R. v. St-Onge Lamoureux, 2012 SCC 57

SUPREME COURT OF CANADA

Date: 20121102 Docket: 33970

Between:

Her Majesty The Queen and Attorney General of Quebec

Appellants

and

Anic St-Onge Lamoureux

Respondent

• and -

Attorney General of Canada, Attorney General of Ontario, Attorney General of Manitoba, Attorney General of British Columbia, Attorney General of Alberta, Barreau du Québec, Association québécoise des avocats et avocates de la défense, Criminal Lawyers’ Association of Ontario and Criminal Trial Lawyers’ Association

Interveners

Official English Translation: Reasons of Deschamps J.

Coram: McLachlin C.J. and LeBel, Deschamps, Fish, Abella, Rothstein and Cromwell JJ.

Reasons for Judgment: (paras. 1 to 101) Reasons Dissenting in Part: (paras. 102 to 180)

Deschamps J. (McLachlin C.J. and LeBel, Fish and Abella JJ. concurring) Cromwell J. (Rothstein J. concurring)

R. v. St‑Onge Lamoureux, 2012 SCC 57, [2012] 3 S.C.R. 187

Her Majesty The Queen and

Attorney General of Quebec Appellants

v.

Anic St‑Onge Lamoureux Respondent

and

Attorney General of Canada, Attorney General of

Ontario, Attorney General of Manitoba, Attorney

General of British Columbia, Attorney General of Alberta,

Barreau du Québec, Association québécoise des avocats et

avocates de la défense, Criminal Lawyers’ Association

of Ontario and Criminal Trial Lawyers’ Association Interveners

Indexed as: R. v. St‑Onge Lamoureux

2012 SCC 57

File No.: 33970.

2011: October 13; 2012: November 2.

Present: McLachlin C.J. and LeBel, Deschamps, Fish, Abella, Rothstein and Cromwell JJ.

on appeal from the court of québec

Constitutional law — Charter of Rights — Presumption of innocence — Statutory amendments affecting evidence that can be adduced to rebut presumption of accuracy and presumptions of identity in context of prosecution for driving with blood alcohol level over legal limit — Exclusion of possibility that “Carter” defence would suffice on its own to cast doubt on breathalyzer test results — Whether new provisions of Criminal Code infringe right to be presumed innocent — If so, whether infringement justified — Criminal Code, R.S.C. 1985, c. C‑46, s. 258(1) (c), (d.01), (d.1) — Tackling Violent Crime Act, S.C. 2008, c. 6 — Canadian Charter of Rights and Freedoms , ss. 1, 11(d) .

Constitutional law — Charter of Rights — Fundamental justice — Right to make full answer and defence — Statutory amendments affecting evidence that can be adduced to rebut presumption of accuracy and presumptions of identity in context of prosecution for driving with blood alcohol level over legal limit — Exclusion of possibility that “Carter” defence would suffice on its own to cast doubt on breathalyzer test results — Whether new provisions of Criminal Code infringe right to make full answer and defence — If so, whether infringement justified — Criminal Code, R.S.C. 1985, c. C‑46, s. 258(1) (c), (d.01), (d.1) — Tackling Violent Crime Act, S.C. 2008, c. 6 — Canadian Charter of Rights and Freedoms , ss. 1, 7 .

Constitutional law — Charter of Rights — Self‑incrimination — Statutory amendments affecting evidence that can be adduced to rebut presumption of accuracy and presumptions of identity in context of prosecution for driving with blood alcohol level over legal limit — Exclusion of possibility that “Carter” defence would suffice on its own to cast doubt on breathalyzer test results — Whether new provisions of Criminal Code infringe protection against self‑incrimination — If so, whether infringement justified — Criminal Code, R.S.C. 1985, c. C‑46, s. 258(1) (c), (d.01), (d.1) — Tackling Violent Crime Act, S.C. 2008, c. 6 — Canadian Charter of Rights and Freedoms, ss. 1 , 11(c) .

L was charged with operating a vehicle with a blood alcohol level over the legal limit. At trial, she argued that the new provisions of the Criminal Code with respect to breathalyzer test results are unconstitutional. The trial judge found that the statutory amendments did not bar L from presenting a Carter defence to rebut the presumption of accuracy. In light of the evidence, he concluded that L’s testimony about her alcohol consumption was not sufficiently serious or probative to raise a reasonable doubt. Finding that the qualified technician’s explanations were sufficient and that the presumptions established in s. 258(1) ( c ) and ( d.1 ) of the Criminal Code applied, he convicted L. The trial judge upheld in part the constitutionality of the new Criminal Code provisions.

Held (Rothstein and Cromwell JJ. dissenting in part): The appeal should be allowed in part. Sections 258(1) ( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code do not infringe s. 7 and s. 11( c ) of the Canadian Charter of Rights and Freedoms , but do infringe s. 11( d ) . Sections 258(1)( d.01 ) and 258(1)( d.1 ), and s. 258(1)( c ) after severance of the second and third requirements for rebutting the presumptions, are justified under s. 1 of the Charter .

Per McLachlin C.J. and LeBel, Deschamps, Fish and Abella JJ.: A statutory presumption violates the right to be presumed innocent if its effect is that an accused person can be convicted even though the trier of fact has a reasonable doubt. The expert evidence filed in this case reveals that the possibility of an instrument malfunctioning or being used improperly when breath samples are taken is not merely speculative, but is very real. The Alcohol Test Committee of the Canadian Society of Forensic Science has made a series of recommendations concerning the procedures to be followed by the professionals who operate the instruments and verify that they are properly maintained. These recommendations shed light on the circumstances that might explain how an instrument malfunctioned or was used improperly. However, Parliament did not adopt the Committee’s recommendations, and the prosecution referred to no alternative mechanisms that would enable a court to find that the instruments are generally maintained and operated properly or that the rate of failure attributable to improper maintenance or operation is insignificant. The trier of fact could therefore entertain a reasonable doubt about the validity of the test results, since he or she will not have shown why they can be relied on in the case of the accused who is on trial. But a judge who entertains such a doubt will nevertheless remain bound by the presumptions of accuracy and identity of s. 258(1) ( c ) of the Criminal Code and will be required to convict the accused unless the accused rebuts those presumptions in accordance with the requirements of that provision. In view of the mechanism for applying the statutory presumptions established in s. 258(1) ( c ), s. 258(1) ( c ) and ( d.01 ) infringe s. 11( d ) of the Charter .

Whether a statutory presumption can be justified under s. 1 of the Charter depends on several factors, including the importance of the legislative objective, how difficult it would be for the prosecution to prove the substituted fact beyond a reasonable doubt, whether it is possible, and how easy it is, for the accused to rebut the presumption, and, as can be seen from this case, scientific advances. The objective of the amendments — to give breathalyzer test results a weight consistent with their scientific value — is pressing and substantial. Section 258(1) ( c ) of the Criminal Code contains three separate and cumulative new requirements that the accused must satisfy to rebut the presumptions of accuracy and identity. These requirements must be considered separately for the remainder of the justification analysis.

First, the accused must raise a doubt that the instrument was functioning and was operated properly. This requirement is rationally connected with Parliament’s objective. According to the scientific evidence on which Parliament relied, if the instrument functions properly and all the relevant procedures are followed, the results should be reliable. In addition, the measure violates the right to be presumed innocent as little as reasonably possible. The reliability of breathalyzer tests has been recognized by the scientific and legal communities. Moreover, the new provisions do not make it impossible to disprove the test results, but require that evidence tending to cast doubt on the reliability of the results relate directly to possible deficiencies in the maintenance of the instruments or in the test process. Finally, the effects of this limit on the right to be presumed innocent are proportional to Parliament’s objective. The objective of the first requirement of s. 258(1) ( c ), as clarified by s. 258(1) ( d.01 ), is to confirm the scientific value and ensure the primacy of breathalyzer test results. This statutory amendment was a response to the serious disconnect that existed in the fact that the Carter defence had a high success rate despite the recognized scientific reliability of the results. Furthermore, the scheme adopted for breathalyzer tests includes certain guarantees that place limits on police action and protect the presumption of innocence.

Second, s. 258(1)( c ) requires evidence tending to show that the malfunction or improper operation of the instrument resulted in a reading according to which the blood alcohol level of the accused exceeded .08. This requirement constitutes a serious infringement of the right to be presumed innocent that cannot be justified in a democratic society. The requirement that the accused raise a doubt that his or her blood alcohol level in fact exceeded .08 constitutes an excessive burden in the context of a statutory scheme under which the evidence must relate directly to the functioning or operation of the instrument.

The third requirement of s. 258(1)( c ) cannot be justified under s. 1 of the Charter . There is no rational connection between the objective of the new legislative measures and the requirement of adducing evidence to raise a doubt that the blood alcohol level of the accused in fact exceeded .08. This requirement is in addition to the requirement of showing that the instrument malfunctioned or was operated improperly. If the accused has already identified a defect that could cast doubt on the reliability of the results, it is difficult to justify requiring the court to nevertheless accept that the results have probative value if the accused has produced no evidence regarding his or her blood alcohol level.

It was open to Parliament to exclude, in s. 258(1) ( d.01 ), the production of evidence of the alcohol consumption of the accused that tends to show that the instrument was malfunctioning or was operated improperly, and to provide that such evidence is legally insufficient to cast doubt on the reliability of the test results. This exclusion does not infringe the rights protected by s. 7 , nor does it render the rebuttal of the presumptions established in s. 258(1) ( c ) illusory.

Section 258(1) ( d.1 ) of the Criminal Code establishes a second presumption of identity according to which a blood alcohol level over .08 at the time of the analysis is presumed to be the same as the blood alcohol level of the accused at the time of the alleged offence. Since s. 258(1)( d.1 ) exempts the prosecution from having to establish the guilt of the accused beyond a reasonable doubt before the accused must respond, it infringes the right to be presumed innocent. To rebut this second presumption of identity, evidence to the contrary adduced by the accused must tend to show two facts: (1) the consumption of alcohol of the accused was consistent with a blood alcohol level that did not exceed .08 at the time when the offence was alleged to have been committed; and (2) the consumption of alcohol of the accused was consistent with the test results. The objective of these requirements is pressing and substantial. A rational connection can easily be established between each of these requirements and the requirement’s legislative objective. They also satisfy the minimal impairment test. Section 258(1)( d.1 ) strikes a fair balance between collective rights and individual rights, and is part of a broader legislative scheme designed to confirm the primacy of breathalyzer test results. It is a justified infringement of the right to be presumed innocent.

The presumption of identity established in s. 258(1)( d.1 ) is based on the usual behaviour of drivers, who do not generally drink a sufficient quantity of alcohol to alter the results either just before or just after being pulled over by the police. It is in fact the exceptional behaviour of the accused, not the statutory presumption in the prosecution’s favour under s. 258(1)( d.1 ), that makes it necessary for the accused to testify. The choice by the accused to testify in this regard flows from a decision that must be made whenever the Crown’s evidence is sufficient to support a conviction. Thus, the protection against self‑incrimination guaranteed by s. 11( c ) of the Charter is not infringed.

In this case, the trial judge erred in holding that L could rebut the presumption of accuracy of s. 258(1) ( c ) of the Criminal Code by presenting a Carter defence, but that error did not affect his conclusion, since, when all is said and done, he did not believe L. L’s conviction is therefore upheld.

Per Rothstein and Cromwell JJ. (dissenting in part): The appeal should be allowed and the constitutional questions should be answered in the negative.

Sections 258(1) ( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code are based on three quite straight‑forward ideas. These ideas are that if all of the statutory requirements for taking and analyzing breath samples are observed: (1) the breathalyzer results are reliable in the absence of some basis in the evidence to doubt them; (2) the estimated blood alcohol concentration (“BAC”) arrived at by consumption and elimination evidence (so‑called Carter evidence) is not sufficiently reliable to be used to challenge the accuracy of breathalyzer results; and (3) the BAC at the time of testing will not be higher than at the time of driving, unless the accused drank a large quantity of alcohol shortly before driving or consumed alcohol between driving and testing.

None of the challenged provisions limits the right under s. 11( c ) of the Charter not to be compelled to testify. Although all of the provisions are challenged under ss. 7 and 11( d ) of the Charter , the constitutionality of the provisions which address the burden of proof are best analyzed under s. 11( d ) , while those which limit the relevance of, or exclude evidence in relation to, particular issues are best analyzed under s. 7 .

The fact that s. 258(1) ( d.01 ) of the Criminal Code excludes Carter evidence to challenge the proper functioning or operation of the approved instrument does not violate s. 7 of the Charter . The parties contesting the provision have not shown that s. 258(1)( d.01 ) limits in any meaningful respect the right to make full answer and defence. In the face of the compelling evidence presented by the Crown about the generally misleading nature of Carter evidence in relation to the accuracy of the breathalyzer, those challenging the exclusion of this evidence had to advance some evidence suggesting that, despite its great potential to mislead, there remained some reason not to restrict the use of Carter type evidence. There is no such evidence in this record. Although hypothetical scenarios can form the basis of a Charter challenge, they must be reasonable. The other ground advanced in support of the s. 7 challenge — that s. 258(1)( d.01 ) makes a defence “illusory” — must also be rejected.

Section 258(1) ( c ) of the Criminal Code restricts evidence in relation to the accuracy of the device to evidence that tends to show three things: (1) that the device malfunctioned or the analysis was performed improperly, (2) that the improper performance resulted in the determination that the accused’s BAC exceeded .08, and (3) that the accused’s BAC was in fact lower than .08 at the time of the offence. The first two of these elements do nothing more than to recognize the reality that breathalyzer readings, when obtained under the statutory requirements, should be taken as accurate absent some reason to think otherwise. Absent some evidence to suggest that the analysis is not accurate, a reasonable doubt based simply on the general notion that technology may be fallible or that there is a hypothetical possibility not founded on the evidence that the device malfunctioned or was not operated properly would not be a rational conclusion. Thus, requiring the inference of accuracy to be drawn absent evidence to the contrary does not limit the right to make full answer and defence. As for the third component, it does no more than set out in statutory form what this Court has consistently held is required as a matter of logic and relevance to rebut the presumption of accuracy.

A provision limits the right to be presumed innocent guaranteed by s. 11( d ) of the Charter if it either (a) relieves the Crown of having to present a case to meet before the accused is called on to answer or (b) creates the risk of conviction even if, without the provision, the trier of fact could have a reasonable doubt about the accused’s guilt. The presumption of accuracy in s. 258(1) ( c ) does not create a risk of conviction in the presence of a reasonable doubt about guilt. It therefore does not limit the right to be presumed innocent and there is no need to consider whether any limitation is justified under s. 1 of the Charter . In requiring some evidence tending to show improper functioning or operation, the provision simply enacts common sense in light of accepted scientific fact. Parliament is entitled to legislate this rather than require the evidence to be called in every “blowing over” prosecution. Furthermore, in order to constitute evidence to the contrary as a matter of logic and relevance, that evidence must tend to raise a doubt that the BAC in fact did not exceed .08. It follows that this third aspect of s. 258(1) ( c ) simply translates that requirement for materiality into the consideration of whether the device functioned or was operated improperly.

With respect to the presumptions of identity in s. 258(1) ( c ) and ( d.1 ) of the Criminal Code , there is overwhelming evidence that a breathalyzer test administered in accordance with the statutory requirements and which reveals an over .08 result is a reliable indication that the accused had a BAC which was equal to or higher than that at the time of driving. There is no infringement of the right to be presumed innocent by deeming that the BAC at the time of testing is the same as at the time of driving. Parliament has simply legislated well‑established facts so that they do not have to be proved in every case. There is no risk of conviction on the basis of a reasonable doubt that has a basis in common sense and logic in the evidence or the absence of evidence. A doubt about the presumptions of identity based on “bolus or intervening drinking” would be speculative, absent evidence supporting the fact that one or the other of those scenarios had actually occurred. The fact of post‑driving drinking is peculiarly in the knowledge of the accused and it would be unduly onerous to require the prosecution to negate this rather unusual possibility in every case even when it had no foundation in the evidence. Also, the challenged provisions do not relieve the Crown of its obligation to present a case to meet before the accused is called on to answer. Where an over .08 breathalyzer test result is obtained in accordance with the statutory requirements, a trial judge cannot conclude that there is no evidence upon which he could reasonably convict an accused person.

