Breathalyzers — and why they don’t work

Posted by Lawrence Taylor on October 17th, 2004

So what did I mean in my earlier post by "unreliable breath machines" and "passing laws against science"? Here’s just one of many examples… The computers inside Breathalyzers actually multiply the amount of alcohol in a DUI suspect’s breath sample 2100 times to get the blood alcohol concentration ("BAC"). This is because it is programmed to assume that the suspect has 2100 units of alcohol in his blood for every unit of alcohol in his breath. This is called the "partition ratio". But this ratio is only an average: actual ratios vary from as low as 900:1 to as high as 3500:1; if individual ratio is different, the BAC result will be different.

Translation: If a suspect has a true BAC of .06% ("not guilty") and a partition ratio of 1300:1, for example, the machine will give a result of .10% ("guilty"). Convicted by a machine. His crime? He was not average.

Well, when juries hear this kind of evidence, they tend to return "not guilty" verdicts. This did not sit well with MADD’s and prosecutors’ lobbyists. The result: in California and other states, drunk driving laws were changed by adding "Percent, by weight, of alcohol in a person’s blood is based upon grams of alcohol per 210 liters of breath". In other words, the law no longer cared what the actual amount of alcohol was in your blood: it was legally (not scientifically) presumed to be 2100 times what is in your breath — even though we know it is not.

The conservative California Supreme Court later found the new DUI law constitutional in People v. Bransford. But a dissenting Justice pointed out the obvious: "The majority…has on its own created the new crime of driving with alcohol in one’s breath". (Justice Joyce Kennard, 8 Cal.4th 894)

Result: today, defense attorneys are prohibited from mentioning anything about partition ratios to a jury. Scientific truth has been banned from the courtroom. And the conviction rate in drunk driving cases has risen dramatically.