This case started like many DUI or OUI cases in Maine. My client was driving home from a bar at around midnight. A Yarmouth, Maine Police officer saw his car driving too slow, watched the vehicle touch the yellow center line several times and noticed a brake light out. The officer stopped the car and had the driver perform field sobriety tests. The traffic stop and roadside tests were recorded on video. The officer saw clues of impairment on the tests and he arrested my client for OUI. The driver was taken to the station to do a breath test on an intoxilyzer 8000. His breath alcohol level tested at .20; the legal limit is .08. Because of the high test, a 2 day jail sentence was mandatory. We had a trial and the jury returned a not guilty verdict.
Maine OUI laws provide that a person is guilty of OUI if they operate a motor vehicle:
- With a breath alcohol level of .08 grams per 210 liters of breath, or
- While impaired to any degree by intoxicants
The state can prove either that the person was impaired or that the alcohol level was .08 or more, they don’t need to prove both. In this case, the State also charged that the alcohol level exceeded .15 grams per 210 liters of breath. That triggers a mandatory jail sentence if proven. To win the case, we had to beat both the impairment and alcohol level allegations.
Proving Impairment: Observations & Field Sobriety Tests
The officer stopped my client for touching the yellow line on several occasions and for having a brake light out. But the officer testified that the driving he witnessed did not show many of the clues that he is trained to look for in drunk driving cases.
- The vehicle did not straddle the center line
- He didn’t weave within his lane or swerve
- He didn’t take any wide turns
- He stopped appropriately and did not try to flea
- He had no issue handing over his license, registration and insurance
- He exited the car without incident
Once out of the car, the driver agreed to do the three standardized field sobriety tests. The officer is trained to administer the each of the tests in a particular way and then to look for specific “clues” of impairment. This officer noted the clues he observed and testified about them at trial:
Horizontal Gaze Nystagmus test, HGN: 5 of 6 Clues
This test looks for an involuntary jerking of the eyes or ‘nystagmus’ as they follow an object moved from side to side across the field of view. There are lots of of other things that cause nystagmus and one would naturally see the symptom when a person tracks a fast moving object, or when the person watches flashing lights.
The officer claimed that he observed 5 of a possible 6 clues of impairment on this test. The problem was that the test was administered improperly. The officer testified on cross examination that he moved his finger (which the driver was instructed to follow) too quickly, and that doing so could cause the nystagmus he saw. Part way through the test, a backup officer arrived with blue lights flashing and those remained illuminated for the rest of the test. The officer agreed on cross that the proper procedure was to turn off all strobe lights since the flashing can cause nystagmus too. The officer also testified that nystagmus looks the same no matter what the cause.
Walk and Turn: 5 of 8 Clues
This test requires the subject to walk a straight line taking 9 steps out, then turn 180 degrees and walk 9 steps back. The steps are taken so that the heel of the front foot should touch the toe of the trailing foot. The officer claimed five clues of impairment: the subject started before instructions were finished, stepped off the line, missed heel toe, turned improperly, and raised his arms for balance. But the clues he claimed were not clearly reflected on the video and the overall performance looked pretty good. The officer admitted on cross examination that his instructions were vague in some respects and that some of the clues he claimed to observe were not really present in the video. My client did take the correct number of steps, turned and returned to the starting point.
One Leg Stand: 2 of 4 Clues
This test requires the subject to balance on 1 foot with arms at his side while looking down at the foot he’s holding off the ground and counting for 30 seconds. The officer claimed my client swayed and raised arms for balance while standing on one leg. Because the test was performed just out of frame, the cruiser video only showed half of my client’s body. Still, it was pretty clear that there wasn’t any swaying or arm raise that would count as a clue in what you could see. The subject did keep his foot off the ground for the full 30 seconds and didn’t hop, or lose balance.
Proving the Alcohol Level: The Intoxilyzer 8000
Click to see the breath test
My client’s intoxilyzer 8000 printout showed no error codes or other issues. It was accepted into evidence and documents a breath alcohol level of .20 grams per 210 liters of breath, 2.5 times the legal limit. Still, the Intoxilyzer 8000 is not a perfect machine and the evidence at trial showed that there were problems inherent in the testing method and specific problems with the device used in this case.
General Problems with Breath Alcohol Testing
The limit is .08 but what does that mean?
It’s important to keep in mind exactly what that Maine OUI law prohibits: operating at or above .08 g of alcohol per 210 L of breath. When I cross-examined the State’s chemist, I tried to put those numbers in perspective. One liter contains 1000 cubic centimeters (CCs) or 1000 milliliters (ml) of volume. A liter of water weighs one kilogram and so 1 CC of water weighs 1 gram. .08 grams is eight hundredths of 1 mL of water; something like a very small droplet. Alcohol is a bit lighter than water but he amounts are basically the same.