Even assuming that ss. 258(1) ( c ), 258(1)( d.01 ) or 258(1)( d.1 ) of the C riminal Code limit the right to be presumed innocent as guaranteed by s. 11( d ) of the Charter , any limitation is reasonable and demonstrably justified in a free and democratic society.

Cases Cited

By Deschamps J.

Considered: R. v. Crosthwait , 1980 182 (SCC) , [1980] 1 S.C.R. 1089 ; R. v. St. Pierre , 1995 135 (SCC) , [1995] 1 S.C.R. 791 ; referred to: R. v. Boucher , 2005 SCC 72 , [2005] 3 S.C.R. 499 ; R. v. Carter (1985), 1985 168 (ON CA) , 19 C.C.C. (3d) 174 ; R. v. Gilbert (1994), 1994 347 (ON CA) , 92 C.C.C. (3d) 266 ; R. v. Orbanski , 2005 SCC 37 , [2005] 2 S.C.R. 3 ; R. v. Milne (1996), 1996 508 (ON CA) , 107 C.C.C. (3d) 118 ; R. v. Coutts (1999), 1999 3742 (ON CA) , 45 O.R. (3d) 288 ; R. v. Huff , [2000] O.J. No. 3487 (QL) ; R. v. Powichrowski , 2009 ONCJ 490 , 70 C.R. (6th) 376 ; R. v. Gibson , 2008 SCC 16 , [2008] 1 S.C.R. 397 ; R. v. Oakes , 1986 46 (SCC) , [1986] 1 S.C.R. 103 ; R. v. Downey , 1992 109 (SCC) , [1992] 2 S.C.R. 10 ; R. v. Vaillancourt , 1987 2 (SCC) , [1987] 2 S.C.R. 636 ; R. v. Whyte , 1988 47 (SCC) , [1988] 2 S.C.R. 3 ; Dubois v. The Queen , 1985 10 (SCC) , [1985] 2 S.C.R. 350 ; R. v. Hummel (1987), 1987 4075 (ON SC) , 36 C.C.C. (3d) 8 ; R. v. Phillips (1988), 1988 198 (ON CA) , 42 C.C.C. (3d) 150 ; R. v. Drolet , 2010 QCCQ 7719 , [2010] R.J.Q. 2610 ; R. v. Edwards Books and Art Ltd. , 1986 12 (SCC) , [1986] 2 S.C.R. 713 ; R. v. Chaulk , 1990 34 (SCC) , [1990] 3 S.C.R. 1303 ; Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567 ; R. v. Duff , 2010 ABPC 319 , 501 A.R. 122 ; R. v. Gillespie , 2010 BCPC 207 () ; R. v. Muzuva (2010), 206 C.R.R. (2d) 18 ; R. v. Cayer , 2010 QCCQ 9352 () ; R. v. Laforge , 2010 QCCQ 7718 , [2010] R.J.Q. 2537 ; RJR‑MacDonald Inc. v. Canada (Attorney General) , 1995 64 (SCC) , [1995] 3 S.C.R. 199 ; R. v. Seaboyer , 1991 76 (SCC) , [1991] 2 S.C.R. 577 ; R. v. Morgentaler , 1988 90 (SCC) , [1988] 1 S.C.R. 30 ; R. v. O’Connor , 1995 51 (SCC) , [1995] 4 S.C.R. 411 ; R. v. Kasim , 2011 ABCA 336 , 515 A.R. 254 ; R. v. Darrach , 2000 SCC 46 , [2000] 2 S.C.R. 443; R. v. Paszczenko , 2010 ONCA 615 , 103 O.R. (3d) 424; R. v. Grosse (1996), 1996 6643 (ON CA) , 29 O.R. (3d) 785; R. v. Hall , 2007 ONCA 8 , 83 O.R. (3d) 641; R. v. Bulman , 2007 ONCA 169 , 221 O.A.C. 210 .

By Cromwell J. (dissenting in part)

R. v. Carter (1985), 1985 168 (ON CA) , 19 C.C.C. (3d) 174 ; R. v. Gibson , 2008 SCC 16 , [2008] 1 S.C.R. 397 ; R. v. Seaboyer , 1991 76 (SCC) , [1991] 2 S.C.R. 577 ; R. v. Crosthwait , 1980 182 (SCC) , [1980] 1 S.C.R. 1089 ; R. v. St. Pierre , 1995 135 (SCC) , [1995] 1 S.C.R. 791 ; R. v. Boucher , 2005 SCC 72 , [2005] 3 S.C.R. 499 ; R. v. Lifchus , 1997 319 (SCC) , [1997] 3 S.C.R. 320; Dubois v. The Queen , 1985 10 (SCC) , [1985] 2 S.C.R. 350; R. v. Appleby , 1971 4 (SCC) , [1972] S.C.R. 303 ; R. v. Oakes , 1986 46 (SCC) , [1986] 1 S.C.R. 103 ; R. v. Vaillancourt , 1987 2 (SCC) , [1987] 2 S.C.R. 636 ; R. v. Whyte , 1988 47 (SCC) , [1988] 2 S.C.R. 3 ; R. v. Schwartz , 1988 11 (SCC) , [1988] 2 S.C.R. 443; R. v. Bulman , 2007 ONCA 169 , 221 O.A.C. 210; R. v. Grosse (1996), 1996 6643 (ON CA) , 29 O.R. (3d) 785; R. v. Hall , 2007 ONCA 8 , 83 O.R. (3d) 641; R. v. Paszczenko , 2010 ONCA 615 , 103 O.R. (3d) 424 .

Statutes and Regulations Cited

Canadian Charter of Rights and Freedoms , ss. 1, 7 , 11( c ), ( d ).

Controlled Drugs and Substances Act , S.C. 1996, c. 19, s. 51.

Criminal Code , R.S.C. 1985, c. C‑46, ss. 212(3), 253(1) ( b ), 254(2) , (3) , 258(1) ( c ), ( d.01 ), ( d . 1 ), ( g ), 276 .

Criminal Law Improvement Act, 1996 , S.C. 1997, c. 18, s. 10(2).

Interpretation Act , R.S.C. 1985, c. I‑21, s. 25(1).

Tackling Violent Crime Act , S.C. 2008, c. 6.

Authors Cited

Canada. House of Commons. House of Commons Debates , vol. 141, 1st Sess., 39th Parl., January 30, 2007, pp. 6185, 6186.

Canada. Senate. Standing Senate Committee on Legal and Constitutional Affairs. Proceedings of the Standing Senate Committee on Legal and Constitutional Affairs , No. 9, 2nd Sess., 39th Parl., February 21, 2008, p. 37.

Canadian Society of Forensic Science. “Recommended Standards and Procedures of the Canadian Society of Forensic Science Alcohol Test Committee” (2009), 42 Can. Soc. Forensic Sci. J. 1.

Cross on Evidence , 7th ed. By the late Sir Rupert Cross and Colin Tapper. London: Butterworths, 1990.

Hodgson, Brian T. “The Validity of Evidential Breath Alcohol Testing” (2008), 41 Can. Soc. Forensic Sci. J. 83.

Martin, T. L., J. G. Wigmore and K. L. Woodall. “A Comparison of Blood Alcohol Concentrations Estimated From Drinking Histories of Drivers Charged with ‘Over 80’ and Their Intoxilyzer® 5000C Results” (2004), 37 Can. Soc. Forensic Sci. J. 187.

Robertson, Robyn, Ward Vanlaar and Herb Simpson. National Survey of Crown Prosecutors and Defence Counsel on Impaired Driving: Final Report . Ottawa: Traffic Injury Research Foundation, July 2008.

Sommers, Marilyn Sawyer, et al. “‘Nurse, I Only Had a Couple of Beers’: Validity of Self‑Reported Drinking Before Serious Vehicular Injury” (2002), 11 Am. J. Critical Care 106.

Wigmore, J. G. “Man vs. Machine: Self‑Reported Alcohol Consumption of Drinking Drivers vs. Evidential Breath Alcohol Tests. Is the Restriction of Evidence to the Contrary Scientifically Valid?” (2009), 54 Crim. L.Q. 395.

APPEAL from a judgment of the Court of Québec (Judge Chapdelaine), 2010 QCCQ 8552 , [2010] J.Q. n o 10077 (QL), 2010 CarswellQue 10716, convicting the accused of driving with a blood alcohol level over the legal limit and upholding in part the constitutionality of ss. 258(1) ( c ), ( d.01 ) and ( d.1 ) of the Criminal Code . Appeal allowed in part, Rothstein and Cromwell JJ. dissenting in part.

Michel Déom , Jean‑Vincent Lacroix , Marie‑Ève Mayer and Patricia Blair , for the appellants.

Patrick Fréchette , for the respondent.

François Joyal and Ginette Gobeil , for the intervener the Attorney General of Canada.

James V. Palangio and Philip Perlmutter , for the intervener the Attorney General of Ontario.

Christian Vanderhooft and Nathaniel Carnegie , for the intervener the Attorney General of Manitoba.

Rodney Garson and Roger F. Cutler , for the intervener the Attorney General of British Columbia.

Jason R. Russell and Robert Palser , for the intervener the Attorney General of Alberta.

Marco LaBrie and Jean‑Philippe Marcoux , for the intervener Barreau du Québec.

Éric Downs and Julie Bolduc , for the intervener Association québécoise des avocats et avocates de la défense.

Patrick Ducharme and Paul Burstein , for the intervener the Criminal Lawyers’ Association of Ontario.

Shannon K. C. Prithipaul , for the intervener the Criminal Trial Lawyers’ Association.

English version of the judgment of McLachlin C.J. and LeBel, Deschamps, Fish and Abella JJ. delivered by

[ 1 ] Deschamps J. — This appeal concerns the constitutionality of certain provisions of the Criminal Code , R.S.C. 1985, c. C‑46 (“ Cr. C. ”), that deal with offences involving driving with a blood alcohol level over the legal limit . The questions raised in it relate to the right to be presumed innocent, the right to make full answer and defence and the protection against self‑incrimination (ss. 11( d ), 7 and 11( c ), respectively, of the Canadian Charter of Rights and Freedoms ).

[ 2 ] The impugned provisions include four new requirements that must be met by a person charged with driving with a blood alcohol level exceeding 80 mg of alcohol in 100 ml of blood (.08) in order to rebut the presumptions that apply in the prosecution’s favour in such a case. Three of these requirements relate to the presumption of accuracy and one of the presumptions of identity that attach to the results of the test to which a person must submit when required to do so by the police. To challenge the reliability of the results, the accused must raise a doubt: (1) that the breathalyzer instrument was functioning and was operated properly; (2) to the effect that the determination that the blood alcohol level of the accused exceeded the legal limit resulted from a malfunction or improper operation of the instrument; and (3) to the effect that the blood alcohol level of the accused would not in fact have exceeded the legal limit at the time when the offence was alleged to have been committed. Moreover, a new requirement must now be met in order to rebut the presumption of identity of the test results showing that the blood alcohol level of the accused exceeded the legal limit with his or her actual blood alcohol level at the time of the alleged offence. This presumption can be rebutted only if the evidence adduced by the accused shows that his or her consumption of alcohol was consistent not only with a blood alcohol level under the legal limit at the time of the offence, but also — and this is the new requirement — with the test results.

[ 3 ] For the reasons that follow, I find that Parliament was justified in requiring that any evidence adduced to cast doubt on the test results be directed at the functioning or operation of the instrument. However, where such evidence casts doubt on the reliability of the results, the imposition of additional conditions does not constitute a reasonable limit on the right to be presumed innocent. I would reject all the other constitutional arguments that have been raised.

[ 4 ] The impugned provisions are one aspect of the broader fight against drinking and driving, a problem that has preoccupied Parliament and the courts for several decades now. I will therefore begin by reviewing the historical background and the legislative history of these provisions before inquiring into their validity. I will conclude by considering the specific case of the respondent.

1. Historical Background

[ 5 ] In 1969, Parliament made it a criminal offence for a person to operate or have the care of a vehicle while his or her blood alcohol level exceeded .08, and made it mandatory under the Criminal Code to provide breath samples for analysis for the purpose of determining whether that offence had been committed. Among other things, the relevant provisions required a person stopped by the police to provide breath samples and created a mechanism by which those samples would be analyzed by designated technicians using approved devices. Parliament also introduced presumptions (of accuracy and identity) that would apply if certain conditions were met and would make it easier for the prosecution to prove that a person had operated or had the care of a vehicle while his or her blood alcohol level exceeded the legal limit.

[ 6 ] According to the presumption of accuracy, the certificate of the technician responsible for the analyses is presumed to provide an accurate determination of the person’s blood alcohol level at the time the breath samples were taken. According to the first presumption of identity, a person’s blood alcohol level as shown by the test is presumed to be the same as his or her blood alcohol level at the time of the alleged offence. Pursuant to a second presumption of identity added by Parliament in 1997 ( Criminal Law Improvement Act, 1996 , S.C. 1997, c. 18, s. 10(2) ), a blood alcohol level that exceeds .08 at the time of the analyses is presumed to have also exceeded .08 at the time when the offence was alleged to have been committed ( R. v. Boucher , 2005 SCC 72 , [2005] 3 S.C.R. 499, at para. 14 ).

[ 7 ] Before the impugned amendments were enacted, the relevant provisions stated that the presumptions could be rebutted by producing “evidence to the contrary”. The Ontario Court of Appeal considered the meaning of the expression “evidence to the contrary” in R. v. Carter (1985), 1985 168 (ON CA) , 19 C.C.C. (3d) 174, and R. v. Gilbert (1994), 1994 347 (ON CA) , 92 C.C.C. (3d) 266 . It held that, under the provisions in force at the time, the testimony of the accused concerning his or her alcohol consumption, combined with an explanation by a toxicologist of the implications of that consumption, could be tendered as “evidence to the contrary” in order to raise a doubt about the results of the breathalyzer test. This defence is known as the “ Carter defence” after one of the Ontario Court of Appeal cases mentioned above.

[ 8 ] In Gilbert , Osborne J.A., although acknowledging the validity of the defence, had expressed doubts about the chances of succeeding with it (at p. 280):

An accused who is charged with an offence, the essence of which is that he was driving with an impermissibly high blood‑alcohol concentration level must be able to lead evidence as to the quantity of alcohol that he consumed at relevant times. I do not think it is necessary that this kind of evidence be accompanied by an attack on the particular breathalyzer machine, or its operator. It may well be that without such an attack it may be difficult for an accused to have the tendered evidence accepted to the point of raising a reasonable doubt. That, however, does not make the evidence inadmissible generally, or, as I have said, inadmissible because it constitutes an indirect attack on the breathalyzer or its manner of operation.

Despite these reservations, the Carter defence proved to be effective, as can be seen from the subsequent cases on this issue.

[ 9 ] Moreover, it was held that the prosecution could not generally use roadside sobriety tests conducted by the police to incriminate a person who had operated or had the care of a vehicle: R. v. Orbanski , 2005 SCC 37 , [2005] 2 S.C.R. 3, at para. 58 ; R. v. Milne (1996), 1996 508 (ON CA) , 107 C.C.C. (3d) 118 (Ont. C.A.); R. v. Coutts (1999), 1999 3742 (ON CA) , 45 O.R. (3d) 288 (Ont. C.A.); R. v. Huff , [2000] O.J. No. 3487 (QL) (Ont. C.A.) . Furthermore, the results of a breathalyzer test could not be used to assess the credibility of an accused who raised a Carter defence ( Boucher , at paras. 43 and 64 ).

