Guilty…of Not Being Average?

Posted by Lawrence Taylor on December 14th, 2012

One of the greatest sources of error in breath-alcohol testing is the consistently recurring fallacy that the individual tested is perfectly average in certain critical physiological traits. Put another way, obtaining an accurate blood-alcohol reading in a DUI case is completely dependent on the validity of a number of assumptions.

Unfortunately for the person being tested, these assumptions are usually incorrect: The person tested is rarely “average” in even one of these critical characteristics, let alone in all of them.

For example, all breath testing devices depend on the assumption that the ratio between alcohol in the exhaled breath and alcohol in the blood is 1 to 2100. In fact, the machine is designed to produce a reading based on that assumption; the accuracy of the reading is directly tied to the accuracy of the presumption. Yet, the actual ratio in any given individual can vary from less than 1:1300 to more than 1:3000. So a DUI suspect with a true blood-alcohol level of .07 but a breath-to-blood ratio of, say, 1:1500 would have a .10 reading on an “accurate” breath testing instrument.

Put simply, these machines do not test individuals. Rather, they test the same “average suspect” over and over again, but using the individual subject’s breath.

Another example of the assumption of “averageness” can be found in urinalysis. When a DUI suspect’s urine is analyzed for blood-alcohol, a presumption exists that there are 1.3 parts of alcohol in the bladder’s urine for every 1 part of alcohol in the blood. This 1:1.3 ratio is as fallacious as the 1:2100 ratio’that is, it is based entirely on the ratio found in the average person. In fact, however, the actual ratio found in any given individual can vary greatly. And as the ratio is in error, so will be the final blood-alcohol reading.

Another example of this constant reliance on averages shows itself when the prosecutor in a DUI trial offers evidence of so-called retrograde extrapolation -- a fancy term for guessing backwards. The blood-alcohol level at the time of testing is not relevant to the charge, of course, and so the state will offer evidence to show what the level was when the defendant was driving. This is commonly done by “extrapolating” backward — that is, computing the earlier blood-alcohol level by estimating how much alcohol had been eliminated or “burned off” in the period between driving and testing.

But this requires two assumptions: The blood-alcohol level was declining and the rate of elimination is known. This second assumption involves the further assumption that the “burn-off” rate was .015 percent per hour (sometimes the assumed rate used is .02 percent). How does the prosecution know that the defendant was eliminating (assuming he was eliminating rather than absorbing) at that rate and not at .005 percent or .3 percent? Quite simply, the prosecution does not know: It merely assumes that the defendant eliminates at the average rate. And, of course, error in such an assumption translates into error in the extrapolation.

This ever-present “average person” in the DUI arena is not limited to chemical analysis. We even find him with the arresting officer in the field. When the officer administers the nystagmus' test (“follow my finger with your eyes”) as part of the battery of field sobriety tests, he operates on the assumption that the suspect is “Mr. Average.” The officer has been trained to “read” at what angle the suspect’s eyes begin jerking. If it begins before 45 degrees, the suspect fails. And where does the magic figure of 45 come from? The average person.

Don Nichols, a pioneer among DUI defense attorneys, would point out to juries that his client is female, Chinese and deceased — despite obvious evidence to the contrary. He then explains that statistically there are more women than men in the world, more Chinese than any other nationality and more dead human beings than living ones. Statistically, then, the average person is female, Chinese and deceased’and so, according to the prosecution, must be his client. He also asks the jury how many of them have 2.3 children(the average in the United States).

So why does the state presume facts that are clearly untrue? Simple. It's convenient: it makes prosecution and conviction much easier.
 

PinterestRedditDiggShare