Focusing On The Wrong Thing

--by Sebastian H

I have a constant frustration with the legal system: that its mechanisms often cause a focus on the wrong thing.  I don't mean that it focuses in areas I think are unimportant--though that happens too.   I mean that even if it is looking in an important area, the structure of the legal system leads it to deal with the issue in the wrong way.

There is a certiorari petition before the Supreme Court on the question of whether or not bringing a drug sniffing dog to somone's home counts as a police search.  It is a legal problem at all because using drug sniffing dogs on a person's car has been deemed "not a search" under the 4th amendment.  This is a surprising result in itself, but leave that aside for the moment.  So having discovered that using a drug sniffing dog around a car isn't a search, some police officer decided to use it around a house.  This ends up implicating all sorts of problems, because constitutionally speaking homes are specially protected areas when it comes to police searches. 

So now various high courts have to decide how to distinguish the case of the drug sniffing dog near a home from a drug sniffing dog near a car.  Great.  But that is the wrong question entirely.  It turns out that drug sniffing dogs have an incredibly high false positive rate--they alert many more times than drugs are actually found once a search is done.  Worse the dogs seem to alert much more often if their handler suspects there should be an alert--independent of whether or not there are actually drugs present.  So the dog launders the police officer's suspicion instead of actually adding new information.

Dr Lit asked 18 professional dog handlers and their mutts to complete two sets of four brief searches. Thirteen of those who participated worked in drug detection, three in explosives detection, and two worked in both. The dogs had been trained to use one of two signals to indicate to their handlers that they had detected something. Some would bark, others would sit.

The experimental searches took places in the rooms of a church, and each team of dog and human had five minutes allocated to each of the eight searches. Before the searches, the handlers were informed that some of the search areas might contain up to three target scents, and also that in two cases those scents would be marked by pieces of red paper.

What the handlers were not told was that two of the targets contained decoy scents, in the form of unwrapped, hidden sausages, to encourage the dogs’ interest in a false location. Moreover, none of the search areas contained the scents of either drugs or explosives. Any “detections” made by the teams thus had to be false. Recorders, who were blind to the study, noted where handlers indicated that their dogs had raised alerts.

The findings, which Dr Lit reports in Animal Cognition, reveal that of 144 searches, only 21 were clean (no alerts). All the others raised one alert or more. In total, the teams raised 225 alerts, all of them false. While the sheer number of false alerts struck Dr Lit as fascinating, it was where they took place that was of greatest interest.

When handlers could see a red piece of paper, allegedly marking a location of interest, they were much more likely to say that their dogs signalled an alert. Indeed, in the two rooms where red paper was present and sausages were not, 32 of a possible 36 alerts were raised. In the two where both red paper and sausages were present that figure was 30–not significantly different. In contrast, in search areas where a sausage was hidden but no red piece of paper was there for handlers to see, it was only 17. 

The dogs, in other words, were distracted only about half the time by the stimulus aimed at them. The human handlers were not only distracted on almost every occasion by the stimulus aimed at them, but also transmitted that distraction to their animals–who responded accordingly. To mix metaphors, the dogs were crying “wolf” at the unconscious behest of their handlers.

Economist

A tracking study was done of drug sniffing dogs in Illinois which found that the searches their 'alerts' triggered found no evidence of drugs 56% of the time.  For Hispanic people searched as a result of the 'alerts' there was no evidence of drugs 63% of the time.  Chicago Times

This is a problem because drug sniffing dogs are supposed to be used as a tool leading to probable cause for a search.  The dog sniffs are used when the police officer does not have probable cause to search someone he is suspicious of.  If the dog alerts, he now has probable cause and can search the suspect, the suspect's car, and if the state attorney general offices get their way they can search a home.