[ 10 ] Because of these rules, it was thought by some that the statutory presumptions attaching to breathalyzer test results did not operate as Parliament had intended. In R. v. Powichrowski , 2009 ONCJ 490 , 70 C.R. (6th) 376, Judge Duncan described what he saw as an impasse faced by the prosecution in certain impaired driving cases under the former legislative scheme (at para. 23) :

An indirect result of the development of the case law, particularly R. c. Boucher as interpreted and distinguished in R. v. Snider (2006), 31 M.V.R. (5th) 296 (Ont. C.J.) and subsequent decisions, was that in cases involving a Carter defence, prosecutors partly abandoned section 258 and attempted to prove their cases the long way around without aid of the statutory presumption of accuracy or, more precisely, the burden of its accompanying jurisprudence. Ironically then, the very legislation that was designed to facilitate proof of the prohibited condition in order to help combat the menace of drinking and driving had become an obstacle to be avoided by the prosecution.

[ 11 ] These difficulties were well known. In a 2006 report prepared for the Department of Justice, Brian T. Hodgson, a forensic toxicology consultant, stressed the importance of re‑establishing the primacy of the test results:

For the continuing use of the statutory legal limit enunciated in subsection 253(b) CCC, over 80, the law needs to reestablish the primacy of scientific evidential results. The defence of “evidence to the contrary” needs to be directed specifically to the factors that impact on the evidential breath alcohol results such as: deficiencies in the test process and/or the drinking patterns of the accused just prior to the time of offence (within 30 minutes) or drinking after the time of offence but before the time of testing. The Supreme Court’s acceptance of the subjective, non‑scientific statements of an accused person about his drinking history leading up to the time of offence without reference to the scientific evidential results is incompatible with the scientific basis of 253(b) CCC. [Emphasis added.]

(“The Validity of Evidential Breath Alcohol Testing” (2008), 41 Can. Soc. Forensic Sci. J. 83, at p. 94)

[ 12 ] On January 30, 2007, Rob Moore, the Parliamentary Secretary to the Minister of Justice, gave an overview of Bill C‑32’s restrictions on the type of “evidence to the contrary” that can be tendered to defend against a charge of impaired driving:

Probably the most important change in this bill is the proposal to ensure that only scientifically valid defences can be used where a person is accused of driving with a concentration of alcohol exceeding 80 milligrams in 100 millilitres of blood. This is known as driving over 80.

( House of Commons Debates , vol. 141, 1st Sess., 39th Parl., January 30, 2007, at p. 6185)

[ 13 ] Bill C‑32 died on the Order Paper, however. Then, on October 18, 2007, the government introduced Bill C‑2, entitled the Tackling Violent Crime Act . Bill C‑2 essentially reproduced Bill C‑32’s restrictions on evidence to the contrary that would be admissible at a trial involving a charge of driving with a blood alcohol level over the legal limit. The Tackling Violent Crime Act (S.C. 2008, c. 6) was assented to on February 28, 2008. Four of the requirements at issue in this appeal came into force on July 2, 2008, while the fifth dates back to 1997. All the provisions in question are reproduced in the Appendix.

2. Changes Resulting from the New Provisions

[ 14 ] Before beginning the constitutional analysis, I should explain how the scheme applicable to prosecutions for driving with a blood alcohol level over the legal limit has been restructured by the statutory amendments.

2.1 Presumptions

[ 15 ] Before 2008, it was settled law that s. 258 Cr. C. established two presumptions of identity and one presumption of accuracy. The amendments have not changed the nature of these presumptions. Section 258(1) ( c ) Cr. C. establishes a presumption of accuracy of the results of the analyses, and a presumption of identity according to which the results are presumed to correspond to the blood alcohol level of the accused at the time of the alleged offence. (In the past, this Court placed the presumption of accuracy in s. 258(1) ( g ) Cr. C. However, the 2008 amendments indicate clearly that Parliament intended them to apply to both the presumption of accuracy and the presumptions of identity, and that it was also incorporating the presumption of accuracy into s. 258(1) ( c ).) Section 258(1) ( d.1 ) Cr. C. establishes a second presumption of identity according to which a blood alcohol level over .08 at the time of the analysis is presumed to be the same as the blood alcohol level of the accused at the time of the alleged offence.

2.2 Standard of Proof

[ 16 ] Nor has the standard of proof that must be met to rebut the presumptions been changed. In R. v. Crosthwait , 1980 182 (SCC) , [1980] 1 S.C.R. 1089 , this Court stated that evidence to the contrary tendered by the accused in respect of the test results was sufficient if it raised a reasonable doubt. In R. v. Gibson , 2008 SCC 16 , [2008] 1 S.C.R. 397, the Court held that the two expressions “evidence tending to show” and “evidence to the contrary” gave rise to the same standard: reasonable doubt (para. 17). The use of the word “conclusive” in s. 258(1) ( c ) Cr. C. does not mean that the presumptions are irrebuttable, as evidence to the contrary can still be presented to counter them.

2.3 Evidence

[ 17 ] The statutory amendments affect the evidence that can be adduced to rebut the presumption of accuracy and the first presumption of identity. The combined effect of the requirements set out in s. 258(1)( c ) and s. 258(1)( d.01 ) Cr. C. is to preclude the Carter defence in its previous form. The accused can no longer simply present a Carter defence. Rather, he or she must (1) raise a doubt that the instrument was functioning or was operated properly, (2) show that the malfunction or improper operation of the instrument resulted in the determination that his or her blood alcohol level exceeded the legal limit, and (3) show that his or her blood alcohol level would not in fact have exceeded that limit at the time when the offence was alleged to have been committed. I cannot accept the interpretation according to which the third of these requirements from s. 258(1)( c ) Cr. C. is not in fact a distinct requirement but follows from proof of the first two ( Powichrowski , at para. 31 ). The wording of the English version of the provision makes it clear that this third requirement is indeed a separate one: the accused must produce evidence tending to show “three things”. Under the provisions as amended, mere evidence that a deficiency in the test process led to a result over .08 is not enough; Parliament also requires that the evidence raise a doubt that the blood alcohol level of the accused in fact exceeded .08. Whereas the evidence needed to satisfy the first two requirements relates to circumstances directly associated with the taking of samples using the instrument, a Carter defence will usually be needed to satisfy the third.

[ 18 ] In short, although the Carter defence may formerly have been sufficient to rebut the presumption of accuracy and the first presumption of identity, this is no longer the case. Two additional requirements must now be satisfied.

[ 19 ] Where an accused challenges the second presumption of identity — according to which, if a person’s blood alcohol level exceeds .08 at the time of the analysis, the same is presumed to have been true at the time of the offence ( s. 258(1) ( d.1 ) Cr. C. ) — he or she is not challenging the test results. Rather, the accused is arguing that, because he or she consumed alcohol shortly before the samples were taken, the result indicating a level exceeding .08 does not correspond to his or her blood alcohol level at the time when the offence was alleged to have been committed. Under the new provisions, the accused can rebut this presumption only by showing that his or her consumption of alcohol was consistent both with a blood alcohol level not exceeding .08 at the time of the offence and with the results of the breathalyzer test. A Carter defence is therefore required to discharge this burden.

3. Issues

[ 20 ] Three statutory provisions are in issue, and several arguments are raised against each of them. I will proceed as follows: First, I will consider whether the three requirements provided for in s. 258(1) ( c ) and clarified by s. 258(1) ( d.01 ) are consistent with the right to be presumed innocent. Because I conclude that the second and third requirements are invalid, only the first requirement will then have to be reviewed in relation to the right to make full answer and defence. The arguments concerning the protection against self‑incrimination will not be considered in relation to s. 258(1) ( c ) and s. 258(1) ( d.01 ) Cr. C. , because they do not apply to them in light of my conclusions with respect to those provisions. I will then consider whether the two requirements established in s. 258(1) ( d.1 ) are valid having regard to the right to be presumed innocent and the protection against self‑incrimination. Finally, I will review the trial judge’s decision and reasons in this case.

4. Compatibility of Section 258(1)( c ) and Section 258(1) ( d.01 ) with the Right to Be Presumed Innocent (Section 11( d ) of the Charter )

4.1 Do these Provisions Infringe the Protected Right?

[ 21 ] If the conditions for the taking of breath samples set out in s. 258(1) ( c ) are met, the trial judge must find that the test results adduced in evidence by the prosecution are, as indicated in that same provision, conclusive proof, for the purposes of the charge, of the blood alcohol level of the accused both at the time when the analyses were made and at the time when the offence was alleged to have been committed unless the accused succeeds in rebutting the presumptions of accuracy and identity.

[ 22 ] In R. v. Oakes , 1986 46 (SCC) , [1986] 1 S.C.R. 103, and in several subsequent cases, the Court considered statutory presumptions adopted to facilitate the prosecution’s task. Under the statutory provision at issue in Oakes , possession of a narcotic gave rise to a presumption against the accused that he or she had the intention to traffic in that substance. The prosecution was thus exempted from proving an essential element of the offence, namely the intention to traffic in a narcotic, and the onus was on the defence to disprove this element. Another presumption was considered in R. v. Downey , 1992 109 (SCC) , [1992] 2 S.C.R. 10. Under s . 195(2) Cr. C. (now s. 212(3) ), living with or being habitually in the company of prostitutes gave rise to a presumption that the person in question was living on the avails of prostitution. Like the provision at issue in Oakes , the one at issue in Downey required the defence to raise a doubt with respect to an essential element of the offence.

[ 23 ] The statutory presumptions established in s. 258(1)( c ) Cr. C. operate differently than the ones at issue in Oakes and Downey . Section 258(1) ( c ) does not exempt the prosecution from proving that the blood alcohol level of the accused exceeded the legal limit, which is an essential element of the offence. However, in proving this essential element, the prosecution can rely on the test results without having to prove that they are valid. In sum, although the prosecution is not exempted from proving an essential element of the offence, the accused must nevertheless raise a doubt about a fact that the prosecution has not established in accordance with the rules of criminal evidence.

[ 24 ] A statutory presumption violates the right to be presumed innocent if its effect is that an accused person can be convicted even though the trier of fact has a reasonable doubt ( R. v. Vaillancourt , 1987 2 (SCC) , [1987] 2 S.C.R. 636, at pp. 654‑56; Downey , at p. 21). In R. v. Whyte , 1988 47 (SCC) , [1988] 2 S.C.R. 3, the Court stressed that the distinction between elements of the offence and other aspects of the charge is irrelevant to the analysis regarding the right to be presumed innocent. “If an accused is required to prove some fact on the balance of probabilities to avoid conviction, the provision violates the presumption of innocence because it permits a conviction in spite of a reasonable doubt in the mind of the trier of fact as to the guilt of the accused” (p. 18). What is important for the purpose of determining whether the right to be presumed innocent is violated is not whether the statutory presumption relates to an essential element of the offence, but whether it exempts the prosecution from establishing the guilt of the accused beyond a reasonable doubt before the accused must respond ( Oakes , at p. 121; Dubois v. The Queen , 1985 10 (SCC) , [1985] 2 S.C.R. 350, at p. 357). Thus, like the presumption at issue in Oakes , the ones established in s. 258(1) ( c ) will violate the right to be presumed innocent if they can result in the conviction of an accused in spite of a reasonable doubt that the accused is in fact guilty.

[ 25 ] It is therefore necessary to inquire into the effect of the presumptions of accuracy and identity provided for in s. 258(1) ( c ) Cr. C. The expert evidence filed in the instant case reveals that the possibility of an instrument malfunctioning or being used improperly when breath samples are taken is not merely speculative, but is very real. The Alcohol Test Committee (“Committee”) of the Canadian Society of Forensic Science (“CSFS”) has made a series of recommendations concerning the procedures to be followed by the professionals who operate the instruments and verify that they are properly maintained: “Recommended Standards and Procedures of the Canadian Society of Forensic Science Alcohol Test Committee” (2009), 42 Can. Soc. Forensic Sci. J. 1 . The Committee states that before collecting a breath sample, the qualified technician must, among other things, observe the test subject for 15 minutes, conduct a system blank test and a system calibration check, and verify the temperature of the alcohol standard, and that the alcohol standard must be changed after a certain number of calibration checks. The Committee also recommends that approved instruments be inspected on an annual basis to ensure that they continue to meet the manufacturer’s technical specifications. According to the Committee, the calibration and maintenance of instruments are essential “to the integrity of the breath test program” (p. 14).

[ 26 ] The Committee’s recommendations shed light on the circumstances that might explain how an instrument malfunctioned or was used improperly. Thus, human error can occur when samples are taken and at various steps in the maintenance of the instruments, which, it should be mentioned, are used Canada‑wide. Hodgson’s report, which the prosecution itself relied on as a source of the statutory amendments, refers to the importance of proper operation and maintenance:

. . . to achieve scientifically sound results in operational use, user agencies must ensure that approved instruments are operated by qualified personnel using procedures based on good laboratory practice. [p. 83]

Moreover, Parliament recognized the importance of following such practices and procedures in s. 258(1) ( c ) and s. 258(1) ( d.01 ), since the accused can rebut the presumptions by showing that the instrument was not properly maintained or operated.

[ 27 ] However, Parliament did not adopt the Committee’s recommendations, and the prosecution referred to no alternative mechanisms that would enable a court to find that the instruments are generally maintained and operated properly or that the rate of failure attributable to improper maintenance or operation is insignificant. The trier of fact could therefore entertain a reasonable doubt about the validity of the test results, since he or she will not have shown why they can be relied on in the case of the accused who is on trial. But a judge who entertains such a doubt will nevertheless remain bound by the statutory presumptions and will be required to convict the accused unless the accused rebuts those presumptions in accordance with the requirements of s. 258(1) ( c ). In view of the mechanism for applying the statutory presumptions established in s. 258(1) ( c ), I find that s. 258(1) ( c ) and s. 258(1) ( d.01 ) infringe s. 11( d ) of the Charter .

[ 28 ] I wish to stress, however, that it is not because the test results could differ from the blood alcohol level of the accused at the time of the alleged offence that s. 258(1)( c ) infringes the right to be presumed innocent. Rather, the infringement lies in the fact that, as Parliament recognized, the instruments can malfunction or be operated improperly, and therefore that the trier of fact could have a reasonable doubt about the guilt of the accused where the only evidence before him or her consists of the test results.

[ 29 ] A clear distinction must be drawn between the stage of determining whether protected rights have been infringed and that of determining whether the infringement is justified. Parliament may have had good reasons for enacting the legislation in question. For instance, the fact that evidence based on DNA analysis is scientifically reliable does not mean that taking DNA samples does not infringe a protected right. If the measures adopted by Parliament infringe Charter rights, the court determining whether the impugned measures are constitutional must consider the justification given for them.

4.2 Is the Infringement Justified?

[ 30 ] This Court has recognized in a number of cases that a statutory presumption that infringes s. 11( d ) of the Charter can nevertheless be justified under s. 1 of the Charter : Whyte ; Downey ; R. v. Hummel (1987), 1987 4075 (ON SC) , 36 C.C.C. (3d) 8 (Ont. H.C.J.); R. v. Phillips (1988), 1988 198 (ON CA) , 42 C.C.C. (3d) 150 (Ont. C.A.). According to the principles stated in those cases, the means available to the accused to rebut the presumption are relevant at the stage of justifying the infringement.

[ 31 ] Whether a statutory presumption can be justified under s. 1 depends on several factors, including the importance of the legislative objective, how difficult it would be for the prosecution to prove the substituted fact beyond a reasonable doubt, whether it is possible, and how easy it is, for the accused to rebut the presumption, and, as can be seen from the instant case, scientific advances.

[ 32 ] The test for determining whether a statutory provision that infringes a Charter right can nevertheless be justified under s. 1 is well known. It was established in Oakes .

[ 33 ] Parliament’s decision to resort to the presumptions of accuracy and identity to help combat the problems resulting from drinking and driving is not at issue in this appeal; rather, what are at issue are the means available to rebut those presumptions. Parliament intended to limit the evidence that can be adduced to raise a reasonable doubt about the reliability of the test results. As can be seen from the legislative history, the objective of the amendments, which form part of a scheme whose purpose is to “reduc[e] the carnage caused by impaired driving” ( Orbanski , at para. 55 ), was to give the reliability of the test results a weight consistent with their scientific value.

