Pre Bill C-2, the defence could rebut the presumption of identity by leading a defence of the client's evidence of alcohol consumption together with evidence of a toxicologist declared an expert on absorption, elimination, and distribution of ethyl alcohol in the human body. The expert would then do a calculation of blood alcohol concentration at time of driving or care or control which could raise a reasonable doubt, pre C-2, as to the Intoxilyzer results. No evidence would be led as to the reasons for the unreliability of the Intoxilyzer on the particular occasion. Many Judges disliked the defence raising issues about the Intoxilyzer reliability because all they were concerned about was reasonable doubt based on drinking scenario and toxicological evidence. Critics labelled this as the "two beer" defence and convinced Parliament to eliminate this defence of truth.

Strictly speaking, pre-C-2 "evidence to the contrary" was not a "defence". Strictly speaking, it was simply raising a reasonable doubt about the presumption of identity that the Crown was relying upon (see . v. Phillips, 1988 CanLII 198 (ON CA) . Most lawyers, however, called it a defence.

Strictly speaking, post-C-2 and pre-C-46 "evidence tending to show" negated the "presumption of identity" by raising a reasonable doubt about the RELIABILITY, not accuracy, of the approved instrument measurement results (see R. v. St-Onge Lamoureux and R. v. Lam). Many Judges and lawyers had difficulties with these concepts and expected the defence to call a "two beer" defence even if there was clear operator error or malfunction, i.e. a major issue with reliability of the measurement results. In this era, strictly speaking, there was no obligation by the defence to contradict the instrument's result. The issue was "operator error" or "malfunction" (malfunction could include physiological presentation not contemplated by the design of the instrument - elevated body temperature). We had great difficulty, however, in Court, with proving the extent to which the operator error or malfunction affected the reliability of the measurement result. Reliability of the measurement result IS NOT THE SAME THING AS accuracy of the measurement result. Crowns would frequently cross-examine defence experts on "how much difference would the operator error or malfunction make to the measurement result?" (i.e. implying to the "accuracy" (a troublesome and misused term in duimetrology) of the measurement result). Unfortunately not enough defence lawyers objected to such questions.

In R. v Lam, 2015 ONSC 2194 (CanLII), the ON SC held:
"[31] ... St. Onge itself makes it clear that an accused person must link the improper operation or the failure to maintain the breath machine directly to an unreliable result. In other words, it is not enough to simply say: “the machine wasn’t maintained properly”. An accused person must be able to say: “the machine wasn’t maintained properly, and it led to a problem with the machine.” "

Please note the use by Justice Goldstein of "an unreliable result" in paragraph 31 above. The wording is not "inaccurate" but rather "unreliable". Post St-Onge and Lam, Crowns, defence, and Judges needed to ask the question "how much difference would the malfunction or operator error affect reliability of the measurement result, not accuracy of the measurement result. A measurement result should, in Canadian law (see Weights and Measures Act section 4), be applied in the context of "4 (1) All units of measurement used in Canada shall be determined on the basis of the International System of Units established by the General Conference of Weights and Measures." The International System of Units is an international system and includes the VIM. See CGPM, BIPM, VIM, VIML, and GUM in this duidictionary. Post C-2 and pre-C-46, we defence lawyers, in the context of St-Onge and Lam, should have been obtaining expert opinion on the effects of the operator error or malfunction on expanded or uncalculable uncertainty of measurement. See "uncertainty of measurement" and "measurement uncertainty" in this duidictionary. If, for example, the QT used the approved instrument or the wet-bath simulator in a manner not contemplated by the manufacturer's design then uncertainty of measurement may be expanded to an unacceptable (or uncalculable) uncertainty, that exceeds acceptable uncertainty (ATC and CFS call it variability but it is not just instrument variability) assumed by the Alcohol Test Committee, the manufacturer, or the Criminal Code construed and applied in the context of the Weights and Measures Act and the Charter. Please remember that "A measurement result is generally expressed as a single measured quantity value and a measurement uncertainty."(VIM Note 2, 2.9 (3.1) measurement result, JCGM 200:2012(E/F))