Minnesota attorney: We can't trust breathalyzers

Getty Images

Getty Images

When Minneapolis attorney Barry S. Edwards goes to parties, people often ask him how he can possibly defend drunk drivers.

Edwards also defends people charged with drug possession, domestic assault, and probation violation. Somehow, this is the one people usually have a problem with.

To a certain extent, he gets it. Drunk driving is a serious problem in Minnesota. According to the Department of Public Safety, 381 people died on Minnesota roads last year, and 84 of those fatalities were “known to be drunk-driving related.”

Organizations like Mothers Against Drunk Driving have done a great job making it way less socially acceptable to drive drunk than in decades past. Edwards thinks some of that fervor to prevent tragic accidents has led to a real snag in the criminal justice system. He says it all comes down to breathalyzers.

We all generally understand how breathalyzers work, even if we’ve never had to use one. You blow into one end, a computer reads your breath tells you your Blood Alcohol Concentration (BAC). If you’re over 0.08 percent, you’re too drunk to legally drive in most of the country. Simple as that.

Assuming the breathalyzer works: A recent New York Times investigation found they often generate “skewed result," citing “thousands” of test results thrown out nationwide because of human error, “lax government oversight,” and programming mistakes.

Minnesota used to use a model called the Intoxilyzer 5000EN—nearly all these machines have names that sound like cartoon robots—which used infrared technology to measure the level of alcohol particulate in the user’s breath. In 2011, an “error” in the Intoxilyzer’s source code was discovered, and in 2012 defense attorneys like Edwards sued to challenge their clients’ drunk driving arrests.

The Minnesota Supreme Court rejected those challenges in a 4-3 decision, despite nearly “a half-dozen... experts” who said the machine wasn’t accurate enough, and could even be skewed by the presence of a nearby cellphone, according to MPR. The ruling preserved breathalyzer's untouchable status in Minnesota courts. 

Around that time, the state’s Bureau of Criminal Apprehension (BCA) switched over to a new model, the DataMaster DMT-G—see what we mean?—which was supposed to be far more accurate. Not only did this one take infrared readings, it used a “fuel-cell” reading to double-check.


As early as 2012, the Star Tribune reports, the reliability of the fuel-cell was called into question, and the bureau told law enforcement agents to simply shut it off until “inconsistencies” in its shelf life were corrected. That meant it was back to just infrared readings. (The New York Times gave Minnesota and the DataMaster a shout-out in its own piece.)

In a statement to City Pages, the BCA said it “remains confident in the reliability of results obtained by DataMaster devices across Minnesota" and "welcome[s] the opportunity to address any challenges in court."

Edwards isn't as optimistic. With a lot of other forms of scientific evidence—urine tests, drug dog sniffs, blood samples—it is possible to challenge their efficacy and relevance to the case before the trial even begins.

“We don’t want jurors to be confused by magic tricks,” Edwards says. “We want someone who is educated in science… to be a gatekeeper.”

Minnesota law states that in any trial, civil or criminal, the results of a breath test will always be admissible in evidence without “antecedent expert testimony” guaranteeing it’s reliable—provided the breathalyzer was approved by the state commissioner of public safety, and the test was administered by someone who knows how to use it.

Lawyers also ask Edwards how he defends drunk drivers. But they mean it literally. They wonder how he can possibly argue with a number on a screen. His answer is he can, and he sometimes does, depending on how amenable judges are to hearing his doubts. Just last year, a Minnesota judge ruled that the DataMaster seemed to be rounding up results and pushing defendants artificially past the legal limit.

Edwards thinks the Times investigation "will help bring some awareness" to problems with breathalyzers, but says the general lack of scrutiny makes his work “frustrating.” He believes there are ways to curb drunk driving without relying on the alleged three-points-past-the-decimal accuracy of these machines.

For example, instead of a legal limit there could be a zero tolerance policy on driving after drinking. Or we could rely more on blood and urine testing, which are much more accurate, but require search warrants.

For his part, Edwards isn't holding his breath.