[ 34 ] The reliability of breathalyzer tests was explicitly mentioned in the abstract of Hodgson’s report:

The scientific basis for evidential breath alcohol testing is well established. Experiments derived from a recognized scientific law in physics have proven the scientific validity of breath analysis to determine alcohol concentration in the blood. Instruments designed to measure breath alcohol content are based on technology that is capable of producing scientifically sound results. Like Canada, every country that embarks on evidential breath alcohol analysis subjects these instruments to a rigorous evaluation process. These processes determine whether the instruments meet the scientific standards for accuracy, precision, reliability and specificity. [p. 83]

[ 35 ] Specific evidence concerning the reliability of the Alco‑Sensor IV‑RBT IV and Intoxilyzer 5000C instruments was first adduced in R. v. Drolet , 2010 QCCQ 7719 , [2010] R.J.Q. 2610, a case also heard by Judge Chapdelaine, who presided over the respondent’s trial. The parties consented to the filing of that evidence in the case at bar ( 2010 QCCQ 8552 () ). In Drolet , Judge Chapdelaine found that, as a whole, the scientific evidence produced by the Attorney General of Quebec, the Barreau du Québec and the Association québécoise des avocats et avocates de la défense provided [ translation ] “ample” proof of the reliability of the instruments in question (para. 189).

[ 36 ] Both Hodgson’s report and the comments of Judge Duncan in Powichrowski illustrate the problems associated with evidence of breathalyzer test results under the former legislative scheme: such evidence could be rejected on the basis of the testimony of the accused, which was sometimes characterized as subjective recollection. Because it was hard to rely on the test results as effective evidence, the presumptions were less useful than they might have been, and the prosecution was hindered in its efforts to combat drinking and driving. I find that the objective of the amendments — to give the results a weight consistent with their scientific value — is pressing and substantial. As I have already mentioned, however, the three requirements of s. 258(1) ( c ) are in fact separate, and cumulative. I will consider them separately for the remainder of the justification analysis under s. 1 of the Charter .

4.2.1 Evidence of the Malfunction or Improper Operation of the Instrument

[ 37 ] Once the objective has been found to be valid, the Oakes test requires that a rational connection be established between the objective and the means adopted to attain it. It is clear from the words of s. 258(1) ( c ) and s. 258(1) ( d.01 ) Cr. C. that evidence relating directly to the instrument itself or to its operation is now required in order to cast doubt on the reliability of breathalyzer test results. A mere inference based on an individual’s rate of absorption or elimination of alcohol, which is what was required for a Carter defence, is no longer enough. The accused must now raise a doubt that the instrument was functioning or was operated properly.

[ 38 ] In my opinion, the requirement that the accused adduce evidence concerning the functioning or operation of the instrument is rationally connected with Parliament’s objective. According to the scientific evidence on which Parliament relied, if the instrument functions properly and all the relevant procedures are followed, the results should be reliable. It is therefore logical to provide that the results can be challenged only by raising problems that can be objectively identified and that relate to possible deficiencies in the instrument itself or in the procedure followed in operating it.

[ 39 ] In addition to establishing a rational connection, the appellants had to show that the measure violates the right to be presumed innocent as little as reasonably possible ( R. v. Edwards Books and Art Ltd. , 1986 12 (SCC) , [1986] 2 S.C.R. 713, at p. 772). In the minimal impairment inquiry, the court must not second‑guess Parliament and try to identify the least intrusive solution. In Downey , this Court stated that “Parliament is not required to choose the absolutely least intrusive alternative in order to satisfy this branch of the analysis. Rather the issue is ‘whether Parliament could reasonably have chosen an alternative means which would have achieved the identified objective as effectively’” (p. 37, quoting R. v. Chaulk , 1990 34 (SCC) , [1990] 3 S.C.R. 1303, at p. 1341). The latitude accorded to Parliament depends largely on the context. Hence, penal legislation that directly threatens a person’s liberty will be assessed differently than a complex regulatory response to a social problem ( Alberta v. Hutterian Brethren of Wilson Colony , 2009 SCC 37 , [2009] 2 S.C.R. 567, at paras. 35 and 37 ).

[ 40 ] The reliability of breathalyzer tests has been recognized in Hodgson’s report, as well as by expert witnesses across the country (for example, Brian Image, James Wigmore, Kerry Blake and Robert Langille), by several courts (the Alberta Provincial Court in R. v. Duff , 2010 ABPC 319 , 501 A.R. 122, the British Columbia Provincial Court in R. v. Gillespie , 2010 BCPC 207 () , the Ontario Court of Justice in Powichrowski and in R. v. Muzuva (2010), 206 C.R.R. (2d) 18 , and the Court of Québec in R. v. Cayer , 2010 QCCQ 9352 () ), and by the judge who heard the respondent’s case. Thus, the validity of Parliament’s chosen method is supported by scientific evidence that is accepted by the scientific and legal communities.

[ 41 ] It should also be mentioned that the new provisions do not make it impossible to disprove the test results. Rather, Parliament has recognized that the results will be reliable only if the instruments are operated and maintained properly, and that there might be deficiencies in the maintenance of the instruments or in the test process. What the new provisions require is that evidence tending to cast doubt on the reliability of the results relate directly to such deficiencies.

[ 42 ] Since the nature and scope of the evidence that might be considered relevant has not been argued on this appeal, it would not be appropriate to rule on the specific limits of that evidence. I will merely note that, in light of the evidence accepted by the trial judge, there are several pieces of evidence that can be provided to a person who is charged under s. 253(1) ( b ) Cr. C. , including the breathalyzer readings , the qualified technician’s certificate and the analyst’s certificate concerning the sample of the alcohol standard.

[ 43 ] In its recommendations, the CSFS Committee also suggested mechanisms for ensuring that the instruments function properly and for assuring the quality of breath alcohol analyses. It can be inferred from these recommendations that the instruments may not function optimally if the suggested procedures are not followed.

[ 44 ] The Barreau du Québec and the Association québécoise des avocats et avocates de la défense argue on the basis of Judge Lortie’s decision in R. v. Laforge , 2010 QCCQ 7718 , [2010] R.J.Q. 2537, that Parliament could have opted for a less intrusive statutory amendment. In Laforge , Judge Lortie expressed the view that Parliament could simply have allowed the accused to present a Carter defence, but authorized the trier of fact to assess the credibility of the accused in light of the breathalyzer test results (at para. 272):

[ translation ] Bill C‑2 was clearly meant to be a response to Boucher , in which it was held that breathalyzer test results could not be considered in assessing credibility. In this context, Parliament could have amended the legislation to authorize such an assessment. Thus, the trial judge would exercise his or her discretion and assess the consumption theory of the accused in light of the evidence as a whole. In other words, a St. Pierre amendment that is less intrusive and would be held to be valid by the courts.

[ 45 ] The scientific data presented at the time of the enactment of the new provisions show that Parliament intended to do more than simply adjust the wording that had been interpreted in Boucher . Apart from the theoretical difficulties involved in assessing the credibility of the accused on the basis of test results that are presumed to be accurate, returning to a Carter defence would make it impossible to meet Parliament’s objective. Absent statutory provisions to the effect that the results are to prevail, judges would still be faced with the problem the amendments were actually intended to solve. If the testimony of the accused concerning his or her consumption of alcohol were accepted, it could raise a reasonable doubt about the reliability of the test results despite the fact that it has now been shown that the success rate of this defence is hard to justify in light of the scientific reliability of the instruments. It was appropriate for Parliament to enact provisions that would spare the prosecution the burden of tendering evidence of scientific reliability in every case.

[ 46 ] I accordingly conclude that requiring evidence aimed at establishing that the instrument malfunctioned or was operated improperly satisfies the minimal impairment test.

[ 47 ] What remains to be determined is whether the advantages of this requirement outweigh its disadvantages. For this, it is necessary to examine the consequences of the measure. The limits that flow from the requirement have a significant effect on the defences available to the accused, as it is now more difficult to rebut the presumptions. The evidence to be adduced is more complex. The accused must retain a technician or an expert to determine whether the instrument malfunctioned or was operated improperly. It is impossible for a layperson to do this. However, it should be borne in mind that the Carter defence also required the accused to retain an expert.

[ 48 ] The prosecution gains a clear, albeit limited, advantage from the requirement, since evidence to the contrary is limited to the real issue: whether the test results are reliable. The evidence to be tendered relates directly to an instrument that is under the prosecution’s control. The prosecution must of course disclose certain information concerning the maintenance and operation of the instrument, but it is free to establish procedures for tracking how such instruments are maintained and operated. Moreover, the prosecution has control over the people who maintain and operate the instruments.

[ 49 ] At first glance, the advantages of limiting the evidence the accused can adduce in order to rebut the presumptions to evidence that the instrument malfunctioned or was operated improperly appear to outweigh the disadvantages of this measure. However, since this particular requirement forms part of a broader legislative scheme, I must consider the other requirements before concluding that it is justified.

4.2.2 Connection Between the Deficiency and the Determination of a Level Exceeding .08

[ 50 ] Section 258(1)( c ) Cr. C. requires evidence tending to show not only that the instrument was malfunctioning or was operated improperly, but also that the malfunction or improper operation resulted in a reading according to which the blood alcohol level of the accused exceeded .08. This requirement furthers Parliament’s objective of giving greater weight to the test results and thus passes the rational connection test.

[ 51 ] Whether the impairment resulting from this requirement is minimal is open to debate. The accused must prove that if his or her blood alcohol level exceeded the allowable maximum, it was because the instrument malfunctioned or was operated improperly. The burden of doing so seems at first glance to be quite heavy. No expert evidence was adduced to show how this connection can be proved. It is conceivable that evidence that an instrument has produced erratic results could raise a doubt that the results concerning an accused are reliable. However, it would be difficult for the accused to identify a specific malfunction and prove that it resulted in a reading according to which his or her blood alcohol level exceeded the legal limit. One can only speculate about the type of expert evidence the accused would need to produce for this purpose, but it would certainly have to be much more specific than the evidence needed to prove that the instrument was malfunctioning or was operated improperly.

[ 52 ] At this step in the defence process, it must be accepted that the judge will not consider evidence showing a connection between a deficiency and the determination that the blood alcohol level of the accused exceeded the legal limit unless the accused has already proved that the instrument was malfunctioning or was operated improperly. At this stage, if the arguments made by the defence are frivolous or trivial, they will not cast doubt on the proper functioning or operation of the instrument, and the defence must fail. The facts of Crosthwait provide a good illustration of this. In that case, the accused had tried to raise a doubt that the instrument had functioned properly by arguing that the technician had not compared the air temperature with the temperature of the solution before making the analyses. The mere possibility that the instrument had malfunctioned was not evidence to the contrary that could cast doubt on the reliability of the results.

[ 53 ] Thus, it is necessary to proceed on the basis that the accused must not simply show that a deficiency is possible, but raise a real doubt that the instrument was functioning or operated properly. In short, if Parliament’s objective was to eliminate frivolous cases, that objective would be achieved through the assessment of the evidence by the trier of fact. To enable the prosecution to benefit from the presumptions even though a real doubt has been raised about the results of breathalyzer tests amounts in practice to a reverse onus. Iacobucci J.’s comments in R. v. St. Pierre , 1995 135 (SCC) , [1995] 1 S.C.R. 791, at par. 56 , regarding the presumption of identity are very relevant here:

If this position is accepted, and the materiality of the evidence of the accused depends upon reference to the legal limit, a grey area exists between the breathalyzer result and the legal limit, and the burden of clarifying this will be placed on the accused when, in fact, the burden should rest with the Crown to prove its case.

[ 54 ] I note that in St. Pierre , the Court commented on the rules with respect to the evidence to the contrary that the accused had to adduce to rebut the presumption of accuracy and distinguished that evidence from the evidence needed to rebut the presumption of identity. But Iacobucci J.’s comments on the presumption of accuracy were based on the conclusions from Crosthwait , in which Pigeon J. had held that to rebut the presumption of accuracy, the accused had to raise a reasonable doubt “as to [his or her] blood alcohol content . . . being over the allowable maximum” (p. 1101). However, not only was it not argued in Crosthwait that the right to be presumed innocent had been infringed, but Pigeon J.’s comment was premised on the fact that “ any evidence tending to invalidate the result of the tests [could thus] be adduced on behalf of the accused in order to dispute the charge against him” (p. 1100 (emphasis added)). The context of the new provisions is completely different. To satisfy the second requirement and raise a doubt that his or her blood alcohol level exceeded the allowable maximum, an accused cannot rely on his or her consumption of alcohol ( s. 258(1) ( d.01 ) Cr. C. ).

[ 55 ] The presumption of accuracy was not at issue in St. Pierre . Nevertheless, I note that the legislative context of this presumption was the same in that case as in Crosthwait , that is, it could be rebutted by “evidence to the contrary” ( s. 25(1) of the Interpretation Act , R.S.C. 1985, c. I‑21, as applicable to s. 258(1)( g ) Cr. C. ; see Crosthwait , at pp. 1099‑1100) . Moreover, this Court has not had to consider the constitutionality of the presumption of accuracy until now. I therefore endorse, in the context of the new provisions on the presumption of accuracy, the following comment that Arbour J.A. had made in dissent in St. Pierre , and with which Iacobucci J. agreed (at para. 22):

To the extent that Crosthwait , supra , held that “evidence to the contrary” in s. 258(1) ( c ) means evidence tending to show that the accused’s blood alcohol content at the time of the offence was below the permissible limit, it should not be applied in a case such as the present one. [para. 21]

[ 56 ] Insofar as the majority in St. Pierre followed the approach adopted by Pigeon J. in Crosthwait with respect to the requirement that the accused show that his or her blood alcohol level did not exceed .08 in order to rebut the presumption of accuracy, their reasons must be reconsidered to take the constitutional argument into account. Although the requirement that the accused raise a doubt that his or her blood alcohol level in fact exceeded .08 could be justified when there were no limits on the evidence the defence could tender to cast doubt on the test results, it constitutes an excessive burden in the context of a statutory scheme under which the evidence must relate directly to the functioning or operation of the instrument.

[ 57 ] A consideration of the advantages and disadvantages of the second requirement of s. 258(1)( c ) reinforces the conclusion that this requirement is not justified. Requiring that a connection be established between the deficiency in the functioning or operation of the instrument and the determination that the blood alcohol level of the accused exceeded the legal limit increases the burden on the defence significantly without reducing the expense to the prosecution. In St. Pierre , L’Heureux‑Dubé J., dissenting, stressed that Parliament had established the presumption in s. 258(1) ( c ) Cr. C. “in clear recognition of the difficulty and expense of requiring expert evidence in virtually every alcohol‑related driving offence” (para. 90). The scheme that has existed since the statutory amendments came into force is designed to require the prosecution to adduce technical evidence to counter an attempt to rebut the presumption of accuracy or the first presumption of identity. An accused who produces evidence to rebut one of these presumptions will do so by calling an expert, and the prosecution will have to call a technician, and possibly an expert. As a result, being the party that has to prove that there is no connection after the accused has adduced evidence to show that the instrument malfunctioned or was operated improperly does not impose a significant additional burden on the prosecution.

[ 58 ] Moreover, it is important to note that, where the accused raises a reasonable doubt that the instrument functioned or was operated properly, this simply means that the prosecution loses the benefit of the presumptions under s. 258(1) ( c ). The prosecution can still tender additional evidence to prove that, despite the proven deficiency, the blood alcohol level of the accused exceeded .08 as shown by the test results.

[ 59 ] In these circumstances, having regard to Parliament’s objective of giving priority to the reliability of the test results, I conclude that requiring an accused to prove not only a malfunction or improper operation of the instrument that is serious enough to raise a reasonable doubt, but also a causal connection between that malfunction or improper operation and the determination that the blood alcohol level of the accused exceeded the legal limit, constitutes a serious infringement of the right to be presumed innocent. This infringement cannot be justified in a democratic society.

