November, 2013 - If Breath Alcohol Can’t be Measured, It Doesn’t Exist

We find no sense in talking about something unless we specify how we measure it; a definition by the method of measuring a quantity is the one sure way of avoiding talking nonsense.1

 An essential fundamental of experimental science is that if something cannot be measured, it does not exist. One noteworthy scientist, Mr. William H. George, phrased it this way: “[S]cience is measurement. If I cannot make measurements, I cannot study a problem scientifically.”2  Understanding this primary tenant, a follow-up question might be: how precise must a measurement be in order to study something scientifically, and furthermore, if the thing being studied is not measured precisely, then does it exist at all?  These somewhat theoretical questions ought to have great significance to trial lawyers and judges, particularly in the realm of forensic science where measurements are often used to “prove” a crime.

 In the context of drunk driving litigation, the relevance of these questions is obvious.  We “know” someone is guilty because his breath or blood measurement says he’s guilty.  But what do we really know about the measurement itself?  Said differently, at what point does a supposedly scientific measurement of a person’s breath become so imprecise that we can confidently say the thing measured “does not exist?”  In truth, the answer to this question might vary depending on what is being measured and who is being asked, but what never changes is the underlying principle – all measurement is uncertain, at least to some degree.  Yet, in order to know that something exists, we must be certain in our ability to measure it.

 As should be expected, the science of measurement, known as “metrology,” holds that all measurements have some degree of uncertainty.  And, the only way to know the degree or amount of uncertainty for any particular measurement is to calculate that measurement’s “uncertainty budget.”  Properly performed, this uncertainty budget must take into consideration all potential sources of error.  A measurement that does not provide an uncertainty budget is meaningless.  Consequently, it can be said that without an uncertainty budget, the thing being measured does not exist.  In order to bring the thing measured into existence, “‘numerical data reported in a scientific paper [must] include not just a single value (point estimate) but also a range of plausible values (e.g., a confidence interval, or interval of uncertainty). This is done to ensure that the conclusions drawn from the [results] are valid.’”3
 Theoretical scientist Gilbert Keith Chesterton put it this way: “[M]en can construct a science with very few instruments, or with very plain instruments; but no one on earth could construct a science with unreliable instruments. A man might work out the whole of mathematics with a handful of pebbles, but not with a handful of clay which was always falling apart into new fragments, and falling together into new combinations. A man might measure heaven and earth with a reed, but not with a growing reed.4  This quote begs the question: does the Michigan State Police Breath Testing Division make it’s measurements with pebbles or with clay and reeds?

 One way to know the difference would be for the Division to program the state’s DataMasters to report, along with the breath test result, the range of uncertainty associated with the result.  Otherwise, and without the uncertainty budget, the measured value, i.e., the breath test result, is next to meaningless. And “as with all other scientific investigations, laboratory analyses conducted by forensic scientists are subject to measurement error and should indicate the uncertainty in the measurements that are made.”5  “For these reasons, the academy specified that breath test results need to be reported, along with a confidence interval that has a high probability of containing the true [breath]-alcohol level.”6

 The International Organization for Standardization (ISO) indicates that “it is necessary to include a measurement’s uncertainty when reporting a result because knowledge of the uncertainty associated with measurement results is essential to the interpretation of the results. Without quantitative assessments of uncertainty, it is impossible to decide … whether laws based on limits have been broken.”7  Thus, without a reported range (confidence interval) it should be impossible for juries to decide if a person is guilty in a drunk driving case.  The sad truth is juries are not provided with this essential information, and worse than this, they do not know they do not know.  And so, in Michigan it is quite possible for a person to be wrongfully convicted of drunk driving.

 Nevertheless, there may yet be hope for changing the way in which science and scientific evidence is handled in our courtrooms. Largely due to the initial efforts of Washing State attorney Ted Vosk, DUI lawyers throughout the country have begun arguing that breath and blood tests ought not be admitted unless accompanied by a properly assessed uncertainty budget.  While these arguments are not always embraced by the judiciary, some judges and prosecutors are beginning to recognize that these arguments are not simply the folly of theoretical science or fanciful defense attorneys.  And thankfully the United States Supreme Court has begun to open the door to legal arguments that serve to legitimize forensic science.  Justice Stephen Breyer tells us that “[i]n this age of science we must build legal foundations that are sound in science as well as in law.”8  This means that where scientific evidence is relied upon, “[t]he law should seek verdicts consistent with scientific reality”9  “This can only be achieved, however, “by requiring scientific evidence to conform to the standards and criteria to which scientists themselves adhere.”10

 Still, the pathetic manner in which our criminal justice system has traditionally dealt with issues of science has had an adverse impact on these hearings, and so, the results of motions addressing the lack of a reported uncertainty budget have been mixed.  And this is to be expected.  Change takes time and occurs in sometimes tiny steps.

 Thus, in order for any of us to win our battle to keep Michigan’s citizens from being wrongfully convicted of drunk driving, we must keep fighting.  Every opportunity we have we must attempt to convince judges and prosecutors to bring only true science to the courtroom. Yes, this takes persistence, but it is only in this way that our courts and our criminal justice system will be strengthened, and therefore assure that only the guilty are convicted.  So persist away. As Roman philosopher Lucretius said: “The drops of rain make a hole in the stone, not by violence, but by oft falling.”

by Patrick T. Barone

 Patrick T. Barone is an adjunct professor at Cooley Law School where he teaches "Drunk Driving Law and Practice."  Mr. Barone is also the co-author of two books on DUI-related issues, including Defending Drinking Drivers (James Publishing), a well-known and highly respected multi-volume national legal treatise.  He is a frequent lecturer on trial practice and drunk driving defense tactics. He can be contacted on the web at: www.baronedefense


1. Sir Hermann Bondi, In Relativity and Common Sense (1964)
2. In 'Musical Acoustics Today', New Scientist (1 Nov 1962), 16 No. 311, 257
3. Uncertainty in the Quest for Justice; but the fight continues, Vosk, Defense, pg 12 (2013), citing National Research Council of the National Academies, Strengthening Forensic Science in the United States: A Path Forward (Washington, D.C.: National Academies Press, 2009), 87.
4. Heretics (1905), 146-7, Science quotes on:  |  Mathematics (350)
5. Vosk, supra.
6. Id.
7. Id.
8. Vosk, supra, quoting Justice Stephen Breyer’s introduction to National Research Council of the National Academies, Reference Manual on Scientific Evidence, (Washington, D.C.: National Academies Press 2011), 9.
9. Vosk, supra, quoting B. Black, “Evolving Legal Standards for the Admissibility of Scientific Evidence,” Science 239, no. 1508: 1512 (1988).
10. Id.