But in Illinois, a dog sniff doesn't lead to discovery of evidence of drugs even 50% of the time.  Dr. Lit's experiment suggests that the false positive for the sniff alert is extremely high.  Essentially the dog can pick up on the suspicions of the handler, and alerts on that basis.  But if the suspicions of the police officer aren't enough for a probable cause search, they shouldn't be enough for a probable cause search when laundered through the dog 'alert' either.  The dog alert is relevant only if it provides strong additional information.  That is the issue which the law should be focusing on.  Fine line distinctions between car 'non-searches' and house 'non-searches' are irrelevant if the dog isn't providing a reliable independent factor for probable cause. 

The fact is that in Illinois (the only state to track it comprehensively) even if a police officer already suspects you of being involved in the drug trade, and then gets a drug sniffing dog out, and then the dog gives an alert, the police officer only finds drugs or evidence of drugs 46% of the time (27% of the time if you're Hispanic).  If a police officer suspected a Hispanic person of being involved in the drug trade, didn't have independent probable cause for a search, but flipped a coin twice and only searched when it came up heads at least once, would we say that the coin flip produced probable cause? 

 

See also Radley Balko on the topic in Reason earlier this year

Comments

[bobbyp]

Seb,

Very interesting stuff....It is not surprising that the handlers unconsciously influence their dogs.

But as to what the 'real problem' is, isn't it the fact that a dog sniff should be considered a search to begin with under any reasonable standard? The auto decision strikes me as appalling...what decision was that, and what was the reasoning of the bench?

[Mike Schilling]

What's the added incidence of car-sniffing-dog false positives when it's parked near a fire hydrant?

[Free Lunch]

The courts have completely made a hash of probable cause and automobiles and every explanation they make for getting further from real warrant requirements is more convoluted and silly. Only "national security" contains more absurd excuses for ignoring the Fourth Amendment.

"The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."

This is simple legal writing. It doesn't make exceptions for young males of non-European heritage, the most common victims of Fourth Amendment shortcuts. It doesn't say carriages are exempt or out-buildings on the farm. It doesn't say the cops get to keep anything they have searched and seized for their own departmental use. It is straightforward and clear.

The Supreme Court should say, "Hey, we screwed up in Katz. Privacy is privacy. Wherever you are you do expect a reasonable level of privacy. If you are engaged in ongoing criminal activity, the police have a right to obtain a warrant, backed by reasonable cause, to monitor almost everything you do, but they do not get to do anything other than casual surveillance on the street without that warrant. Your car is one of your effects that is secure from unwarranted searches and seizures. Dogs cannot be used without a warrant, particularly since the evidence is clear that police engage in discriminatory behavior in the way they use dogs and the dogs are not particularly reliable. Oh, finally, you cannot seize any goods from anyone who has been accused until they are convicted and you have shown in the trial that those goods are a direct result of the criminal activity, and even then, if you seize the goods, the law enforcement department cannot get any benefit from that seizure."

If the police have a problem getting warrants quickly, then the governmental units responsible have to make certain there are judges available at all times to review the probable cause and issue warrants, but time, no matter how much it is of the essence, cannot be used as an excuse to throw the Fourth Amendment away.

[hairshirthedonist]

I'm thinking David Berkowitz should go free, given our deference to dogs.

[Phil]

I'd love to see one of these dog cases go back to the Supreme Court now, given their recent Sixth Amendment decisions. You can't question a dog on the stand, after all.

[Doctor Science]

The study is here, though of course behind a Springer paywall. #boohiss

The principle investigator, who has been a detection-dog handler, says:

It is important to recognize that these findings do not mitigate the abilities of these handler/dog teams to perform successfully.

Oh yes they do.

[CharlesWT]

"It turns out that drug sniffing dogs have an incredibly high false positive rate--they alert many more times than drugs are actually found once a search is done."

From law enforcement point of view, this is a feature, not a defect.

[byomtov]

I don't know a lot about dogs, but one thing I do know is that they are absolutely incredibly good at picking up cues from humans. That they react to their handlers' suspicions shouldn't surprise anyone.