4.2.3 Evidence of a Blood Alcohol Level Not Exceeding .08

[ 60 ] Section 258(1)( c ) Cr. C. also requires the accused to adduce evidence tending to show that his or her blood alcohol level would not in fact have exceeded .08 at the time of the offence. Such evidence will generally be that of a Carter defence.

[ 61 ] In Oakes , this Court stressed that legislative measures adopted for the purpose of attaining an objective must be “carefully designed to achieve the objective in question” (p. 139). If Parliament’s objective were simply to ease the burden the prosecution must discharge in combatting the problems associated with drinking and driving, the rational connection would be clear, since this provision requires the accused to raise an additional doubt about his or her guilt.

[ 62 ] As the legislative facts show, however, Parliament’s objective is not stated in such general terms that it can encompass all measures taken to combat drinking and driving. Moreover, care must be taken not to state the objective too broadly: RJR‑MacDonald Inc. v. Canada (Attorney General) , 1995 64 (SCC) , [1995] 3 S.C.R. 199, at para. 144 . As I mentioned above, the objective was to give the test results a weight consistent with their scientific value. In this context, I do not see a rational connection between this objective and the requirement of adducing evidence to raise a doubt that the blood alcohol level of the accused in fact exceeded .08. Such evidence is not aimed directly either at the process of taking samples using authorized instruments or at the test results.

[ 63 ] The inquiry into whether the impairment of the right to be presumed innocent is minimal confirms that the requirement of evidence that the blood alcohol level of the accused was under the legal limit is not justified. I reiterate that this requirement is in addition to the requirement of showing that the instrument malfunctioned or was operated improperly. If the accused has already identified a defect that could cast doubt on the reliability of the results, it is difficult to justify requiring the court to nevertheless accept that the results have probative value if the accused has produced no evidence regarding his or her blood alcohol level. This amounts to saying that, where a court has a doubt about an essential element of the offence, it must nevertheless convict unless the accused can present evidence tending to show that he or she is innocent. I accordingly find that the third requirement of s. 258(1) ( c ) cannot be justified under s. 1 of the Charter .

4.2.4 Conclusion Concerning the First Requirement

[ 64 ] In light of my conclusion that the second and third requirements are not justified, the first requirement is the only one that can limit the evidence an accused may tender to cast doubt on the test results. What therefore remains to be determined is whether the effects of this limit on the right to be presumed innocent are proportional to Parliament’s objective. In the proportionality analysis required by s. 1 , it is important to consider the impugned provision in the context of the entire legislative scheme of which it forms a part.

[ 65 ] In Downey , Cory J. explained that in determining whether an infringement is proportional to the legislative objective, it is necessary to balance societal and individual interests (p. 38). The objective of the first requirement of s . 258(1) ( c ) Cr. C. , as clarified by s. 258(1)( d.01 ), is to confirm the scientific value and ensure the primacy of breathalyzer test results. The purpose of this statutory amendment was to remedy a situation that was common before it came into force: test results could be rejected on the basis of testimony that was considered subjective. The amendment was a response to the serious disconnect that existed in the fact that the Carter defence had a high success rate despite the recognized scientific reliability of the results.

[ 66 ] Moreover, the scheme adopted for breathalyzer tests includes certain guarantees that place limits on police action and protect the presumption of innocence. Under s. 254(2) Cr. C. , for example, a peace officer may not require a person who has operated or had the care or control of a vehicle to perform physical co‑ordination tests unless the officer has reasonable grounds to suspect that the person has consumed alcohol. Under s. 254(3) Cr. C. , an officer may not require a person to submit to a breathalyzer test unless the officer has reasonable grounds to believe that the person is committing, or has at any time within the preceding three hours committed, the offence of driving or having care of a vehicle with a blood alcohol level exceeding .08. The breath samples must be analyzed by a qualified technician using an approved instrument. The samples must be taken not later than two hours after the time when the offence is alleged to have been committed, and with an interval of at least 15 minutes between the times when they are taken ( s. 258(1) ( c )(ii) Cr. C. ).

[ 67 ] From this perspective, if the second and third requirements provided for in s. 258(1) ( c ) Cr. C. are severed, I consider Parliament’s response to be a measured one. In light of the objective of this provision, the scientific evidence in the record and the guarantees that form part of the evidential blood alcohol analysis scheme, I find that the limit on defences that is established in s. 258(1)( c ) and s. 258(1)( d.01 ) Cr. C. is a justified infringement of the right to be presumed innocent. As a result of this conclusion, my consideration of the other arguments against finding s . 258(1) ( c ) and s. 258(1) ( d.01 ) to be constitutional will be limited to this requirement.

5. Compatibility of Section 258(1)( c ) and Section 258(1) ( d.01 ) with the Right to Make Full Answer and Defence (Section 7 of the Charter )

[ 68 ] The first argument based on s. 7 of the Charter is that the combined effect of s. 258(1)( c ) and s. 258(1)( d.01 ) Cr. C. is to limit the right to make full answer and defence.

[ 69 ] In enacting the impugned provisions, Parliament excluded evidence of the alcohol consumption of the accused insofar as it is adduced to show that the instrument was malfunctioning or was operated improperly ( s. 258(1) ( d.01 )). Moreover, the presumption of accuracy of the results and that of identity can be rebutted only by casting doubt on the proper functioning or operation of the instrument. In other words, evidence of the alcohol consumption of the accused is not legally sufficient to raise a doubt under s. 258(1) ( c ) as to the reliability of the results. Because I have already found that the second and third requirements of s. 258(1) ( c ) are unconstitutional, evidence of the effect of the consumption of the accused on his or her blood alcohol level will be admissible only if the accused chooses to tender it or if the presumptions of s. 258(1) ( c ) are inapplicable (where, for example, the samples were taken too late) or have been rebutted.

[ 70 ] Since Carter , Canadian courts have accepted that there is a logical connection between evidence of alcohol consumption and the accuracy of breathalyzer test results. I agree that such evidence can logically tend to discredit both the results of a breathalyzer test and — indirectly — the proper functioning and operation of the instrument itself.

[ 71 ] However, the fact that evidence is relevant does not necessarily make it admissible. This Court has recognized that relevant evidence can be excluded if its exclusion is justified by a ground of law or policy ( R. v. Seaboyer , 1991 76 (SCC) , [1991] 2 S.C.R. 577, at p. 609). Professors Cross and Tapper quote Wigmore’s comment that “[a]dmissibility signifies that the particular fact is relevant and something more, — that it has also satisfied all the auxiliary tests and extrinsic policies” (R. Cross and C. Tapper, Cross on Evidence (7th ed. 1990), at p. 60; see Seaboyer , at p. 692, per L’Heureux‑Dubé J.).

[ 72 ] In the context of the case at bar, as I mentioned above, the expert evidence accepted by the courts over the past few years has established that breathalyzer tests are very reliable, provided that the instruments are operated and maintained properly. At the same time, many reports have shown the testimony of accused persons regarding their alcohol consumption to be unreliable (see J. G. Wigmore, “Man vs. Machine: Self‑Reported Alcohol Consumption of Drinking Drivers vs. Evidential Breath Alcohol Tests. Is the Restriction of Evidence to the Contrary Scientifically Valid?” (2009), 54 Crim. L.Q. 395; T. L. Martin, J. G. Wigmore and K. L. Woodall, “A Comparison of Blood Alcohol Concentrations Estimated From Drinking Histories of Drivers Charged with ‘Over 80’ and Their Intoxilyzer® 5000C Results” (2004), 37 Can. Soc. Forensic Sci. J. 187; M. S. Sommers et al. , “‘Nurse, I Only Had a Couple of Beers’: Validity of Self‑Reported Drinking Before Serious Vehicular Injury” (2002), 11 Am. J. Critical Care 106).

[ 73 ] Moreover, a study commissioned by the federal Department of Transport and presentations made to the Standing Senate Committee on Legal and Constitutional Affairs show that the Carter defence resulted in a high rate of acquittal ( Proceedings of the Standing Senate Committee on Legal and Constitutional Affairs , No. 9, 2nd Sess., 39th Parl., February 21, 2008, at p. 37; R. Robertson, W. Vanlaar and H. Simpson, National Survey of Crown Prosecutors and Defence Counsel on Impaired Driving: Final Report (July 2008), at p. 72 ). I infer from these documents a criticism to the effect that a high rate of acquittal is not justified in light of the probative value of the test results. Although certain aspects of this evidence cannot be considered scientific, it cannot be disregarded.

[ 74 ] This evidence shows both that the probative value of the testimony of the accused regarding his or her alcohol consumption is debatable and that to admit such testimony for the purpose of determining whether the test results are valid entails a real risk of perverting the fact‑finding process. These studies and presentations were not available to the courts when the Carter defence became established in our law. The law had to be updated, and that is what the government endeavoured to do in amending the legislation. The courts must take this into consideration. Requiring an accused to tender evidence related directly to the functioning or operation of the instruments used to take samples contributes to truth‑finding and favours the integrity of the trial. In my opinion, the exclusion of evidence of the alcohol consumption of the accused that is set out in s. 258(1) ( d.01 ) is based on a policy consideration that is consistent with Seaboyer .

[ 75 ] In Seaboyer , the majority of this Court held that the effect of s. 276 Cr. C. was to exclude evidence that might be “of great importance to getting at the truth and determining whether the accused is guilty or innocent under the law” (p. 616). Thus, even though the purpose of s. 276 was to help judges and juries arrive at just verdicts, its effect in practice could be the opposite:

Accepting that the rejection of relevant evidence may sometimes be justified for policy reasons, the fact remains that s. 276 may operate to exclude evidence where the very policy which imbues the section — finding the truth and arriving at the correct verdict — suggests the evidence should be received. [p. 620]

[ 76 ] The same cannot be said in the instant case, however. Unlike in Seaboyer , the statutory provisions at issue here do not exclude evidence that might be necessary for a judge or jury to arrive at a just verdict. There is no reasonable risk that these provisions frustrate the objective of truth‑finding. In my opinion, it was open to Parliament to exclude the production of evidence of the alcohol consumption of the accused that tends to show that the instrument was malfunctioning or was operated improperly ( s. 258(1) ( d.01 )), and to provide that such evidence is legally insufficient to cast doubt on the reliability of the test results. This exclusion does not infringe the rights protected by s. 7.

[ 77 ] The second argument based on s. 7 of the Charter is that the new provisions create a defence that is so difficult to attain as to be practically illusory. In R. v. Morgentaler , 1988 90 (SCC) , [1988] 1 S.C.R. 30, Dickson C.J. commented on the principle on the basis of which such an argument could be made (at p. 70):

One of the basic tenets of our system of criminal justice is that when Parliament creates a defence to a criminal charge, the defence should not be illusory or so difficult to attain as to be practically illusory.

[ 78 ] Although Parliament now requires evidence tending to establish a deficiency in the functioning or operation of the instrument, this does not mean that there are limits on the evidence that can reasonably be used by the accused to raise a doubt in this regard. The accused can request the disclosure of any relevant evidence that is reasonably available in order to be able to present a real defence. If the prosecution denies such a request, the accused can invoke the rules on non‑disclosure and the available remedies for non‑disclosure (see R. v. O’Connor , 1995 51 (SCC) , [1995] 4 S.C.R. 411). In short, the accused might rely, for example, on a maintenance log that shows that the instrument was not maintained properly or on admissions by the technician that there had been erratic results, or he or she might argue that health problems had affected the functioning of the instrument (see R. v. Kasim , 2011 ABCA 336 , 515 A.R. 254).

[ 79 ] It should be noted that the defence created by Parliament is not illusory simply because accused persons will rarely succeed in raising a reasonable doubt that the instrument was functioning or was operated properly. The existence of a defence must not be confused with how often those presenting it are successful. As Judge Duncan noted in Powichrowski (at para. 69 ):

While it may be that the defendant, having explored every avenue, will be unable to meet the requirements of the section and rebut the presumption, that is what often happens when a defendant is faced with credible and reliable evidence against him.

[ 80 ] Finally, it should be mentioned that the conclusion that the first requirement of s. 258(1) ( c ) does not violate s. 7 is in no way incompatible with a finding that it constitutes a justifiable infringement of the right to be presumed innocent. In light of the scientific evidence in the record, as we have seen, Parliament was justified in requiring evidence directly related to the operation or functioning of breathalyzers. It was therefore open to Parliament, without violating s. 7 or s. 11( d ) of the Charter , to exclude the possibility that a Carter defence would suffice on its own to cast doubt on the test results.

[ 81 ] For these reasons, I find that the rebuttal of the presumptions established in s. 258(1) ( c ), as clarified by s. 258(1) ( d.01 ), has not been rendered illusory by Parliament and that this provision does not infringe the right to make full answer and defence.

6. Compatibility of Section 258(1)( d.1 ) with the Right to Be Presumed Innocent (Section 11( d ) of the Charter )

[ 82 ] To facilitate a discussion on the amendments that affect the second presumption of identity, this presumption of identity must be clearly distinguished from the first one. The first presumption of identity, which is established in s. 258(1) ( c ), allows the prosecution to use the test results as a substituted fact in order to prove an essential element of the offence, namely the blood alcohol level of the accused at the time when the offence was alleged to have been committed. If the accused argues that the results do not correspond to his or her blood alcohol level at the time of the offence on the basis that the instrument was malfunctioning or was operated improperly, the evidence the accused can adduce will be limited by s. 258(1) ( c ) and s. 258(1) ( d.01 ) Cr. C.

[ 83 ] If the accused does not challenge the functioning of the instrument, but instead argues that, contrary to the results, his or her blood alcohol level at the time when the offence was alleged to have been committed did not exceed .08 and that the results were distorted by the fact that he or she had consumed alcohol shortly before or after the alleged offence, the evidence the accused can tender to rebut the presumption is circumscribed by s. 258(1) ( d.1 ) Cr. C. The type of consumption that can be used to rebut the second presumption was explained as follows by Rob Moore during the debate that preceded the enactment of Bill C‑32 ( House of Commons Debates , at p. 6186):

This could happen if, for example, the person downed several drinks and was arrested before the alcohol was absorbed. It could also occur that after driving, but before testing, the person consumed alcohol and it was absorbed by the time the approved instrument test was taken.

[ 84 ] As a result of the statutory amendments, evidence to the contrary adduced by the accused must tend to show two facts: (1) the consumption of alcohol of the accused was consistent with a blood alcohol level that did not exceed .08 at the time when the offence was alleged to have been committed; and (2) the consumption of alcohol of the accused was consistent with the test results.

6.1 Does Section 258(1)(d.1) Infringe the Protected Right?

[ 85 ] As I mentioned in my analysis with respect to s. 258(1) ( c ) Cr. C. , a statutory presumption violates the right to be presumed innocent if a judge can convict an accused even though there is a doubt that the accused is actually guilty. For instance, where breathalyzer test results according to which the blood alcohol level of the accused exceeded .08 are tendered in evidence, the trier of fact could have a reasonable doubt that the blood alcohol level of the accused was the same at the time of the alleged offence, because the accused could have drunk shortly before or after being pulled over. Since s. 258(1)( d.1 ) Cr. C. exempts the prosecution from having to establish the guilt of the accused beyond a reasonable doubt before the accused must respond, I must conclude that it infringes the right to be presumed innocent.

6.2 Is the Infringement Justified?

[ 86 ] The evidence that can be adduced to rebut the presumption of identity established in s. 258(1)( d.1 ) Cr. C. was clarified in two stages. The first requirement was incorporated into the Criminal Code in 1997 (S.C. 1997, c. 18, s. 10(2)). Under the provision that established it, the accused had to adduce evidence tending to show that his or her blood alcohol level at the time when the offence was alleged to have been committed did not exceed .08. This requirement is now found in the first subparagraph of s. 258(1)( d.1 ) Cr. C. Another requirement was added at the same time as the amendments to s. 258(1)( c ) Cr. C. discussed above. According to this new requirement, the evidence tendered by the accused must be consistent with the test results.

