Police Use Junk Science To Secure Convictions.
The NYT is doing it’s job and trying to expose the flaws in the technology of alcohol breath test machines.
A million Americans a year are arrested for drunken driving, and most stops begin the same way: flashing blue lights in the rearview mirror, then a battery of tests that might include standing on one foot or reciting the alphabet.
What matters most, though, happens next. By the side of the road or at the police station, the drivers blow into a miniature science lab that estimates the concentration of alcohol in their blood. If the level is 0.08 or higher, they are all but certain to be convicted of a crime.
But those tests — a bedrock of the criminal justice system — are often unreliable, a New York Times investigation found. The devices, found in virtually every police station in America, generate skewed results with alarming frequency, even though they are marketed as precise to the third decimal place.
Judges in Massachusetts and New Jersey have thrown out more than 30,000 breath tests in the past 12 months alone, largely because of human errors and lax governmental oversight. Across the country, thousands of other tests also have been invalidated in recent years.The machines are sensitive scientific instruments, and in many cases they haven’t been properly calibrated, yielding results that were at times 40 percent too high.
A breath test claims to be able to determine how much alcohol is in your blood just from your exhalations. It sounds legit, right? Alcohol is in the blood, the blood exchanges gases in the lungs, and some of the alcohol in your blood should pass out of the blood and into the air in your lungs as gases exchange. But even so, a couple of breaths isn’t a lot of data to work from. When medical researchers are studying how much oxygen and carbon dioxide and other metabolites are in the air a person exhales, they use a full face mask and have a person breathing through it for a while. They collect a lot of data.
But the police? They use a briefcase sized machine, or a handheld one, to run the test, and in a lot of places, that is the extent of the physical evidence they gather to convict you with. Remember Theranos, that claimed to be able to do blood analysis with a pin prick of blood?
Some places are smart, and they use the machine to basically establish probable cause for a blood test , and then get a blood draw; but a lot of places never bother with the blood test, and the machines? The machines are, ideally, sensitive scientific instruments. Machines that are often left out in the open, where who knows what kinds of abuse they are subjected to. Machines whose guts are protected by secrecy and tight IP, so no one can challenge the veracity of the machines (a classic black box). And we are trusting the police to properly maintain and calibrate these machines.
I’ve said before that we should do away with drunk driving laws. Not because I think people should be allowed to drive impaired1, nor do I think that we should not have additional penalties for drivers who operate impaired, but that the police should focus on the behavior of a driver, and not the results of a black box test that encourages them to.
Relatedly, a family that was the result of similar lazy policing gets a court victory. Again, a case of police using a black box (in this case, a field chemical test) to be the sole driver of an investigation, rather than the first step. And this case is a great example of why we shouldn’t trust the use of police breath tests as the sole bit of evidence for anything other than a warrant. Because the police can not be bothered to understand the tool they are using.
Balko had a bunch of posts about this back when he was automatically assumed to be Team Evil by Good People and it was from him that I learned that you can make a breathalyzer’s numbers go up by shaking it.
So if the cop doesn’t like you and you’re borderline (only had one beer for a 175 pound man, say), he can shake the breathalyzer behind his back for a few seconds before looking at it.Report
Balko was the first to turn me onto the fact that such technology was unreliable, and worse, prone to being gamed. But yep, being team Evil-Lib, no one really listened to him. Hopefully now that the NYT has said something, people will listen.
Pollyanna-ish of me, I know.
PS Speaking of the NYT publishing unexpected good.Report
Im skeptical it will do anything. The facts have been out since the 2009 National Academy of Sciences report on the lack of scientific rigor underlying police forensics. I’m not aware of any significant changes in rules of evidence even a decade later.
Lawyers and judges are bad at science and the system is overwhelmed. There’s also a real lack of political will. Conservatives fetishize law enforcement too much. Progressives are so up their own asses on demographic disparities they forget to actually do anything substantive, not to mention have their own carceral instincts.Report
No liberal ever listens to Balko? That is a newsflash. I guess i’ll have to wipe my memory and start tracking down all those libs who have talked him up for years.
Slightly less snarkily. If people want less “Team X vs Team Y” bull squat then we need to stop acting like that describes the entire world.Report
Back when Balko was writing for Reason, I’d link to his stuff on FB, and a lot of my friends would give me the “ewwww, he writes for Reason!” response.
As much as I enjoy Reason, I’m kinda glad he got picked up by WaPo, it eliminated a lot of the guilt by association.Report
That’s what you get for linking political stuff on FB.
OK, that’s one of the many bad things that you get for linking political stuff on FB.Report
Oh i’m sure that has happened. Reason often is blinded by their ideology and can suck hard. But as pillsy said, that is FB and people do stupidly ignore stuff. However i think i first read balko many years ago from a link from mother jones or the nation. I’ve seen many liberals quote out his stuff.Report
Wouldn’t it be great to have the Swiss health care system, or something like it, where you had to buy private insurance and if you couldn’t afford it you’d be subsidized. Why does that sound familiar?Report
And an FDA that was more like the European Medicines Agency.Report
And all beer was brewed to the Reinheitsgebot. Wait… no not that one.Report
I’d be fine with that one. It’s not like we wouldn’t quickly come up with a new term for the grain sodas that are so popular here.Report
i think a lot of the problem is that cops, like most humans – can’t handle uncertainty very well. So give them an instrument, and they get to duck the intellectual problems of someone who blows above 0.08 but actually is in full control and poses little threat. Its a common problem in translating actual science into mass use and effective policy.Report
The problem with that theory is that police don’t write laws. Legislatures do, and they do that in response to the public. The police use things like the breathalyzer machines in response to the public cry to get DUI’s off our streets. The efficacy of them has little to no meaning on the use of them. They are used simply to enforce social control.
If the public agitates for a legal limit that has a number attached, then there has to be a method to determine that number. Actual drunkenness or ability to drive isn’t the point of the law, that number is. So, enforcement is geared to finding that number. Nothing more, and sadly, nothing less.Report