BTW, isn't this just one more example of the legal system doing a poor job of dealing with probabilities and statistics, and science in general? I suppose that's part of Sebastian's broad point.

[John Herbison]

I wonder whether dogs who do not alert prolifically are actually used in the field. My suspicion is that such dogs are winnowed out during the training process.

[y81]

In answer to the original post, the accuracy of drug-sniffing dogs could be challenged in an appropriate case, but it wouldn't be a constitutional issue, so the Supreme Court probably wouldn't be the court that decided the issue.

In answer to Free Lunch, the drafters of the Constitution might have written, "no searches without a warrant," but that isn't what they did. They wrote "no unreasonable searches" and "no warrants without probable cause." Those are the applicable standards.

[Countme-In]

I expect Rick Perry will come out tomorrow in favor of executing anyone caught in possession of sausage.

Ron Paul will express his opinion that sausages are not prohibited by the Constitution, but newsletters he wrote, read, and published years ago will reveal sausage-hatred of a fascist kind.

Newt will begin his statement regarding sausages as follows: It is very simple ....... sausages are fundamentally anti-American ... etc ,, though Callista will roll her eyes in the direction of sausage.

Michelle Bachmann will drone about the caliber of the sausage.

Mitt Romney promoted sausage in Massachusetts, but now calls it socialist salami.

Rick Santorum has no opinion on sausage and for good reason, although he will note that sausages bring to mind ... well, just about everything.

This is a good post, Sebastian, but what I notice is that despite the prevalence of dogs sniffing out unwrapped sausages, despite their cop handlers having the sausage sandwich for lunch that day, our prisons are full of blacks, not Italians.

What's with that?

[Countme-In]

I suspect if sausage-sniffing dogs were permitted before the Supreme Court, Clarence would recuse himself, but not before noting that sausage is not mentioned in the Constitution.

[Laura Koerbeer]

There's an orca poop sniffing dog who helps scientists monitor the health and eating habits of the local orcas. Do you think he gives false positives?

I'm puzzled by this false positive phenomenon. I'm not puzzled that the dogs pick up cues from humans. I'm puzzled that the dog is confused about what to respond to. An effective trainer would not reward a dog for false positives. An effective trainer would not be, in effect, training the dog to respond to human cues. The goal is supposed to be to train the dog to respond tothe smell of (whatever).

It's like all these trainers are training dogs to the standard of Clever Hans.

Is this pattern of false positves just a drug sniffing thing? I wonder if bomb siffing dogs give false positives at the same high rates, or the dogs that are supposed to sense the onset of epilepsy.

[Snarki, child of Loki]

The system won't change until high-level federal judges and conservative politicians are subjected to abusive searches, seizures and arrests as a result of spurious dog "alerts".

Preferably on a weekly basis, until they see the light. After all, if they have done nothing wrong, they have nothing to fear.

[Laura Koerbeer]

This is really bugging me. I used to spend a lot of time with a woman who had a highly trained service dog. Her dog knew ninety verbal commands. Many of the commands involved finding and bringing an object or acting upon an object that was out of sight. "Bring Phone", for example, whenthe phone was in a different room. Sometimes there were cues other thanthe verbal one to help the dog understand but sometimes not. The phone, for eample, was often misplaced so the owner could not give cues to its whereabouts. The dog would look and look until she found it.Sometime she'd need reminders to keep looking, but she didn't bring random objects just because she thought the owner was cuing her toward them. She knew she was supposed to find the phone and she knew which object the phone was.

Surely the smell of pot or coke or heroin is distinctive enough that a well trained dog would be able to distinguish it from regular household or yard or car smells. Heck I can tell pot when I smell it!

I think high positves indicate either lousy training, sloppy police work ( the cops not usingthe dogs properly) or the maybe the ability of dogs to accurately sniff out drugs is just overhyped and overrated.