[ 87 ] No specific evidence has been introduced concerning the objective of s. 258(1)( d.1 )(i) Cr. C. , but that objective can be inferred from the legislative history. The amendment was passed shortly after — and most likely in response to — this Court’s decision in St. Pierre , so the comments made in that decision are helpful. The majority in St. Pierre held that the presumption of identity could be rebutted by any evidence showing a difference between the blood alcohol level of the accused at the time of the alleged offence and his or her blood alcohol level at the time of the analyses. L’Heureux‑Dubé J., dissenting, was concerned that the presumption would be rebutted “in every case where the accused invokes either the ‘last drink’ defence or the ‘post‑driving drinking’ defence, where there is not even an iota of proof to suggest that the discrepancy occasioned by the alcohol consumption would be of any legal relevance to conviction or acquittal on a charge of ‘over 80’” (para. 90). Parliament seems to have shared this concern and to have tried to ensure that the prosecution would not be required to have recourse to experts to explain the rate of absorption between the time of the offence and that of the analyses. I have no difficulty finding that this was a pressing and substantial objective.

[ 88 ] The requirement established in s. 258(1)( d.1 )(i) must be considered in conjunction with the 2008 amendment, which provides that the accused must also show that his or her consumption was consistent with the test results. This change to the presumption of identity, which was made at the same time as the amendments to s . 258(1)( c ) Cr. C. , was motivated by the same objective as those amendments. Parliament intended to give the test results a probative value consistent with their scientific reliability. I accepted above that this was a pressing and substantial objective, and the same conclusion applies here.

[ 89 ] A rational connection can easily be established between each of the requirements set out in s. 258(1)( d.1 ) and the requirement’s legislative objective. In the first case, the measure is linked to Parliament’s wish to ensure that the prosecution does not have to have recourse to experts to prove the blood alcohol level of the accused at the time of the alleged offence where the difference is not significant. In the second case, the measure is linked to Parliament’s wish to confirm the scientific value of the test results and to establish an explicit correlation between those results and a variation in blood alcohol level due to consumption of alcohol by the accused shortly before or after the time of the alleged offence.

[ 90 ] In my opinion, s. 258(1)( d.1 ) Cr. C. also satisfies the minimal impairment test. Whereas requiring the accused to show a connection between a malfunction of the instrument and the determination that his or her blood alcohol level exceeded .08 imposes an undue burden on the accused, the same is not true of requiring the accused to show that his or her consumption of alcohol shortly before or after the alleged offence was consistent with a blood alcohol level that did not exceed .08 at the time of the alleged offence. In such situations, the accused does not challenge the test results, but invokes his or her own unusual behaviour. It is the accused — and not the prosecution — who knows when he or she drank, and how much. What is more, it is also the accused — and not the prosecution — who would decide to analyze his or her capacity to absorb and eliminate alcohol, and to adduce evidence in this regard. I do not therefore consider it unduly onerous to require the person who has this information and is in a position to tender relevant evidence to show not only that he or she had a “last drink”, or drank after being pulled over, but also that the difference resulting from that consumption is relevant to the determination of his or her guilt or innocence. I should also note that the cases in which such a defence is raised should be rare, and that such a case would denote either significant irresponsibility with regard to public safety or a pathological reaction by the accused. As L’Heureux‑Dubé J. pointed out in St. Pierre (at para. 106 ),

[i]n most cases, moreover, there is good reason to suspect that post‑driving drinking (or just the claim thereof) is an act of mischief intended to thwart police investigators. All such cases, at the very least, involve a significant degree of irresponsibility and a cavalier disregard for the safety of others and the integrity of the judicial system. This Court should not encourage or, at the very least, lend legitimacy, to such behaviour.

[ 91 ] As for the second requirement of s. 258(1)( d.1 ) Cr. C. , it also infringes the right to be presumed innocent as little as reasonably possible. Where the reliability of the test results is not in dispute, requiring that evidence to the contrary be consistent with those results means that the defences raised by the accused must be consistent with one another. If the results would not have shown a blood alcohol level over .08 without the “last drink” or “post‑driving drinking”, then the consumption of alcohol by the accused will necessarily be consistent with the test results. If the defence relates to the reliability of the results, then the accused must impugn the test process itself under s. 258(1)( c ) Cr. C. and show that the instrument was malfunctioning or was operated improperly. An accused cannot rely on s. 258(1)( d.1 ) Cr. C. to rebut the presumption of accuracy or the first presumption of identity by raising a Carter defence — that would undermine the integrity of the entire legislative scheme.

[ 92 ] Section 258(1)( d.1 ) Cr. C. strikes a fair balance between collective rights and individual rights, and is part of a broader legislative scheme designed to confirm the primacy of breathalyzer test results. It is a justified infringement of the right to be presumed innocent.

7. Compatibility of Section 258(1)( d.1 ) with the Protection Against Self‑Incrimination (Section 11( c ) of the Charter )

[ 93 ] The protection against self‑incrimination does not apply in every case in which the accused risks being found guilty if he or she does not present a defence. In R. v. Darrach , 2000 SCC 46 , [2000] 2 S.C.R. 443, this Court noted that “[t]here is an important difference between a burden of proof with regard to an offence or an evidentiary burden, and the . . . need to respond when the Crown establishes a prima facie case, in order to raise a reasonable doubt about it” (para. 50). The need for the accused to testify to raise a doubt after the prosecution has produced evidence — where, for example, he or she wishes to rely on an alibi defence — results from a decision over which the Crown has no control. The prosecution is responsible neither for the choice of the accused nor for the consequences of that choice. The decision to testify in such circumstances is not incompatible with the protection against self‑incrimination.

[ 94 ] Section 258(1)( d.1 ) places an evidentiary burden on the accused. Since the effect of the presumption is that the prosecution need not prove that the consumption pattern of the accused is irrelevant to the reliability of the test results, the onus is on the accused to prove its relevance in order to ensure that those results do not stand as proof of his or her blood alcohol level at the time of the offence. Since the burden results from a statutory presumption and is not based on proven facts, it might at first glance be inferred from Darrach that the evidentiary burden is incompatible with the protection against self‑incrimination. But this inference does not withstand scrutiny. The presumption is based on the usual behaviour of drivers, who do not generally drink a sufficient quantity of alcohol to alter the results either just before or just after being pulled over by the police. It is in fact the exceptional behaviour of the accused, not the statutory presumption in the prosecution’s favour under s. 258(1)( d.1 ), that makes it necessary for the accused to testify.

[ 95 ] This conclusion can best be understood by considering the situation that arises in a case in which the prosecution cannot rely on the presumption of identity established in s. 258(1)( d.1 ), where, for example, more than two hours passed between the time of the alleged offence and that of the test. In such a case, an expert will take the blood alcohol level of the accused at the time of the test and use it to try to calculate retroactively what that level would have been at the time when the accused was pulled over. To do this, the expert must make certain factual assumptions, for example, that the accused did not consume a large quantity of alcohol within approximately one half hour before the alleged offence (in other words, that a portion of the alcohol the accused consumed had already been absorbed when he or she was pulled over), or between the time when he or she was pulled over and that of the test. If nothing in the evidence makes it possible to cast doubt on the expert’s assumptions, the court may make a deduction, based on common sense, that a person will not generally ingest large quantities of alcohol immediately before driving or while driving, or after being pulled over by the police ( R. v. Paszczenko , 2010 ONCA 615 , 103 O.R. (3d) 424; R. v. Grosse (1996), 1996 6643 (ON CA) , 29 O.R. (3d) 785 (C.A.) ; R. v. Hall , 2007 ONCA 8 , 83 O.R. (3d) 641 ; R. v. Bulman , 2007 ONCA 169 , 221 O.A.C. 210 ) .

[ 96 ] In sum, even without the presumption of identity, the accused might be required to raise a doubt about his or her unusual alcohol consumption if nothing in the evidence indicates that the expert’s assumptions are erroneous. It therefore seems artificial to say that requiring the accused under s. 258(1)( d.1 ) to testify about his or her alcohol consumption imposes an evidentiary burden on the accused. The choice by the accused to testify in this regard flows from a decision that must be made whenever the Crown’s evidence is sufficient to support a conviction. Thus, s. 11( c ) of the Charter is not infringed.

8. Application to the Facts of This Case

[ 97 ] The respondent was charged under s. 253(1)( b ) Cr. C. with operating a vehicle with a blood alcohol level over the legal limit. A qualified technician took three breath samples from her using an Intoxilyzer 5000C instrument . The analyses showed blood alcohol levels of 164 mg, 124 mg and 130 mg in 100 ml of blood. Before Judge Chapdelaine, the respondent challenged the application of the presumption of accuracy on the basis of the differences in the test results. She argued that the technician should not have taken the result of the second sample into account if he did not think it reflected her actual blood alcohol level. In the respondent’s view, because of the difference of more than 20 mg between the third analysis and the first, the technician should have taken a fourth sample. She contended that the prosecution could not benefit from the presumption of accuracy in these circumstances. The respondent also argued that the new provisions on breathalyzer test results are unconstitutional.

[ 98 ] Judge Chapdelaine found that the qualified technician’s testimony was sufficient to explain the differences in the results of the three analyses, and that the technician’s certificate was proof of its content. The technician had taken a third breath sample from the respondent because of the difference of more than 20 mg between the first two results. He testified that the respondent had not blown hard enough when the last two samples were taken (three seconds for the second sample and four seconds for the third), mainly because she was crying. In his opinion, however, the instrument had functioned effectively, and the last two samples were valid even though the results obtained from them were below the respondent’s actual blood alcohol level.

[ 99 ] Regarding the constitutional challenge, Judge Chapdelaine expressed the opinion that the statutory amendments did not bar the respondent from presenting a Carter defence to rebut the presumption of accuracy. He therefore assessed the probative value of her testimony concerning her consumption of alcohol. On the basis of what she said she had consumed, her blood alcohol level at the time she was pulled over would have been 58 mg in 100 ml of blood.

[ 100 ] In light of the evidence, Judge Chapdelaine concluded that the respondent’s testimony about her alcohol consumption was not sufficiently serious or probative to raise a reasonable doubt. Finding that the qualified technician’s explanations were sufficient and that the presumptions established in s . 258(1)( c ) and s. 258(1)( d.1 ) Cr. C. applied, he accordingly convicted the respondent of operating a vehicle with a blood alcohol level over the legal limit. In short, Judge Chapdelaine erred in holding that the respondent could rebut the presumption of accuracy of s. 258(1)( c ) Cr. C. by presenting a Carter defence, but that error did not affect his conclusion, since, when all is said and done, he did not believe the respondent. The conviction is therefore upheld.

[ 101 ] For these reasons, I would allow the appeal in part and answer the constitutional questions as follows:

1. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C‑46, infringe s. 7 of the Canadian Charter of Rights and Freedoms ?

Answer: No

2. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: It is not necessary to answer this question.

3. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C‑46, infringe s. 11(c) of the Canadian Charter of Rights and Freedoms ?

Answer: No

4. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: It is not necessary to answer this question.

5. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C‑46, infringe s. 11( d ) of the Canadian Charter of Rights and Freedoms ?

Answer: Yes

6. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: Sections 258(1)( d.01 ) and 258(1)( d.1 ), and s. 258(1)( c ) after severance of the words “all of the following three things —” and “, that the malfunction or improper operation resulted in the determination that the concentration of alcohol in the accused’s blood exceeded 80 mg of alcohol in 100 mL of blood, and that the concentration of alcohol in the accused’s blood would not in fact have exceeded 80 mg of alcohol in 100 mL of blood at the time when the offence was alleged to have been committed”, are justified under s. 1 of the Charter .

The reasons of Rothstein and Cromwell JJ. were delivered by

Cromwell J. (dissenting in part) —

I. Introduction

[ 102 ] I have had the advantage of reading the reasons of my colleague Deschamps J. I agree with her that the conviction should be upheld. With respect to the constitutional issues, I also agree that s. 258(1) ( c ) and s. 258(1)( d.01 ) of the Criminal Code , R.S.C. 1985, c. C-46 (“ Cr. C. ”), do not violate s. 7 of the Canadian Charter of Rights and Freedoms and that s. 258(1)( d.1 ) Cr. C. does not violate s. 11( c ) of the Charter .

[ 103 ] However, I respectfully am not persuaded that ss. 258(1) ( c ), 258(1) ( d.01 ) or 258(1) ( d.1 ) limit the right to be presumed innocent as guaranteed by s. 11( d ) of the Charter . Even assuming, without deciding, that they do, I conclude that any limitation is reasonable and demonstrably justified in a free and democratic society. I would answer the relevant constitutional questions in the negative and allow the appeal.

[ 104 ] In my respectful view, the record before us shows that the results of a breathalyzer analysis, performed in accordance with the statutory requirements, are sufficiently reliable that it would not be reasonable to doubt that a reading exceeding 80 mg of alcohol in 100 ml of blood (“.08”) accurately reflects a blood alcohol level that is over .08 at the time of testing and at the time of the alleged offence, absent evidence raising a realistic concern about the proper functioning or operation of the device. That in my view is the effect of the provisions and this meets the requirements of s. 11 ( d ).

[ 105 ] If I am wrong in that conclusion, I agree with Deschamps J. to the extent that she finds that aspects of the scheme are justified limitations, but respectfully disagree that some other aspects are not.

II. The Statutory Scheme and the Challenged Provisions

A. Nature of the Statutory Scheme and the Challenged Provisions

[ 106 ] The provisions in issue are interrelated and for the purposes of the constitutional analysis, they must be examined in light of their overall effect. Before turning to the provisions in detail, therefore, it will be helpful to describe the statutory scheme and the place of the challenged provisions in it.

[ 107 ] To begin, I underline that the offence to which these provisions relate is operating, or having care or control of, a motor vehicle (or vessel or operating or assisting in the operation of an aircraft or of railway equipment) “having consumed alcohol in such a quantity that the concentration in the person’s blood exceeds eighty milligrams of alcohol in one hundred millilitres of blood”: s. 253(1)( b ). It is not necessary for the Crown to prove any particular blood alcohol concentration (“BAC”). Rather, the accused’s guilt or innocence will depend on proof that his or her BAC exceeded the legal limit (.08) at the time of the alleged offence (hereafter “driving”).

[ 108 ] Evidence of this element of the offence will most often come from the results of an analysis of breath samples provided by the accused in compliance with a demand by a peace officer. The Criminal Code sets up a detailed scheme governing how these samples may be lawfully obtained, how they may be analyzed, how the results may be admitted into evidence and how the results may be challenged at trial. The provisions in issue in this appeal fall into this last category and it is useful to place them in the broader context of the breath analysis scheme.

[ 109 ] In order to make a lawful demand for a breath sample, a peace officer must have reasonable grounds to believe that a person is committing or has, within the preceding three hours, committed an offence under s. 253: s. 254(3). If several requirements are met, the prosecution will have the benefit of the presumptions facilitating proof of the accuracy of the results and of the fact that the results correspond to the accused’s BAC at the time of driving. These requirements include time limits for taking the breath samples, the use of approved containers for taking the samples, the use of approved instruments for analyzing them and the operation of the instrument by a qualified technician.

[ 110 ] The challenged provisions which form part of this complex statutory scheme are based on three quite straight forward ideas. These ideas are that if all of the statutory requirements for taking and analyzing samples are observed: (1) the breathalyzer results are reliable in the absence of some basis in the evidence to doubt them; (2) the estimated BAC arrived at by consumption and elimination evidence (so-called Carter evidence; see R. v. Carter (1985), 1985 168 (ON CA) , 19 C.C.C. (3d) 174 (Ont. C.A.)), is not sufficiently reliable to be used to challenge the accuracy of breathalyzer results; and (3) the BAC at the time of testing will not be higher than at the time of driving, unless the accused drank a large quantity of alcohol shortly before driving or consumed alcohol between driving and testing. It will be helpful to say a few words about each of these ideas.

