As a citizen of the D.C. metropolitan area, I am in favor of laws that protect us all from drunk drivers. But the laws need to reasonable. They need to punish the behavior we seek to deter.
D.C.’s per se law on driving while intoxicated (DWI) is not only a mess; it is also of questionable constitutionality. Unlike Virginia and many other jurisdictions in which breath tests are used as a surrogate measure for blood alcohol concentration, D.C. punishes breath alcohol directly. In other words, according to my reading of the statute, you can have alcohol on your breath and not a drop of alcohol in your blood and still be found criminally liable for DWI.
Sound crazy? Take a look at the statute. According to Section 2201.05(b)(1)(A)(i)(I) of Title 50 of the D.C. Code, no person “shall operate or be in physical control of any vehicle in the District . . . [w]hen the person’s alcohol concentration at the time of testing is 0.08 grams or more either per 100 milliliters of blood or per 210 liters of breath or is 0.10 grams or more per 100 milliliters of urine.”
Note that the statute says “alcohol concentration,” not “blood alcohol concentration.” Note that “alcohol concentration” is to be measured either by a blood sample (which would directly measure blood alcohol concentration), by urine sample (which would be a slightly more indirect measure) or by breath sample (which is even more indirect).
Now compare this with Virginia’s DWI statute. According to Section 18.2-266 of the Virginia Code, it is unlawful for a person to “drive or operate any motor vehicle . . . while such person has a blood alcohol concentration of 0.08 percent or more by weight by volume or 0.08 grams or more per 210 liters of breath as indicated by a chemical test as administered as provided in this article.” (Emphasis added.) Note that, unlike the D.C. statute, the Virginia law does tie all measurement to blood alcohol concentration. Breath alcohol concentration is used as a surrogate for measuring blood alcohol concentration; it is not an evil in itself.
Thinking that the D.C. statute must be a recent innovation designed to facilitate drunk driving prosecutions, I pulled up the old version of the law to find that it was even worse. It does begin with a focus on blood alcohol concentration. But it adds a prohibition on breath alcohol in a virtually incomprehensible parenthetical on “alveolar air,” which, according to current technologies, would be impossible to measure accurately. It is therefore no surprise someone decided this language needed to be changed.
So how did we end up where we are today?
Ideally, all DWI per se laws – the laws that proscribe driving a car with a blood alcohol concentration above a certain level without regard to impaired driving – would focus on the evil we are seeking to prohibit: blood alcohol concentration. It is, after all, alcohol in the bloodstream – not on the breath, not in the urine – that causes intoxication. Unfortunately, blood tests are messy, invasive, and have sometimes led to lawsuits. It can also take a while to get back the results from blood tests.
Developers of the Breathalyzer, Intoxilyzer, Intoximeter and other breath test machines sold these technologies as a cheaper and more efficient way for measuring blood alcohol concentrations. But, as with any surrogate measure, the technologies have relied on certain assumptions, such as the blood-breath partition ratio I have discussed earlier.
Criminal defense attorneys have successfully challenged many of the underlying assumptions at trial. These challenges make prosecutions more difficult and costly for the government. They sometimes result in acquittals. In response, instead of either improving the technology or forcing the measurements back to blood alcohol, many jurisdictions have simply written the assumptions made by the breath test technologies into the laws themselves.
This, for example, is what Virginia has done. The “per 210 liters of breath as indicated by a chemical test” language essentially adopts the assumption made by the breath test developers that there is a correlation of 1 to 2100 between the amount of alcohol on the person’s breath and the level of alcohol in the person’s blood. It no longer matters that the actual ratio of breath to alcohol in any given individual can vary from 1:1300 to 1:3000 and that, by assuming the 1:2100 ratio, the machine could substantially overestimate actual blood alcohol concentration.
The D.C. statute seems to have taken this a step further. It too codifies the 1:2100 breath to alcohol ratio. But, by dropping any connection of the breath test to blood alcohol concentration, it also criminalizes the very act of having alcohol on your breath at the time of testing whether or not there is any connection between the breath alcohol and blood alcohol. This, I would think and to put it mildly, raises a number of due process concerns.
{ 3 comments… read them below or add one }
I have to agree that DC has peculiar standards. Shouldn’t those trying to convict using the breath analysis device ideally have to make a showing that 210 liters of breath were sampled and from that 0.08 grams of alcohol in gaseous form or more were found to be present? Or being more reasonable, since I would guess that 210 liters of breath would probably be the rough equivalent of blowing up a small dingy, the test must scale the input breath volume down to a few liters. Suppose it divided both values by 100 to make it apply to 2.1 liters of breath, then only 0.0008 grams (eight ten thousandths of a gram) of alcohol or more would have to be present to convict. But if any liquid portions of that night’s drinking were present in spittle or other residues such as saliva below the tongue, etc. and entered the meter as either liquid or vapor, the integrity of the reading would be compromised and the scaling would have introduced further uncertainty and possibly error.
I agree completely. And the problem you identify is not specific to D.C. Most jurisdictions now use the 210 liters of breath standard.
As I understand it, 210 liters is roughly equivalent to the volume that would fit into a 55-gallon drum. The lung capacity of a human being is about 4-6 liters of breath, and the breath machines capture only about 50 cubic centimeters of this breath for analysis. Write Taylor/Oberman in their book, Drunk Driving Defense, “the machine is capturing a breath sample that is a tiny amount relevant to the charge: The statute’s 210 liters is about 4,200 times more than the sample being tested by the machine. Assuming a reading on the machine is 0.08 percent, this 50 cc breath sample would contain about one-millionth of a fluid ounce of alcohol. Is the machine capable of reliably and accurately measuring such an infinitesimal amount?. . . What would be the impact of even the slightest error if magnified 4,200 times?”
Note that the quote from Drunk Driving Defense and our normal use is to speak of 0.08 as a percent, but the statute uses 0.08 as a weight in grams. So the resulting ratio is weight to volume (grams to liters), not percent. My high school chemistry is not good enough to calculate the volume of gaseous alcohol molecules implied by a particular weight, but there is no reason to think that whatever the resulting percentage might be that it is 0.08. But by chosing 210 liters the authors of the statute must have reasoned that the resulting ratio, whatever it might be, always reflected the equivalent of the blood/alcohol volume ratio.
Now imagine two suspects with identical blood/alcohol readings but very different sizes and weights. The statute assumes that whatever volume of air the first needs to supply oxygen to double or triple the number of cells than the other suspect will not matter. They will both get the same breath machine readings. But do they in fact? It would seem like a relatively easy test to perform.