[liberal japonicus]

Actually, the fulltext is available here

This blog post and this blog post both have a better take on the original article than the Economist article, which I think seems to have a notion of 'ha! sniffer dogs don't work!' (though that dump all the caveats in the last paragraph the way the Economist does is noted). From the last one

These were experienced teams, who have clearly proven their abilities to find hidden contraband. So what was happening? It appears that the handlers, thinking there were scents present where there weren't, cued the dogs. At least, that is what has been reported. I read about the paper in Scientific American, whose headline read, in part, "When dog handlers believed there were drugs or bomb materials, their dogs called more false alarms".

Is that true? It's a subtle point, but the headline isn't quite right. The team called more false alarms. That is, the handlers reported that the dog found the drug. Two things could have happened, and the study's data does not distinguish between them. Either the handlers tipped off the dogs to what they thought were correct locations -- the Clever Hans "effect" -- or the handlers simply misreported what the dogs were doing. They might have "thought" or "sensed" that they saw the dog alert to a scent. This is classic confirmation bias: you see what you expect to see. An experienced handler expects to see his dog find the scent. (emphasis original)

[jrudkis]

I think a lot of this comes down to "Dog Logic." Dog Logic is something I read about when I got my last dog, where the point is you do not necessarily teach your dog what you think you are teaching.

A man had a dog that was not house broken. He would scold him for accidents when he observed them and would walk the dog out side whenever it seemed the dog had to go, but the dog would not poop on the walks. When they went inside, however, the dog would run behind the couch and poop. In the article, the professional dog trainer said the guy was trying to teach his dog to poop out side, but in fact taught his dog not to poop where his owner can see him.

My experience with something similar had to do with a truck load of explosives in Iraq, that managed to get past several bomb sniffing dogs. All the dogs apparently passed testing periodically where the dogs could identify minute amounts of explosives. Somehow, none of them found the massive truck borne explosives.

After investigating the failure, the best explanation we got was that the dogs were trained to find trace amounts of explosives, and did not think that huge piles were the same thing. So the dogs were doing what they thought they were trained for (trace amounts), not what we thought they were trained for (explosives).

Or possibly a lot of those dogs were just untrained and put there for profit by contractors. Hard to know.

[hairshirthedonist]

The team called more false alarms. That is, the handlers reported that the dog found the drug.

Putting aside controlled experiments for the moment, in the real world, doesn't it come down to a matter of trusting individual police officers not to assert that their dogs found something any time those officers felt like searching someone's car or house? Who's to say? And isn't the whole point of the 4th amendment not to place that level of trust into the hands of the people who carry out the searches? Is a dog any sort of check at all, really, on the whims of the police?

(I can't imagine internal affairs has dogs trained to interview drug-sniffing dogs about the circumstances of searches initiated based solely on officer-claimed detections by their dogs - and I'm nuts - the world looks like a Simpson's episode to me.)

[Turbulence]

It is strange that there has been no 'evidence-based justice' revolution in the criminal justice system the way there has been in medicine. As Seb points out, the whole notion of cops laundering their random guesses through dogs is absurd, but it fits in nicely with other criminal justice techniques that have been repeatedly proven to perform poorly*.

Then again, for all the noise and discussion generated by evidence-based medicine, it doesn't seem to have changed medical practice nearly enough.

*Say, eyewitness identifications, polygraphs, the fire forensics that killed Cameron Todd Williams, bite forensics run by that crazy "Doctor" in Mississipi, etc.

[sapient]

It is strange that there has been no 'evidence-based justice' revolution in the criminal justice system the way there has been in medicine.

That's because it's hard to get anyone to care about stuff unless it's happening to them. Medical things happen to most people randomly. Drug investigations happen much more narrowly to certain communities, mostly those with little political power.

Decriminalizing drugs would do a lot to improve the integrity of the criminal justice system. Anti-vice enforcement has never worked well.

[Turbulence]

That's because it's hard to get anyone to care about stuff unless it's happening to them.

