The Florida Department of Law Enforcement has total faith in the Intoxilyzer 5000, the machine police agencies across the state use to conduct breath tests in DUI cases. It's been in use since 1992, and the FDLE swears it is accurate.
But in Seminole County, things aren't so clear-cut. In 2005, four county judges have refused to admit breath tests in hundreds of cases, causing some slam-dunk convictions to be dismissed. The judges have been pilloried in the media, especially on the Orlando Sentinel's editorial page, which declared that they were unfairly hindering the state's drunken-driving cases.
At issue are the Intoxilyzer's source codes, or the software programming instructions that are kept under lock and key in the Kentucky offices of CMI, Inc., the Intoxilyzer's manufacturer. CMI says the source codes are a trade secret and refuses to share them with anyone, even the state. But defense attorneys have started to demand them, saying there's no way a defendant can analyze the machine's accuracy without access to those codes. When the state declined to force CMI to turn over the codes, defense attorneys filed motions to suppress, and in Seminole County, the judges agreed. The state is appealing the ruling to the 18th Judicial Circuit. From there, it's likely to go to the 5th District Court of Appeals in Daytona Beach.
Defense lawyers say this about a defendant's constitutional right to confront the evidence used against him. I spoke with Rigo Armas, the lawyer and engineer who contracts with the 18th Judicial Circuit's public defender's office, to argue the need to reveal the source codes.
OW: Please explain how all this got started, and how you came to argue these cases for the public defender's office.
Armas: Well, my wife worked for the public defender's office when we first arrived here in Orlando, and well, obviously since I'm an attorney as well, she'd always consult me and what my opinions were on different cases. And when the DUI cases came up, because of the scientific nature associated with the Intoxilyzer, she asked me to look into it.
OW: Your wife [assistant public defender Andrea Armas] feels the judges have been beaten up on because of uninformed reporting . Do you share that view?
Armas: I feel that yes, that they have not been represented correctly in the media, because I believe that the popular view is to simply say DUI, drunk driving, that's bad, everything has to be done against a [defendant] and whenever a judge takes a stand that in any way seems to be helpful to someone that is charged with something like that, it's just good to beat them up. … I think if detailed research were done of what occurred during these cases, one would find that the judges really tried to do everything they could to continue the prosecution, and were simply left with no choice … but to withhold use of the Intoxilyzer.
OW: How is this not a case of judges allowing otherwise guilty defendants off on a technicality?
Armas: Well, first of all, they don't get off … because there is other videotape evidence that can be presented; the officers can testify; there is a lot of other evidence that can be used to prosecute any given individual. What's being prevented is the use of this machine because of the lack of the ability to study the machine. It's not a technicality to prevent the state from using "a secret machine." … If you do detailed research on that science, you'll realize that a significant number of experts do not agree on the validity of [this type of testing]. Therefore, there are significant challenges that can be made to the results that these machines produce. But what the state does is say, "Look, this machine is good, the information associated with it is [a] trade secret and we're not going to let you look into it. The manufacturer doesn't want you to, and that's it, period."
OW: What could you glean from having access to the source codes?
Armas: Well, I can tell you that the machine is checked monthly, and when it's checked, it's checked through a simulation port under very strict, controlled atmospheric conditions: pressure, volume, temperature. … The machine, at that point, sucks the vapor from the simulators into it. Well, when a human being blows into the machine, they blow through a different pathway to a tube that's heated. With a human they have … to heat the tube for the sake of accounting for humidity and other things of that nature, so that's a different pathway. And when you're looking at something from an engineering point of view, you want to know everything you possibly can about it before you make any conclusions. By getting specific design specifications for the chamber size and what the tolerances are of that chamber, you could have a better chance of determining what types of errors could be going on.
OW: Do you think if you had access to the source codes, there would be significant reason to question their accuracy?
Armas: I do, because I don't believe in breath science, in the science of it. What you have to understand is that there are a significant amount of experts that would challenge it, and of course the state would present a significant amount of experts to say it's OK. What it comes down to, it's a ballpark figure when you're measuring human beings.
OW: Is there a more accurate or consistent way to test one's blood alcohol level than breath tests?
Armas: What you have to remember is that there is nothing wrong with you having alcohol in your breath; there is nothing wrong with you having alcohol in your blood; the problem is having alcohol in your brain. That's where it affects you. But we can't put a drill to your brain to figure out how much alcohol you have in there in a given situation. So the next best place to read it is your blood. And then they decide that's expensive. Because you have to have a lab tech, it can be [an] invasion [of] your privacy, storing it can be an issue, and there are a lot of ramifications associated with it. So then you get the next solution, which is bearing the cost of buying the [breath-test] machine. The training of the individuals is fairly inexpensive … and the mouthpiece that each individual is tested with is 28 cents, from what I've heard, so that's a pretty inexpensive way to test them. And I've always said you get what you pay for.
OW: Do the cops need the tests to get convictions?
Armas: No. There's plenty of other evidence associated with the specific case. And to be honest, especially in Seminole County, they're on notice that this evidence from this machine the judges are not happy, especially since statute requires full information to be provided [to defendants]. You would think that they would try to investigate, prosecute these cases in other ways that would give them more evidence other than the use of this machine.
OW: How many cases have been thrown out?
Armas: I don't think anything's really being thrown out, it just gets pled differently. I'm not speaking from direct knowledge on that issue because I concern myself more with the initial motion to prevent the machine from being used.
OW: Is there a way to protect trade secrets and allow access to the machine?
Armas: In theory, yes: nondisclosure agreements. I would sign one and state that I would not release the information for any use other than the use of defending my client.
OW: Is the fear that you or someone else is going to look at their source codes and create their own machine?
Armas: Well, if I were going to guess, and this is my personal opinion, I think initially CMI seeing that I was an engineer, they might be concerned with my desire to reproduce the machine. I think after years of doing this they realize that's not my interest, but unfortunately, regardless of my personal desires, the issue becomes how you protect this information from other people that would want to do that. But my thought is it's a necessary problem for the business to deal with if they're going to deal with generating evidence within the courtroom. If you want to use these machines for DOT or employment areas or things of that nature, fine, there's enough business there for them to do and keep secret. But you can't use secret technology to prosecute a human being. Not in the United States anyways.
OW: Other jurisdictions have turned down motions to suppress. How are the ones in Seminole County different?
Armas: I haven't necessarily seen the arguments presented in the other counties, but from reading the orders, I can say that when I first started this set of arguments I asked for the information so that I could show that the machine that they were using was an approved machine. My concern was I really didn't want to have to study a machine and spend the amount of time that it was going to take me to read the code and really analyze it, and then have them come back and say, "Oh that's really not the approved machine, this is." I initially said, "I want the information and I want you to show me that the information you're giving me is the information of the approved machine." And it became evident that they couldn't even [do] that. They just did not fully document in 1993 what they checked out and what they accepted from CMI.
OW: How would you rate your chances as you head up the appellate ladder? There's obviously pressure on the judges to allow the cops latitude.
Armas: I've given up trying to think of that. I've started this fight, I've quit it, I've started it, I've quit it, and now I just, I really can't put a handle on what a judge is going to do. I can tell you that if you look at this from a critical analysis and legal point of view as well as scientific point of view, there is no one that can convince me there is any other decision other than either you exclude the machine, or provide the firstname.lastname@example.org