, and 75% is probably too low realistically.

  22. 22
    namelesscynic

    Adorable, by the way, how you all continue to ignore one minor, minor, almost insignificant point.

    1. Green had $7K in cash, and 1.5 kilos of cocaine. So he was just out for a Sunday drive?

    (Wow. I almost feel like I’m quoting somebody here…)

  23. 23
    augustpamplona

    namelesscynic wrote:

    Adorable, by the way, how you all continue to ignore one minor, minor, almost insignificant point.

    1. Green had $7K in cash, and 1.5 kilos of cocaine. So he was just out for a Sunday drive?

    It’s not an almost insignificant point, it is a completely insignificant point so no one is ignoring it (we know it’s there but it just does not happen to matter). This is about Green’s attorney requesting that the exclusionary rule be applied on the grounds of there having been no probable cause to search and the judge disagreeing.

    The fact that he was carrying 1.5 kilograms of cocaine bears no relevance as to whether police had probable cause to search in the first place. None whatsoever. That’s not how it works. If it did work that way, if finding out someone is guilty after doing something improper made the impropriety proper, then there would be no such thing as an exclusionary rule.

  24. 24
    Rick Pikul

    @Blondin: Actually, they did do a “how to evade the drug sniffing dog” episode. (It was combined with a follow-up to the evade-the-tracking-dog episode.)

    Basically, as long as the dog isn’t distracted (bitch in heat) and finds the drugs quickly, they’re pretty good (in a very stylized testing situtation) at finding their training smell.

    Next time you watch that episode, pay close attention: On several of the tests the only reason the dog succeeded was because the handler knew that there was something to find.

    One thing you can’t trust on Mythbusters is the result on any ‘beat the cops’ myth. I’ve seen quite a few where they rig the test, (generally by testing the wrong thing[1]), or bend over backwards to declare the myth busted.

    This is understandable: If they started saying things like “this way of beating the cops works”, I think they would quickly lose access to various police resources they need for other myths.

    [1] e.g. the ‘Beat the Breathalyzer’ myths. About half of the techniques were things to do to keep the breathalyzer from being used in the first place.

  25. 25
    Bronze Dog

    Why are you going on about Green, namelesscynic?

    This is about the inaccuracy and abuse potential of drug sniffing dogs. Green is now just irrelevant historical trivia about how the dog came to be tested. If Green is innocent, it changes nothing about the abuse potential of drug sniffing dogs. If Green is guilty, it changes nothing about the abuse potential of drug sniffing dogs. Why are you allowing this irrelevancy to distract you?

  26. 26
    georgewiman

    I think they should name all the dogs Clever Hans because they seem to be channeling their trainers’ intentions.

    The fact that the guy was carrying drugs doesn’t justify using a method with a 75% error rate. That boils down to; “We just search whoever we feel like”.

  27. 27
    georgewiman

    OMFG – I got home yesterday and the drug-dogs episode of Dragnet was on TV. In it supercops Joe Friday and Bill Gannon introduce the idea of drug-sniffing dogs. Watch the whole episode on Hulu to understand how this was sold to the public. http://www.hulu.com/watch/55145

  1. 28
    » Drug dogs and the 4th Amendment Decrepit Old Fool

    [...] Federal Court: 25% Alert rate good enough for drug dogs. Drug-sniffing dog Ginger and supercop Joe Friday from Dragnet. Click picture to watch entire episode for free on HULU [...]

Leave a Reply

Switch to our mobile site