No, I don't think that's it at all. Evidence based medicine is a phenomena largely driven by medical practitioners themselves. And it seems to me that judges and prosecutors and detectives have a big interest in whether or not their methods are bullshit, just like doctors do. It is true, they get paid either way, but that's true for doctors too.

Decriminalizing drugs would do a lot to improve the integrity of the criminal justice system.

It wouldn't have saved Cameron Todd Williams, right?

Decriminalization is not a panacea for all criminal justice system problems.

[sapient]

Evidence based medicine is a phenomena largely driven by medical practitioners themselves. I'm not sure that's entirely correct but, even so, the medical establishment has traditionally fought a lot of other initiatives that might have been helpful to patients when there are broader social implications (such as medicare or other "socialization" initiatives). There are probably a number of defense attorneys who attempt to argue against the efficacy of drug-sniffing dogs, but the more powerful popular political anti-crime forces successfully argue for less delicate ways to interfere with certain suspect groups.

Decriminalization is not a panacea for all criminal justice system problems. Did I say or imply that it was? No, I did not. Medical malpractice will never go away either, even though "evidence based medicine" is widely accepted. My point was that vice squads have historically been particularly egregious about overstepping their mandate, and that a lot of 4th Amendment violations involve drug searches. Not to mention the fact that our prisons are overcrowded with nonviolent felons who have violated drug laws.

[byomtov]

It is strange that there has been no 'evidence-based justice' revolution in the criminal justice system the way there has been in medicine. As Seb points out, the whole notion of cops laundering their random guesses through dogs is absurd, but it fits in nicely with other criminal justice techniques that have been repeatedly proven to perform poorly*.

I agree, but it seems the legal system has a very hard time with the sort of arguments and issues Seb discusses here. Dare I suggest that lawyers don't understand numbers very well?

In my experience this extends not just to matters of probability and statistics, but to dealing with accounting and financial information also.

A friend who taught law relates that he was in class one day and asked a student to look at a certain section of the textbook. When the student couldn't find the material my friend told him to:

"Look at the bottom of page xx, where the Arabic numerals are."

Came the response: "Arabic? I don't know any Arabic."

[wj]

Decriminalization is not a panacea for all criminal justice system problems.

But Turb, it doesn't have to be a panacea to be something that would make a significant improvement. Which is all that sapient was asserting.

[avedis]

This is a great post, Sebastian. Yes the courts and the police state have been making a bludgeoned mess of the Constitution as of late. I think the arguments are badly framed and the questions wrong intentionally; not as a result of an artifact of the legal mechanisms.

Anecdotally, I was pulled over at the Canadian border by US officials a couple years ago. I was on a return trip from spending time with a seriously ill close to death family member.

The reason I was stopped was that two drug/money/explosive sniffing dogs were being walked through through the hundred meters or so of backed up traffic and one of them stopped in front of my car and barked. What followed was like a scene out of a bad cop movie.

In a matter of seconds I had two agents with pistols drawn and two more shouting commands at me. One reached in the open window and grabbed the keys out of the ignition. The next moment I was spread across the hood of my car, frisked, and my wallet and ID were taken by an agent who quickly disappeared with them.

I hadn't said a word.

I was then quickly whisked into a building, through several rooms and finally into an interogation room (totally locked down, of course).

There were two agents, one on either side of me shouting questions like I was a recruit and they were DIs.

The gist of their beef was that their dogs said I had drugs/money/explosives stashed in my car. The sooner I told them the where in my car these things would be found, the better (why better I don't know).

At this point I spoke for the first time; politely telling these over enthusiastic protectors of freedom that their dog was mistaken. There was nothing illegal in my car.

That really pissed them off. I guess it shouldn't be surprising that such misanthrops would value their canine relationships more than human ones with their fellow citizens.

Any how, they proceeded to take my car apart. As the search continued to yield no contraband and the batteries on their power screw drivers wore down, they'd come in and yell at me for a couple minutes, then go back outside to find some new piece of car to remove and disassemble.

