Each Test Stands on Its Own
Sample cross-examination on the ATC and CFS Position Paper policy that every test stands on its own - that "reliability" has nothing to do with tests or data before or after the subject tests.
Purposes of this cross-examination:
To challenge the ATC hypothesis or policy that "each test stands on its own".
To identify the hypothesis as faith or policy, not scientific opinion.
To demand the wording of the hypothesis and the empirical testing methodology if it is a scientific opinion.
To identify exactly what other anchoring information ATC/CFS considers essential to reliability notwithstanding this sweeping statement.
To identify exactly what other information they label irrelevant.
To obtain an admission that this is not scientific opinion.
To educate the Court as to the difference between scientific opinion and technical opinion.
To separate the difference between a priori logical opinion and an opinion based on empirical science.
To identify the lack of empirical studies using "sampling" of aging instruments in the field.
To establish that the ATC Position Paper is not a consensus document generally accepted by all forensic toxicologists.
A. You need to look at that information for the
purposes of the breath test that was done in that particular
case. It's irrelevant to the test that was done in this
Q. Now, that concept, that every test stands on
its own, that's not a statement of scientific opinion; that's
a statement of policy?
A. Sorry, a statement --
Q. Your evidence in-chief...
Q. ...was that every subject test stands on its
Q. That's a statement not of scientific opinion,
it's a statement of policy?
A. It's a scientific opinion, not policy.
Q. All right. How can that opinion be stated as
a hypothesis and tested? 1.
A. So as I've said previously, each test is
separate and independent, so there's a particular breath test
sequence that occurs for a particular subject test, right?
When that subject test is over, nothing else that happened at
other times has an effect on that test. What happened and
what effects that test, whether it's radio frequency
interference, ambient alcohol, mouth alcohol, or any other
number of messages, is relevant to that test at that time.
And, again, the next test is conducted, again, at a random
moment in time, and you have two tests that, according to the
recommendations, is that the two tests have to be within good
agreement and within 20 milligrams of alcohol in 100
millilitres of blood. Again, that increases the scientific
confidence in the reliability of the results.
Q. My suggestion to you is that that's a
statement of policy.
A. It's a recommendation at the Centre of
Forensic Sciences and the Alcohol Test Committee. It's
clearly in all those documents.
Q. It's a statement of policy by the Alcohol Test
Committee and by the Centre of Forensic Sciences, it's not a
A. It's a scientific -- scientifically
established parameter that has been in use long before I was
a forensic toxicologist.
Q. It's a guideline, it's a protocol, but it's
not a scientific opinion.
A. It's a recommendation based on science.
Q. And how would one -- if it is based on
science, how could that statement be expressed as a
scientific hypothesis and tested?
A. I can't think of a way to do that right now.
Q. You see, my suggestion to you is that it may
be a very, very good technical recommendation in an ordinary
case, where everything else is working normally, where the
police are following all the other recommended practices. It
may be a good approach, a good standard operating procedure,
if you will, but it is not a scientific opinion. It is not
an opinion that can be stated as a hypothesis and tested.
A. I would disagree with you. There are numerous
papers that's have been published by respected authors who
have said that two tests that don't differ by a particular
parameter, that are taken within a certain period of time and
considered to be in good agreement, is scientific reliability
and accuracy of the readings that were obtained at that time.
Q. Conclusive --
A. From the United States and Canada.
Q. Conclusively. Not just as a general
guideline, but that it is a conclusive determination
scientifically. Where's the empirical authority for saying
A. Again, because you have tests that are taken
at random moments in time, and the likelihood of one --
something happening on one test is possible, but the
likelihood of anything happening to two tests at two random
moments in time is so unlikely as to be a impossible. And
Q. So your first...
A. ...the basis for that.
Q. So you're expressing that statement as a
matter of logic, not as a matter of scientific opinion that
A. I'm not sure how you would be able to test
Q. Well, let me just give you some possibilities.
A. All right.
Q. One possibility might be for the Centre of
Forensic Sciences or the Alcohol Test Committee, or somebody
in the system, to collect together to sample - I think you
know what I mean by sample - to sample some aging instruments
that are out in the field and bring them in and do empirical
testing and see if this what I suggest is a policy,
empirically works or whether it does not work. That would be
one methodology that might be attempted in order to check
whether or not this what I am suggesting is a statement of
policy, holds scientifically in empirical testing.
A. The proper working order of an instrument is
determined at the time of testing.
Q. Well, I understand that's the policy of your
A. Recommendations. You keep saying "policy".
They're recommendations. The Alcohol Test Committee's
recommendations. The manual that the Centre of Forensic
Sciences has produced is recommendations, not policy.
Q. But not every scientist who is a member of the
Canadian Society of Forensic Sciences agrees with that
policy. They say that there are gaps in it.
A. If you would like to provide me with a list of
those people, I can review that list.
Q. Well, we...
A. But the authoritative bodies in Canada agree
with those recommendations...
Q. We saw...
A. ...and that includes Ontario and Canada.
Q. We saw in Exhibit 17 and then 17B - those were
the exhibits - they were the correspondence between
Mr. Kupferschmidt and the editor of the Canadian Society of
Forensic Science Journal?
Q. And Mr. Kupferschmidt said that there are
problems with the policy, and one of the big problems is that
there is no entity that is out there that is supervising the
police in terms of metrological control. I don't know
whether he used the words "metrological control". But he
said that there's no body that's out there that's supervising
the police to make sure that they are following the best
practices of the Alcohol Test Committee.
A. That's correct, yes.
Q. So there are scientists that are out there
that disagree with the policy that you can determine
everything on the basis of just the subject test and the data
that comes out of the subject test, right?
A. You've given me the name of one individual.
A. And there's always going to be experts out
there who disagree with what is going on there, what's going
on in the breath program, what's being done with respect to
breath testing and instruments and the reliability.
Q. But my suggestion to you is, there are no
empirical studies out there. I mean, if I look through the
Wigmore book again that you referred to on the previous
Q. Everything that we would ever want to know
about alcohol testing. It's filled with empirical studies,
including some of your own.
Q. There's no empirical study in that book, or
anywhere else out there, that suggests that you can
scientifically close the door on problems with accuracy and
precision by conducting just a single pair of subject tests,
following the rules that you have outlined.
A. The program that's in place is fit for its
purpose, it works well, and that's all I can... I mean, we
can continue to argue about this point forever, but we
disagree -- we agree to disagree.
1. Reference: NJI Science Manual for Canadian Judges