210 L on the other hand, is a pretty huge volume. I used a 5 gallon bucket at trial and was able to establish through the chemist that the bucket contains approximately 20 L. It would therefore take 10.5 five gallon buckets to contain 210 L of breath, that’s 55 gallons of breath. So in order to render a result in terms of grams per 210 L of breath the machine needs to precisely determine an extremely small quantity of alcohol in an extremely large volume or breath.
Of course, to give that result, the intoxilyzer doesn’t actually test 210 L of breath since thats dozens of times larger than the lung volume of an adult human. In practice, a person who blows into the machine produces something between 1.5 and 3.5 L of breath. Even so, the machine does not actually test all of the breath that the subject exhales.
In fact, the chemist’s testimony established that the sample chamber on and Intoxilyzer 8000 has a volume of approximately 26 ml. It’s the breath in that chamber which the machine actually tests. 26 ml is a bit less than 1/8000 of the 210 L that Maine’s law contemplates. So the machine works by extrapolating from a minuscule sample to render a result in the terms of the huge volume that the statute uses. By using this method, even a tiny error in the sample chamber will be magnified by a huge multiplier.
How Does the Intoxilyzer Reach a Result?
To test the breath sample, the Intoxilyzer 8000 shines infrared light through that 26 CC sample chamber to reach a sensor at the other end. Alcohol molecules will absorb more of that light than normal breath air. The machine knows how much light is emitted and measures how much light arrives at the sensor. The idea is that by comparing these two values the machine can determine the amount of alcohol in the sample chamber.
Of course, the amount of light reaching a sensor doesn’t mean anything until it’s interpreted by the machine’s software. The State chemist testified at trial that the intoxilyzer 8000 is “very software driven” and that the software controls everything from user interface to calculation of the ultimate breath alcohol level. On cross examination I asked the chemist what calculations the software does to turn the light received the sensor into an alcohol test result. He answered that he didn’t know. I asked him what formulas the machine used and he said he didn’t know. I asked him what kind of math the machine would have to do to render that result. He answered that it was all done in terms of voltage but that he was not an engineer and the mathematical questions were “above his pay grade.”
I really feel for my colleagues in Ohio like Douglas Riddell and Charles Rowland. Even with all these issues, rules in that state bar attorneys from attacking the general reliability and accuracy of breath alcohol testing.
Problems with the Yarmouth, Maine Intoxilyzer 8000
Maine law requires only that each breath test machine be certified by a state chemist once every seven months. Still, many police departments, including Yarmouth, do their own calibration checks approximately once every two weeks. In these checks, a certified operator runs a test using an alcohol and water solution mixed at a concentration that should measure .09 g per 210 L. Maine’s regulations require that the machine return a result within .01 of the .09 known value.
I got the calibration check records from this machine and they showed six of the biweekly checks performed over the four months following my client’s test were outside of the .01 tolerance. Each of the 6 checks was low, reading the .09 solution at .072-.079. The State Chemist admitted on cross examination that the regulations require .01 discrepancies be investigated and that if the problem is with the machine it should be taken out of service. He testified that he never investigated the .01 discrepancies and none of the records he had from his office or from Yarmouth PD showed that there was ever any follow-up to those calibration check problems.
On redirect, the chemist tried to explain why the calibration checks might be low. He testified that, though he was never notified of the calibration check problems, he was consulted because Yarmouth PD’s Intoxilyzer room was too cool. My guy’s test was in April and at the end of the summer they added a heater to the room. The cool ambient temperature was causing the metal components like the simulator solution lid to form alcohol condensation. The chemist speculated that this might be a cause for the low calibration check results since alcohol could have condensed out of the simulator solution vapor before reaching the sample chamber.
This testimony was pure speculation and didn’t do much to explain the calibration issues. While the state offered this testimony to help it’s case, I took it as proof that the machine was not operating properly at the time of my client’s test. I was able to argue in closing that it was possible the machine had accumulated alcohol condensation which could cause the test to be artificially high.
The OUI Case Goes to the Jury
In my closing argument, I summarized all the of the problems with the state’s evidence of impairment, the problems inherent in testing on the intoxilyzer 8000, and the specific problems with this machine which we learned about from the prosecution’s expert witness. That witness had also testified that he knew about the clinical effects of alcohol on the body. He established on cross examination that an alcohol level of .50 is almost always fatal. One of the best arguments in closing was that my guy couldn’t be .20, almost halfway to dead, and still perform as he did on field sobriety tests. I argued that there were too many uncertainties, and that it was reasonable to to have questions and doubts about what the evidence meant and how reliable it was. I explained to the jury that if they still had those questions and those doubts, they had already reached their verdict.
The jury began deliberations and almost immediately wanted to see the cruiser video again. Though the officer claimed clues of impairment they were not obvious on the recording and, in watching the video from start to end, my client looked quite good. The jury resumed deliberation and then sent back a note claiming that they were hopelessly deadlocked after only 90 minutes out. The judge instructed them that they must continue to deliberate. They continued to work and then asked for my cross examination of the state’s chemist to be read back to them. After the read back, they deliberated another 90 minutes before returning a not guilty verdict.