Trying out a kindler genter tact, I suggested that perhaps, just perhaps, the dogs were not totally mistaken, but that they had confused the scent of burnt gun powder (about a week previously I had sighted in a rifle and was resting on the hood of the car when I did that). Nope! Impossible! The dogs know the difference.

Attempting humor I suggested that maybe the dogs had earlier made a real bust and had gotten wasted off powder residue and weren't on their A game for the rest of the day. Red faces and popping veins were the response.

Of course, nothing was found because, I am not actually totally stupid and I don't take so much as a pot seed across the border because I know that there are drug dogs on the border. Also, I do smoke once in a while, but only in places other than my car and I don't believe there has ever been any amount of drugs in my car for even a minute. I don't have huge stashes of cold cash and I don't have access to explosives and, if i did, I wouldn't take them to Canada because, for one thing, I like Canada and Canadians.....

......any how. Four hours later they came back to the lock down room for the last time and literally threw my keys at me and told me to "get out of here".

Me, "what? I can go? So does that mean that Fido was wrong?"

My tax funded civil servant, "The dogs are never wrong. You had something in your car earlier and you got rid of it before you got here".

Which left me wondering if i was going to be under surveillance. Probably was, but I never heard any more about this.

[russell]

Decriminalization is not a panacea for all criminal justice system problems.

No, but it's a very useful remedy for problems caused by criminalizing stuff that isn't particularly criminal on its face.

Humans like to get high. Always been so, probably always will be so. It appears to be something our nervous systems have an affinity for.

The war on drugs crapola is an expensive waste of time and money, and is one of the most significant pretexts for undermining personal civil rights in the US.

Thanks for a good post Seb! And for making time to jump in.

[avedis]

"No, but it's a very useful remedy for problems caused by criminalizing stuff that isn't particularly criminal on its face."

Yes.

On top of that it is a good way to avoid creating perverse incentives that further erode civil liberties. I am talking about the exponential growth of bureaucracies like the DEA, et al, but more importantly, the privately owned penal facilities. These facilities are a multibillion $ industry and they tend to be located in economically depressed areas; which means that the owners have a lot of sway over law makers by using jobs creation (and lobbying $s) in those districts as leverage to push for ever more draconian drug laws and easier access to our homes, etc by police, what with drug offenders representing something like a third of those in prison.

[Laura Koerbeer]

DId they put your car back together?

Many many years ago a friend and I were returning from a cross continental bicycle trip. We were crossed into the US in Detriot.

There were no dogs. The US Immigration pricks were assholes all on their own. They tore everything apart. They even openned a box of maple sugar candy. They shouted, threatened. The men did anyway. There was a female agent who never raised her voice and seemed kind of embarrassed. They separated my friend and I and screamed at us. I really thought I might get raped or something.

Then all of a sudden it was over. They told us to get our stuff and leave. We had to repack everthing. Our belongings were thrown all over the room.

[avedis]

Laura, they did put it back together, albeit badly. I had to go to a shop and have a few parts re-adjusted. And they did leave my personal belogings strewn about (they had been neatly packed).

What is frightening to me is that these same pricks can sweep you away to the GITMO or worse without charges being leveled under the anti-terrorism whatever BS laws.

Americans are a bunch of spoiled frightened pukes that don't deserve to be free. Ooohh......eeeeeh....there's a one in ten billion chance a terrorist might kill me......oh my god.....drugs could possibly make my children go crazy like reefer madness or Charles Manson's zombies even though that chance is statistically miniscule...SHRED the Constitution! WHATEVER IT TAKES! make the boogie men go away.....They'd rather be "safe" from phony threats than free. Except they're not even safe from their own government any more (whether that govt be local or fed).

BTW. would it be ok if i called the dog a "bitch"?

[Laura Koerbeer]

Yes! After all I referred to the male officers as "pricks"!