1. Accuracy of Results

[ 111 ] The first idea is that, provided the analysis of breath samples is made in accordance with the statutory requirements, its results should be accepted as accurately reflecting the accused’s BAC at the time of testing unless there is some reason to question the proper functioning or operation of the device. This idea is based on the demonstrated accuracy of approved instruments when used properly by qualified technicians. The challenged provisions require this inference of accuracy to be drawn unless there is evidence capable of giving rise to a doubt about whether the approved device, as a result of improper functioning or use, generated a reading that wrongly showed the accused’s BAC as exceeding the legal limit.

[ 112 ] More specifically, s. 258(1) ( c ) Cr. C. provides that evidence of the results of the analysis is conclusive proof of the accused’s blood alcohol limit at the time of testing. There is, therefore, now clearly a presumption of accuracy in this provision. (This is subject to the analysis meeting the requirements as to the taking of each sample and analysis of it by means of an approved instrument operated by a qualified technician. I should add that if the two samples result in different levels, the lower is deemed to be the relevant blood alcohol level. I should also add that the provision deems the results of the analysis to reflect the accused’s BAC at the time of driving as well. I will return to this aspect of the provision in a moment.)

[ 113 ] To avoid the operation of the presumption that the results of the analysis are accurate, the accused must point to evidence raising a reasonable doubt about each of the following three matters: (1) that the approved instrument was malfunctioning or was operated improperly; (2) the malfunction or improper operation resulted in the determination that the accused’s BAC was over .08; and (3) the accused’s BAC was not over .08 at the time of driving. In other words, to avoid the inference that an approved device operated by a qualified technician yielded an accurate result, the accused must raise a doubt that, but for improper functioning or operation of the device, the reading would have been within the legal limit.

[ 114 ] My colleague Deschamps J. is of the view that this limits the right to be presumed innocent guaranteed under s. 11( d ) of the Charter and therefore must be justified under s. 1 to survive Charter scrutiny. For reasons I will set out in my analysis, my view is that it does not.

2. Consumption/Elimination Evidence

[ 115 ] The second idea is that Carter evidence should not be admitted to challenge the proper functioning or operation of the device. Carter evidence seeks to establish a BAC by applying estimates of absorption and elimination of alcohol in the blood on the basis of how much alcohol the accused consumed. The purpose of one of the challenged provisions, s. 258(1) ( d.01 ), is to exclude this sort of evidence if it is directed to showing that the device malfunctioned, was operated improperly or that the analysis was improperly performed.

[ 116 ] My colleague concludes that this provision, in combination with s. 258(1) ( c ), limits the right to be presumed innocent, but that it does not infringe the right to make full answer and defence. While I agree that it does not limit the right to make full answer and defence, my view is that it does not limit the right to be presumed innocent either.

3. BAC at Time of Testing and Driving

[ 117 ] The third idea underpinning the scheme and the challenged provisions is that the accused’s BAC at the time of driving will generally not have been lower than at the time of testing unless the accused drank a large amount of alcohol shortly before driving (this is often referred to as “bolus drinking”) or consumed alcohol between driving and testing (I will refer to this as intervening drinking). This idea is addressed in two places in the challenged provisions and it is important to take both into account in the constitutional analysis. In s. 258(1)( c ), there is the first “presumption of identity”. It stipulates that the BAC at the time of testing (or the lowest of the results of multiple tests) is conclusive proof that the BAC at the time of the offence was the same. (This of course is subject to the tests being administered in accordance with the statutory scheme, and to there being no reasonable doubt that the readings were over the limit as a result of malfunction or improper operation or that the accused’s BAC at the time of driving was in fact under .08.) The second presumption of identity is found in s. 258(1)( d.1 ). It provides that, if the results of analyses show a BAC over .08, that is proof that the BAC at the time of the alleged offence was over .08, in the absence of evidence tending to show that the accused’s consumption of alcohol was consistent with both a BAC over .08 at the time of testing and under .08 at the time of the offence. This provision addresses the possibility of bolus and intervening drinking.

[ 118 ] As noted earlier, Carter evidence is not admissible to show that the device malfunctioned or was operated improperly (s. 258(1)( d.01 )). Thus, such evidence cannot be used to rebut the first presumption of identity by attacking these aspects of the accuracy of the test results. However, and I think this is an important qualification of the exclusion of evidence, s. 258(1)( d.1 ) permits Carter evidence to rebut this second presumption of identity, provided that the evidence is consistent with both the readings and innocence. In other words, the admission of this evidence for the purpose of rebutting this presumption is premised on the accuracy of the readings. While the statute is not as clear as it might be, I understand that these two presumptions of identity are not intended to conflict. In other words, if an accused raises a reasonable doubt by means of Carter evidence under s. 258(1)( d.1 ), the presumption of identity in s. 258(1)( c ) does not continue to operate.

[ 119 ] My colleague finds that these provisions limit the presumption of innocence and I respectfully am of the view that they do not.

4. Section 1 Justification

[ 120 ] My colleague concludes that limiting the challenge to the accuracy of the readings to evidence raising a doubt about whether the instrument was malfunctioning or was operated improperly (the first requirement of s. 258(1)( c )) is a justified limitation of the right to be presumed innocent, as are the limitations on the admissibility of Carter evidence as set out in ss. 258(1)( d.01 ) and 258(1)( d.1 ). However, my colleague concludes that the other two requirements in s. 258(1)( c ), that is the burden on the accused to raise a doubt that the malfunctioning or improper operation of the device resulted in the over .08 reading and that the accused was in fact below the legal limit at the time of driving, are not justified and therefore violate s. 11( d ) of the Charter . I respectfully disagree. In my view, all that these provisions do is to require that any doubt about the proper functioning or operation of the instrument be material to the issue of whether the accused’s BAC at the time of driving was in fact below .08. Even assuming for a moment that s. 258(1)( c ) did in fact violate s. 11 ( d ), this requirement of materiality would in my view constitute a reasonable and demonstrably justified limitation on the right to be presumed innocent.

5. Summary

[ 121 ] To summarize, the nature of the scheme is this:

• A reading obtained in accordance with the statutory requirements is deemed accurate, absent some evidence suggesting that the reading exceeded .08 as a result of instrument malfunction or improper operation and that the accused’s BAC did not exceed .08 at the time of driving.

• Carter evidence is not admissible to raise a reasonable doubt that the instrument malfunctioned, or was operated improperly. When concerned with the first presumption of identity, it is only admissible to raise a reasonable doubt that the accused’s BAC exceeded .08 at the time of driving. It follows that the presumption that the BAC at the time of driving is the same as at the time of testing cannot be rebutted by Carter evidence alone challenging the accuracy of the reading.

• Carter evidence is admissible to rebut the second presumption of identity provided that the evidence is consistent with both the BAC indicated by the results at the time of testing and with a BAC at the time of driving which is below .08. In other words, consumption evidence in the cases of bolus and intervening drinking is admissible provided that it is consistent with both the test results and innocence.

B. Overview of Conclusions

[ 122 ] In overview, my position respecting the constitutionality of this scheme is as follows. First , Parliament in formulating these provisions was entitled to act on the basis of widely accepted scientific evidence and the Court should take such evidence as is properly before the Court into account in assessing the constitutionality of the provisions. Second , the evidence about the reliability of the breathalyzer analysis conducted under the statutory conditions is such that it would be speculative to have a reasonable doubt about its accuracy in the absence of any evidence supporting the contention that an over .08 reading should have been an under .08 reading. On this basis, my view is that the presumption of accuracy does not limit the right to be presumed innocent. Third , the scientific evidence supports the view that so-called Carter evidence is such an unreliable indicator of the accuracy of an approved device that such evidence which is advanced for that purpose may be excluded without limiting the right to make full answer and defence. Fourth , Parliament is entitled to legislate to give effect to these widely accepted notions rather than to require them to be proved by evidence in every “blowing over” trial. Finally , if there is any limitation of the right to be presumed innocent, it is reasonable and demonstrably justified.

III. Analysis

A. Introduction

[ 123 ] I agree with my colleague that none of the challenged provisions limits the right under s. 11( c ) of the Charter not to be compelled to testify. I also agree, although for somewhat different reasons, that none of the provisions limits the right to make full answer and defence as guaranteed under s. 7 of the Charter . I respectfully do not agree with my colleague’s analysis or conclusion in relation to the presumption of innocence guaranteed by s. 11( d ) .

[ 124 ] Taking a broad view of the respondent’s position, there are two types of constitutional issues advanced. The first relates to the fact that the provisions place a burden on the accused to point to evidence giving rise to a reasonable doubt on the issues of accuracy and identity — that is, that the test results are accurate at the time of testing and represent the BAC at the time of driving. The second relates to the fact that the provisions impose restrictions on the types of evidence that are admissible to raise a reasonable doubt. As outlined earlier, Carter evidence is not admissible to challenge the proper functioning or operation of the device and, in order to raise a doubt about accuracy, the accused must point to evidence that not only the device malfunctioned or operated improperly, but as well that this resulted in the over .08 reading and that his or her BAC in fact did not exceed .08 at the time of driving.

[ 125 ] Although all of the provisions are challenged under ss. 11 ( d ) and 7 , my view is that the constitutionality of the provisions which address the burden of proof are best analyzed under s. 11 ( d ), while those which limit the relevance of, or exclude evidence in relation to, particular issues are best analyzed under s. 7 . Simply put, the provisions which deal with the burden of proof most directly engage the presumption of innocence whereas the provisions which deal with limitations on defence evidence most directly engage the right to make full answer and defence as guaranteed through s. 7 . I will first address s. 7 and then turn to s. 11 ( d ).

B. Section 7 of the Charter

1. Section 258(1)( d.01 )

[ 126 ] Section 258(1)( d.01 ), it will be remembered, is the provision which excludes Carter evidence to challenge the accuracy of the test results. It was argued that s. 258(1)( d.01 ) violates s. 7 of the Charter in two respects: first, because it excludes logically probative defence evidence and second because it creates hurdles which make the prospect of successfully rebutting the presumptions of accuracy and identity contained at s. 258(1)( c ) all but illusory.

[ 127 ] Turning to the first ground of challenge, I agree with Deschamps J.’s rejection of it. In my view, the exclusion of Carter evidence to challenge the proper functioning or operation of the approved instrument does not violate s. 7 of the Charter .

[ 128 ] There is no doubt that s. 258(1)( d.01 ) has the effect of restricting the use of evidence which has already been recognized by this Court as being logically probative of the accuracy of the test results: see, e.g., R. v. Gibson , 2008 SCC 16 , [2008] 1 S.C.R. 397, at paras. 64 and 78 . The exclusion of this evidence for the purpose of raising a reasonable doubt about the accuracy of the results therefore engages the right to make full answer and defence as guaranteed under s. 7 . The Court has held that it is a principle of fundamental justice that defence evidence may only be excluded if its probative value is substantially outweighed by its prejudicial effect on the trial process: see R. v. Seaboyer , 1991 76 (SCC) , [1991] 2 S.C.R. 577, at pp. 609-12; Gibson , at para. 64 .

[ 129 ] In holding that s. 258(1)( d.01 ) does not violate this principle, Deschamps J. points out that, while consumption evidence could logically tend to discredit the validity of breathalyzer test results, expert testimony and reports show that its reliability for this purpose is low as compared to the reliability of the breathalyzers themselves. Furthermore, testimony before a Senate Committee and a study completed on behalf of Transport Canada are to the effect that the unrestricted use of consumption evidence leads to a number of acquittals that is disproportionate in relation to the actual likelihood of breathalyzer inaccuracies. From this, my colleague concludes, and I agree, that the evidence submitted before the Court supports the conclusion that, in the absence of evidence directly putting in doubt the proper functioning or operation of a breathalyzer, the results obtained are reliable and consumption evidence challenging this reliability is misleading and likely to lead to incorrect decisions. There are also alternative and non-misleading means of challenging the accuracy of the readings. All of this being the case, the restrictions imposed by Parliament upon the use of consumption evidence do not run contrary to the Seaboyer principles.

[ 130 ] Underlying this reasoning is an important principle with which I agree. In formulating a statutory provision which excludes relevant evidence, Parliament is entitled to act on the basis of scientific and other evidence about the actual probative value of the evidence in question. It is not required to act simply on the basis of how a reasonable trier of fact might weigh the evidence in question, without the benefit of that scientific and other evidence. In my view, the same principle applies to the courts when they assess the constitutionality of the provision. They are entitled to take that evidence into account provided that it is properly before them and to uphold the provision if satisfied that it passes constitutional muster in light of that evidence.

[ 131 ] My colleague applies this principle in her s. 7 analysis. Although a reasonable trier of fact might not question the utility of challenging the accuracy of the device by means of Carter evidence, the Court is entitled to assess the constitutionality of the provision in light of the evidence about its unreliability and propensity to produce wrong results. As I will discuss shortly, my view is that the same principle ought to be applied when considering the s. 11 ( d ) challenge.

[ 132 ] The basis of my colleague’s s. 7 analysis is the determination that Carter evidence is not sufficiently probative and tends to mislead when adduced to challenge the accuracy of readings. I prefer to base my s. 7 determination somewhat more narrowly. In my view, it is enough to hold that the parties contesting the provision have not shown on the record before the Court that s. 258(1)( d.01 ) limits in any meaningful respect the right to make full answer and defence. In the face of the compelling evidence presented by the Crown about the generally misleading nature of Carter evidence in relation to the accuracy of the breathalyzer, those challenging the exclusion of this evidence had to advance some evidence suggesting that, despite its great potential to mislead, there remained some reason not to restrict the use of Carter type evidence. There is no such evidence in this record.

[ 133 ] A key submission was that there are undetectable errors with breathalyzer results. The argument points to hypothetical scenarios in which irrefutable consumption evidence is presented, such as a video of an accused not drinking throughout a night at a bar. Such irrefutable consumption evidence would conclusively prove that he or she could not have had a BAC over .08 at the time of taking a breathalyzer test. This, it is suggested, is an example of a situation in which, even though there was no discernible malfunction or improper operation of the device, an innocent person would be convicted by virtue of the provision barring consumption evidence to challenge the accuracy of the reading. I cannot accept this contention.

[ 134 ] Although hypotheticals can form the basis of a Charter challenge, they must be reasonable. In the context of the evidence presented to us regarding the general reliability of breathalyzers, I am unable to find that the hypotheticals submitted in support of the position that breathalyzers may generate undetectable errors are in any way persuasive. I do not exclude that a future challenge may be mounted with stronger evidence in this regard, but the evidence in this case does not reach that level.

[ 135 ] In sum, as in any constitutional challenge, an appropriate evidentiary foundation is required. I am of the view that this foundation is lacking in relation to s. 258(1)( d.01 )’s alleged incompatibility with s. 7 of the Charter .

[ 136 ] I turn, therefore, to the other ground advanced in support of the s. 7 challenge — that the provision makes a defence “illusory”. I agree with my colleague Deschamps J.’s rejection of this submission. I would add only that I do not understand her reasons to be establishing any new principle in relation to the Crown’s obligation to make disclosure to the defence.

2. Section 258(1)( c )

[ 137 ] As noted, s. 258(1)( c ) restricts evidence in relation to the accuracy of the device to evidence that tends to show three things: (1) that the device malfunctioned or the analysis was performed improperly, (2) that the improper performance resulted in the determination that the accused’s BAC exceeded .08, and (3) that the accused’s BAC was in fact lower than .08 at the time of the offence.

[ 138 ] In my view, the first two of these elements do nothing more than to recognize the reality that breathalyzer readings, when obtained under the statutory requirements, should be taken as accurate absent some reason to think otherwise. In other words, absent some evidence to suggest that the analysis is not accurate, a reasonable doubt based simply on the general notion that technology may be fallible or that there is a hypothetical possibility not founded on the evidence that the device malfunctioned or was not operated properly would not be a rational conclusion. Thus, requiring the inference of accuracy to be drawn absent evidence to the contrary does not limit the right to make full answer and defence. As for the third component, which requires some evidence that the accused’s BAC was in fact lower than .08 at the time of the offence, it does no more than set out in statutory form what this Court has consistently held is required as a matter of logic and relevance to rebut the presumption of accuracy.

