The DUI Exception to the Constitution

Continued from page 3....

One of the problems is called retrograde extrapolation. It's all well and good the the suspect was a .11 at the time that he breathed into the machine at the police station. But it's not against the law to be over a .08 in a police station. It's against the law to drive a car over .08. What was the blood alcohol level at the time he was driving?

Well, that caused prosecutors a lot of problems. And so most states, almost all states, passed a new law -- at the urgin of the Mothers Against Drunk Driving. And that law said, If any test within three hours of driving results in a blood alcohol reading of .08, it shall be presumed that it was the same at the time of driving. Even though we know absolutely, as a matter of scientific fact, that it was not.

Well, that again raised the conviction rate, except that it was a rebuttable presumption. In other words, you could introduce evidence that that simply wasn't true. And so now, to make a depressingly long story short, some states are beginning to pass laws saying that the crime is having over .08 at the time you breath into the machine. And they don't care what you were when you were driving the car.

Notice how we've gotten further and further and further away from the evil we were trying to cure in the beginning: Were you impaired by alcohol when you were driving your car?

Okay. Let's take a look at this machine. And in most cases, because it's cheaper, easer and faster than blood analysis, the police are going to have you breathing into one of these machines. Let me just give you a real quick rundown on breath test device theory. Incidentally, the book I wrote, Drunk Driving Defense, is about 1200 pages in length, of which nearly 400 pages are just on the technology of breath machines, so this is going to be very cursory. But, I think, for purposes of illustrating some of the problems, it will help.

Basically, these machines capture breath from your lungs into a sample chamber. In the common Intoxilyzer 5000, for example, it's a little 81 cc nickel-plated tube. And, by the way, there are a lot of different manufacturers of these different machines, different types, but we'll get into that in a moment. These machines rust and corrode inside. The sample chamber gets pitted, and it absorbs or retains in these pits alcohol from previous subjects. But let's just say for purposes of theory it captures a sample, a given volume of alveolar air from the lungs. At one end it has a projector that sends a beam of infrared energy, infrared light, through the chamber. The light passes through the breath sample that is passing through that little chamber. Now, at the other end of the chamber is a detector that receives the light from the other side. It then measures how much of this infrared energy gets through -- that is, is not absorbed by the breath and does not reach the other side of the chamber. It measures this dmunition of light energy by comparing the original intensity of the light to the diminished intensity of the light, multiples it by 2100, and spits out the blood alcohol concentration.

Now the theory of infrared spectroscopy, as applied to DUI cases, is based upon the fact that any compound containing the methyl group of molecules will absorb the energy from this 3.61 micron light wave. One of those compounds is ethanol, also called ethyl alcohol. Molecules of ethyl alcohol contain the methyl group as part of its structure. It is resonant with this particular frequency of light wave. So, the more ethyl alcohol in the sample chamber, the more energy is going to be absorbed, the less will get through to the receptor, the higher the blood-alcohol reading will go. Actually, the concept is fairly simple. Except, again, the light waves are absorbed not only by ethyl alcohol, but also by any methyl group in any compound. In other words, it is a stupid machine. It does not differentiate between ethyl alcohol and any other compound containing the molecular group. It is what we call non-specific for ethanol. Any compound on your breath that contains the methyl group will be detected as alcohol -- and reported as alcohol. If you happen to have 32 different compounds containing the methyl group on your breath, it not only will report them all as alcohol, it is cumulative: the machine will add all of those compounds, including any ethyl alcohol, and then report it as alcohol in the blood.

So, do any of these methyl groups exist in the human breath? There are a number of scientific studies -- one of which indicates that there are 102 different compounds found in the human breath that can contain the methyl group. So what you are getting is not necessarily alcohol. What you are getting is some unknown cumulative reading of any of these compounds on your breath. If you had been painting a house yesterday, today you would be registering alcohol on our breath machine. If you had been using solvents, or thinners or glue or anything like this, same result. If you had pumped gasoline into your car and inhaled any of the fumes, hours, even days later, you could be breathing out vapors containing compounds with the methyl group in it. If you are a diabetic, or have been on a low-carbohydrate diet, youo would be getting high blood alcohol readings.

Now, I can go on for quite awhile, but the last example I will give you is called the mouth alcohol problem. The machine assumes that the alcohol, or whatever it is measuring, comes from the air in your lungs, and that's why it's multiplying by 2100. Obviously, if it is getting alcohol directly from your stomach or your throat or your mouth, it's going to fool the machine and the results are going to be extremely high. It would take a minuscule amount of alcohol in your mouth, throat or stomach to fool the machine and create a pretty high BAC reading, since it would be incorrectly multiplied 2100 times. This is called the trapped alcohol or mouth alcohol problem.

So, for example, if you burp or belch, and any gases from your stomach come up, or you have an acid reflux condition, or a hiatal hernia, and any of those gases or liquids come up to your esophagus or mouth, remnants of the saturated gas will stay there for about 15 to 20 minutes before saliva dissipates it. It will be breathed into the machine, if you are being tested. The machine will report a reading, which will be falsely high. It does not mean you're under the influence. It does not mean you're really an .08. It's simply that you had alcohol in your mouth or your throat. The police officers are supposed to guard against this by observing you for 15 minutes and making sure you don't burp. They are supposed to sit down and watch you for a quarter of an hour before giving the test. In all the years that I have been defending DUIs, or for that matter prosecuting them, I don't think I've ever encountered an officer who actually did that. They are far too busy to fool around with things like that. They will check the box, and say they did, but it rarely happens; I'm not sure they could even tell if a belch did happen. The 15-minute observation period is a safeguard, but one that the police ignore.

So, these are just a couple of examples. There are hundreds of things wrong with these machines, not just theoretically, but applied to the machines themselves. How accurate are they? They're close enough for government work. In California, for example, the legal standards of accuracy are that you must have duplicate analysis (two sequential tests) and each result must be within .02% of the other. That means you'd have to take two tests. If the first one (just to use a number to make it mathematically easy) is .10%, the next one has to be .08, .09, .10, .11 or .12. Think about it: a 40% range of error is scientifically accurate in a case where the state must prove your guilt beyond a reasonable doubt. In most states, a 40% range of error is considered to be acceptable accuracy. Close enough for government work.

The people that make these machines keep coming out with new models. They'll come out with a model and call it state-of-the-art, foolproof, fail-safe, and then two years later they come out with a new model that takes care of all of the problems found to have existed with the earlier model. Meanwhile 100,000 people have been tested on the old device (which is quietly retired from service), yet none of those convicted can re-open their guilty pleas, or undo the damage done to their lives and careers by the machine. Then a competing breath manufacturing company comes out with a new and improved model that takes care of the problems with their competitor's new model. And this has become a fairly regular battle of the manufacturers, year after year.

4 of 5« Previous | Next »

Back to Top