I do think that an excess of selfindulgence or perhaps the inability to distinguish real problems from ginned up ones...Republicans are always yapping about the decline of moral values, not that a Republican politician would be able to recogize a moral value if it bit him or her in the ass. I don't know about decline, but I do think that we Americans are sufferig from a significant absence of a sense of perspective and along with an absence of the ability to make reasonable moral judgements which I precieve as decadence.
Jingoism instead of patriotism
Social Darwinism instead of the social contract
Pro-life means sanctify fertilized eggs but cut Medicare, Medicaid, ignore globalwarming, kill a million Iraqis
A corporate media that thinks "unbiased" means to report lies as facts
Half the country living in poverty but we can't afford big government programs unless the money goes to red states or military contractors! And it would be unfair to tax the rich!
And so on.

Has this nation always been this morally and intellectually bankrupt?
So many peoiple have their values basackwards.

[Countme-In]

"By plane, car, horse,camel, elephant, tractor, bicycle and steam roller, on foot, skis, sled, crutch and pogo-stick the tourists storm the frontiers, demanding with inflexible authority asylum from the 'unspeakable conditions obtaining in Freeland," the Chamber of Commerce striving in vain to stem the debacle: 'Please to be restful. It is only a few crazies who have from the crazy place outbroken.'"

William S. Burroughs "Naked Lunch"

Now the few are legion and have hollowed out the Republican Party, and it lurches like a husk through the land, yammering to the poisonous laugh track of sadistic delighteds who long for genocide.

Happy New Year.

[Countme-In]

"So pack your ermines, Mary, we are getting out of here right now."

[CharlesWT]

"...They'd rather be "safe" from phony threats than free. Except they're not even safe from their own government any more (whether that govt be local or fed)."

If humanity cannot live with the dangers and responsibilities inherent in freedom, it will probably turn to authoritarianism. This is the central idea of Escape from Freedom, a landmark work by one of the most distinguished thinkers of our time, and a book that is as timely now as when first published in 1941. [...]

Escape from Freedom

[Hartmut]

Welcome to border control late GDR style ;-)
That was the way people travelling from or to West Berlin by car were treated. US border control people are (I assume) still amateurs doing selected individual cars instead of going full assembly line like those old commies.

[Gordon Danning]

The "dog sniff is not a search" rule actually goes back to United States v. Place, 462 U.S. 696 (1983), in which a dog was used to sniff luggage at an airport.

The Court's rationale was as follows: "A 'canine sniff' by a well-trained narcotics detection dog, however, does not require opening the luggage. It does not expose noncontraband items that otherwise would remain hidden from public view, as does, for example, an officer's rummaging through the contents of the luggage. Thus, the manner in which information is obtained through this investigative technique is much less intrusive than a typical search. Moreover, the sniff discloses only the presence or absence of narcotics, a contraband item. Thus, despite the fact that the sniff tells the authorities something about the contents of the luggage, the information obtained is limited. This limited disclosure also ensures that the owner of the property is not subjected to the embarrassment and inconvenience entailed in less discriminate and more intrusive investigative methods."

The case re: the dog sniffing a car was Illinois v. Caballes, 543 US 405 (2005); interestingly, the defendant "argue[d] that the error rates, particularly the existence of false positives, call into question the premise that drug-detection dogs alert only to contraband," but the Court rejected the argument because the defendant had presented no evidence re: the frequency of false positives. So perhaps the studies cite here could be employed to reverse the dog sniff rule.

However, even a fairly high false positive rate is not necessarily fatal to the use of dogs, because the dogs need only be reliable enough to provide probable cause for a search. That could easily be as less than 50% reliability, because probable cause "does not demand any showing that such a belief be correct, or more likely true than false." Texas v. Brown, 460 U.S. 730 (1983)

[dangermouse]

However, even a fairly high false positive rate is not necessarily fatal to the use of dogs, because the dogs need only be reliable enough to provide probable cause for a search. That could easily be as less than 50% reliability, because probable cause "does not demand any showing that such a belief be correct, or more likely true than false." Texas v. Brown, 460 U.S. 730 (1983)

So "probable" doesn't exclude "literally less successful than a coinflip"?

Way to go, legal system, you've done it again.

[dbt]

less than 50% success rate is not less probable than a coin flip.

If the dog examines 1000 locations, identifies 15 of them, and only 5 of them contain contraband, then only 33% of identifications were correct, but they were correct on 990 locations, a 99% success rate.

(I am generally opposed to using techniques like dogs that "get around" constitutionally mandated 4th amendment rights, just pointing out that your math is wrong.)

[Gordon Danning]

@dangermouse:

Nope, "probable" still means more likely than not. "Probable cause" doesn't. It's a term of art, much like "due process," which obviously does not mean the process that is payable on demand.

[Hartmut]

dbt, ...provided that in the 985 other locations there were indeed no drugs despite real suspicion, i.e. false negatives are excluded. Otherwise it would be possible to boost the 'success' rate by simply 'checking' locations where it is known that no drugs are.

[avedis]

I am admittedly not knowledgeable concerning legal theory, etc, but it seems to my (very) layman mind that there is a disconnect between the results of lie detector tests not being admissable evidence, or even enough to establish probable cause, due to unreliability, but dogs being sufficient for probable cause despite unreliability. What is going on here - legally speaking?

[envy]

I must agree with avedis that polygraphy was just about the first thing that came to mind when I read this. My rather cynical instincts say that polygraph tests are an unreliable, questionable heuristic that can and have been leveled at Our Kind of People, whereas drug-sniffing dogs are a rock-solid tool of forensic science that has time and again proven themselves reliable in our unifying national crusade against the Scourge of Drugs and Those People that use them.

Which doesn't even try to answer what's going on, legally speaking. Sadly, however, I'm not sure it really needs to, under the circumstances.

[avedis]

Envy, I too have some very cynical thoughts around this. One is the same as yours. Another is that there is a subtle unconscious - yet real - strong affiliation/identification between authoritarian personalities and dogs; particularly of the German Shepard variety. A lot of cops actually see the dog as another cop, a buddy a partner in LE.

The machine, the polygraph, they don't like because it is not a cop. It is an unbiased technology and it can turn on a cop.

Still doesn't explain the court's persdpective; unless that attitude filters up - which it very well might.

[Kratoklastes]

@byomtov - re "A friend who taught law relates that he was in class one day and asked a student to look at a certain section of the textbook. When the student couldn't find the material my friend told him to:

"Look at the bottom of page xx, where the Arabic numerals are."

Came the response: "Arabic? I don't know any Arabic."

Well, your friend is only slightly less-dumb than the student, for verily I say unto you that xx is a ROMAN numeral, not an Arabic one.

I take it this was an American 'school', yeah? That would not surprise me in the least.

As to the gist of the article: it is silly that people act surprised when 'law' enforcement attempts to do an end-around on a section of the Bill of Rights, and then goes and gets a rubber stamp from some robed charlatan whose entire career has been underwritten by the government.

Codified Bills of Rights, when overseen by the political-parasite class, morph from "non-exhaustive sets of constraints on government" to "the last set of things you have left". And THEN, the parsing starts by the aforementioned robed charlatans.

4th amendment says "persons, houses, papers, and effects"? Well, a car is not an "effect" because some tax-feeder says so. And besides (continues the 'judge'), if a badged goon says he smells pot, the 4th can go fuck itself.

8th amendment prohibits cruel and unusual punishment? Well (sayeth the robed charlatan), locking someone up for 14 years for civil contempt doesn't violate the 8th because it wasn't PUNISHMENT, it was COERCION.

See how words work?

John Yoo - another charlatan in tax-funded sinecure - claimed that anything that didn't cause death or organ failure was not torture (and therefore not a violation on Geneva and common law prohibitions on same). To which I responded: give me John Yoo and a range of power tools and I will "non-torture" him until he changes his mind.