[ 139 ] Since at least the time of this Court’s decision in R. v. Crosthwait , 1980 182 (SCC) , [1980] 1 S.C.R. 1089, it has been clear that an accused wishing to rebut a presumption that the analysis was accurate when taken had to be able to point to evidence capable of raising a doubt that his or her blood alcohol content was below the legal limit at the time of testing . Pigeon J. for the Court noted that “[w]hat is necessary to furnish evidence to the contrary is some evidence which would tend to show an inaccuracy in the breathalyzer or in the manner of its operation . . . of such a degree and nature that it could affect the result of the analysis to the extent that it would leave a doubt as to the blood alcohol content of the accused person being over the allowable maximum ”: p. 1101 (emphasis added). This interpretation of what is required to rebut a presumption that the readings were accurate has never been questioned by this Court: see, e.g., R. v. St. Pierre , 1995 135 (SCC) , [1995] 1 S.C.R. 791, at paras. 34-42 ; R. v. Boucher , 2005 SCC 72 , [2005] 3 S.C.R. 499, at para. 16 . This principle applies in my view to the presumption of accuracy now found in s. 258(1)( c ). To the extent that the challenged provisions require some evidence that an accurate breathalyzer reading would have been below .08, the provisions simply reflect what is required to constitute logically probative “evidence to the contrary” in this setting.

C. Section 11(d) of the Charter

[ 140 ] The presumption of innocence, guaranteed by s. 11( d ) of the Charter , is fundamental to our understanding of the criminal process. As Cory J. put it in R. v. Lifchus , 1997 319 (SCC) , [1997] 3 S.C.R. 320, at para. 27 : “If the presumption of innocence is the golden thread of criminal justice then proof beyond a reasonable doubt is the silver and these two threads are forever intertwined in the fabric of criminal law.”

[ 141 ] From the time of this Court’s first pronouncements, the presumption of innocence has been understood as having three elements. First, the Crown bears the burden of proving the accused’s guilt; second, the standard of proof required is proof beyond a reasonable doubt; and third, the Crown has the obligation of making out a case to answer against the accused before he or she need respond, either by testifying or by calling evidence: see Dubois v. The Queen , 1985 10 (SCC) , [1985] 2 S.C.R. 350, at p. 357. Laskin J. concisely summed up these requirements in R. v. Appleby , 1971 4 (SCC) , [1972] S.C.R. 303, at p. 317: “[T]he presumption of innocence gives an accused the initial benefit of a right of silence and the ultimate benefit . . . of any reasonable doubt.”

[ 142 ] I will set out my analysis of the s. 11( d ) issue under four headings: (1) an evidence-based approach to the provisions; (2) what s. 11( d ) requires; (3) the nature and effect of the provisions; and (4) do these provisions limit the right to be presumed innocent?

1. An Evidence-Based Approach to the Provisions

[ 143 ] Parliament, in formulating provisions about the burden of proof, is entitled to take into account scientific and other material about the true probative value of certain categories of evidence. The courts, in assessing the constitutionality of these provisions, are entitled to consider the same sort of material, provided of course that it is properly before the court. Thus, in considering the accuracy of breathalyzer tests administered in accordance with the statutory requirements, the courts are entitled to consider scientific and other material properly admitted in relation to that issue. This material may be taken into account in deciding whether a provision violates the presumption of innocence. So, for example, in considering a challenge under s. 11 ( d ) to s. 51 of the Controlled Drugs and Substances Act , S.C. 1996, c. 19, which makes an analyst’s certificate setting out the results of an analysis proof of that fact absent evidence to the contrary, the courts would be entitled to consider the reliability and validity of analyses carried out under the statutory requirements.

[ 144 ] If the courts do not apply this approach, Parliament would be prevented from legislating that these sorts of sound factual inferences should be drawn in the absence of some reason not to do so. The result would be that the scientific basis for drawing these inferences would have to be adduced in evidence in every case.

2. What Section 11( d ) Requires

[ 145 ] My colleague and I are in essential agreement about what s. 11 ( d ) requires. A provision limits the right to be presumed innocent guaranteed by s. 11 ( d ) if it either (a) relieves the Crown of having to present a case to meet before the accused is called on to answer or (b) creates the risk of conviction even if, without the provision, the trier of fact could have a reasonable doubt about the accused’s guilt.

[ 146 ] As noted earlier, the requirements of s. 11 ( d ) were set out in Dubois , at p. 357: “ Section 11 ( d ) imposes upon the Crown the burden of proving the accused’s guilt beyond a reasonable doubt as well as that of making out the case against the accused before he or she need respond, either by testifying or by calling other evidence.”

[ 147 ] R. v. Oakes , 1986 46 (SCC) , [1986] 1 S.C.R. 103, is another seminal case on s. 11 ( d ). Mr. Oakes challenged the constitutionality of then s. 8 of the Narcotic Control Act , R.S.C. 1970, c. N-1 (repealed), which commanded that a trier of fact infer from an accused’s guilt for possession of narcotics that such possession was for the purposes of trafficking in them, unless the accused established that it was not so. The Court held that s. 8 violated s. 11( d ) of the Charter because it required that an accused disprove, on a balance of probabilities, the existence of a presumed fact which was an important element of the offence charged, making it possible for an accused to be convicted despite the existence of a reasonable doubt: at pp. 132 and 134.

[ 148 ] The intervener Association québécoise des avocats et avocates de la défense (AQAAD) submits that the provisions limit the right to be presumed innocent in part because the proof of the readings generated by the device does not “lead inexorably” to their accuracy or their identity with the accused’s BAC at the time of driving. In my view, the “leads inexorably” analysis is not useful in this case for reasons which I will explain.

[ 149 ] R. v. Vaillancourt , 1987 2 (SCC) , [1987] 2 S.C.R. 636, is the origin of the “inexorably leads” test. The case concerned the constitutionality of then s. 213( d ) Cr. C. , which provided that culpable homicide was murder whenever a person caused the death of another while committing certain crimes and using or bearing a weapon in so doing. Lamer J. began with a careful analysis of the challenged provision in the context of the other murder provisions of the Code in order to determine its “true nature and scope”: p. 644. He concluded that proof that the accused did one of the acts listed in s. 213( a ) through ( d ) (i.e., facilitating the commission of an offence or flight, administering a stupefying or overpowering thing for that purpose, stopping the breath of a human being for that purpose, or using or having a weapon during the commission of, or flight from, an offence) “is substituted for proof of any subjective foresight or even objective foreseeability of the likelihood of death”: p. 646. He then held that for the purposes of the challenge to s. 213( d ), s. 7 of the Charter requires that in order to be convicted of murder, the accused be proved to have at least objective foresight of death. In other words, for the purposes of the analysis of that provision, the Charter was taken to require as an essential element of the offence of murder at least objective foresight of death: p. 654.

[ 150 ] Having established that this was an essential element of the offence, Lamer J. moved on to consider s. 11 ( d ), noting that it requires that the Crown establish all of the essential elements of the offence beyond a reasonable doubt including those required by s. 7 of the Charter . He concluded, at p. 655, that “[a]ny provision creating an offence which allows for the conviction of an accused notwithstanding the existence of a reasonable doubt on any essential element infringes ss. 7 and 11 ( d )”. As he put it, at p. 655, “[i]t is clear . . . that what offends the presumption of innocence is the fact that an accused may be convicted despite the existence of a reasonable doubt on an essential element of the offence, and I do not think that it matters whether this results from the existence of a reverse onus provision or from the elimination of the need to prove an essential element.” Vaillancourt did not change in any way the ultimate test for compliance with s. 11( d ) of the Charter . In assessing s. 213( d )’s compliance with the presumption of innocence, the Court remained of the view that

the acid test of the constitutionality of s. 213 is this ultimate question: Would it be possible for a conviction for murder to occur under s. 213 despite the jury having a reasonable doubt as to whether the accused ought to have known that death was likely to ensue? [Emphasis in original; at p. 657.]

[ 151 ] From this, I conclude the following. First, before moving to the s. 11 ( d ) analysis, it is important to fully understand the provision’s “true nature and scope”, assessed in its statutory context. Second, the risk of conviction in the presence of a reasonable doubt remains the “acid test” of s. 11 ( d ). This point is reinforced by the Court’s decision in R. v. Whyte , 1988 47 (SCC) , [1988] 2 S.C.R. 3, where the Court stated that the distinction between elements of the offence and other aspects of the charge is irrelevant and that the “real concern is not whether the accused must disprove an element or prove an excuse, but that an accused may be convicted while a reasonable doubt exists. When that possibility exists, there is a breach of the presumption of innocence”: p. 18. It is also the basis upon which the Court held, in R. v. Schwartz , 1988 11 (SCC) , [1988] 2 S.C.R. 443, that placing an evidentiary burden on an accused to point to evidence raising a reasonable doubt about guilt did not, in itself, limit the right to be presumed innocent. In order to have that effect, the evidentiary burden had to require conviction even in the presence of a reasonable doubt: pp. 485-86. Third, the “inexorably leads” test was used in Vaillancourt to assess whether legislative substitution of an element for a constitutionally required element of the offence ran afoul of s. 11 ( d ). It was used as a way of testing, in those circumstances, whether there was any risk of the accused being convicted of the offence in the presence of a reasonable doubt about the essential element. It is in these sorts of cases that the “inexorably leads” analysis is most useful.

[ 152 ] Only if there is a rational basis to have a doubt about the accused’s guilt once the elements of a presumption are established is the presumption of innocence implicated. As stated in Lifchus , “a reasonable doubt is a doubt based on reason and common sense which must be logically based upon the evidence or lack of evidence” (para. 30); it does not involve proof to an absolute certainty or proof beyond any doubt; a reasonable doubt is not an imaginary or frivolous doubt: paras. 31 and 36. This is the standard that must be applied when considering whether a presumption risks requiring conviction in the presence of a reasonable doubt about guilt. What is required is not proof to a certainty, but proof beyond any reasonable doubt that could according to logic and common sense arise from the evidence or an absence of evidence.

[ 153 ] I turn then to explain what in my view is the nature and effect of the provisions in this case and why they do not limit the right to be presumed innocent and therefore do not require justification under s. 1 of the Charter .

3. The Nature and Effect of the Provisions

[ 154 ] I take a different view of s. 258(1)( c ) than does my colleague Deschamps J. This difference arises from my interpretation of the three matters under s. 258(1)( c ) in respect of which the accused must point to some evidence raising a reasonable doubt; that is, first, that the device malfunctioned or was operated improperly; second, that the improper performance resulted in an over .08 reading; and third, that the accused’s BAC at the time of driving was under .08.

[ 155 ] The first condition requires that an accused raise a doubt as to whether “the approved instrument was malfunctioning or was operated improperly”. This, to me, does not require that an accused limit his or her evidence to the precise accuracy of the results. It requires him or her to address the breathalyzer’s functioning or the way it was operated. This means that the first condition could be met even if a breathalyzer had a malfunction which caused it to under estimate an individual’s BAC, or if a qualified technician failed to follow standard procedure in operating a breathalyzer, but that his shortcomings nonetheless had no effect on the integrity of the results. Indeed, on plain reading, the first condition is not concerned with the consequences of a breathalyzer malfunction or improper operation; it is only concerned with the fact that it happened.

[ 156 ] The consequences on the other hand are scrutinized when examining whether the second condition to rebutting the presumptions is met (raising a doubt that the malfunction or improper operation resulted in the determination that the accused’s BAC was over .08). It requires that an accused raise a doubt that there is a link between the malfunction or improper operation, whatever it is, and the over .08 result. What this requires, in my view, is evidence that, but for the malfunction or improper operation, it is reasonably possible that the breathalyzer would not have produced an over .08 result. This avoids making the Crown lose the benefit of the presumptions simply because, for example, a malfunction caused a breathalyzer to under estimate an individual’s BAC or because of an immaterial operation error. It also ensures that only those results whose reliability is put in doubt to the point of making it unsafe to rely upon them to convict an accused will be given any credence. In short, this simply requires that the malfunction or improper operation be material to the question of whether the accused’s BAC was over the legal limit.

[ 157 ] When the first and second conditions are interpreted as I suggest, the third condition (raising a doubt that the accused’s BAC was not over .08 at the time of driving) loses much of its significance. Once an accused successfully raises a doubt that a breathalyzer malfunction or its improper operation caused it to generate an over .08 result, a trier of fact has logically no choice but to have a doubt that the accused’s BAC at the time of testing was below .08 as well. And if that is the case, then a trier of fact equally has, in my view, no other choice but to conclude that there is a doubt that the accused’s BAC was over .08 at the time of driving, as he would have no reliable evidence upon which one could extrapolate the accused’s BAC at the time of driving. In that sense, it may be said that the second and third conditions are intimately tied to one another, such that the satisfaction of the latter will almost inevitably flow naturally from the satisfaction of the former.

[ 158 ] I return to the nature of the presumptions in s. 258(1)( c ). Taken at face value, the section deems the results of the analysis to be accurate (the presumption of accuracy) and identical to the BAC at the time of driving (the first presumption of identity). The true nature and effect of these provisions, however, are not what they appear to be at first glance.

[ 159 ] The presumption of accuracy in s. 258(1)( c ), on its own, has no effect on proof of the offence. It has that effect only when coupled with the presumption of identity. This is so, of course, because the offence is having a BAC of more than .08 at the time of driving, not having a BAC at that level at the time of testing. In relation to the proof of the offence, the effect of the provision is thus twofold: a reading of over .08 is deemed to be accurate and, secondly, a person with such a BAC at the time of testing is deemed to a have a BAC of over .08 at the time of driving. While the provision specifies that the reading is to be taken as exactly accurate and to correspond exactly to the BAC at time of driving, this is not its effect, having regard to the definition of the offence.

[ 160 ] To rebut the accuracy of the initial result, the accused must raise a doubt about the three matters I mentioned earlier: that the device malfunctioned or the analysis was performed improperly, that this resulted in the determination that the person’s BAC was over .08 and that the BAC was in fact below .08. The effect of this is that a reading of over .08 is deemed accurate unless the accused can point to evidence that raises a doubt that there was a problem with the testing or analysis resulting in the over .08 reading.

[ 161 ] (I pause here to note that we need not consider further the limitation in s. 258(1)( d.1 ) and 258(1)( d.01 ) on the use of Carter evidence to rebut this presumption. In my analysis of the s. 7 challenge, I have concluded that such evidence is so unreliable in comparison to the reliability of the breathalyzer that it may properly be excluded in relation to that issue. That conclusion, in my view, also resolves any challenge to the exclusion of Carter evidence in relation to the presumption of innocence. As I see it, one cannot conclude, on the one hand, that Carter evidence is so unreliable and prejudicial to accurate fact finding that the defence may be prevented from presenting it, but then conclude, on the other hand, that excluding it offends the presumption of innocence because such evidence could support a reasonable doubt about the accuracy of the results. If presenting the evidence is not a valid way of challenging the accuracy of the readings for the purposes of s. 7 , excluding the evidence in an effort to raise a reasonable doubt cannot offend s. 11 ( d ). The s. 11 ( d ) analysis, like the s. 7 analysis, must be based on the true probative value of the excluded evidence.)

[ 162 ] The situation is more complicated in relation to the nature of the first presumption of identity — that is, that the readings at the time of testing reflect the BAC at the time of driving. Once again, while that is the form of the provision, we must bear in mind that the offence is not having a particular BAC, but having a BAC over .08. We must also bear in mind that it is generally known that BAC constantly changes as alcohol is absorbed and eliminated. Parliament should not be taken to have legislated so as to require that the opposite of this incontestable fact be presumed. It is in my view more realistic to describe the effect of the provision as presuming that a person who has a BAC of over .08 at the time of testing would not have a lower BAC at the time of driving . This description of the provision’s effect is reinforced by the second presumption of identity found in s. 258(1)( d.1 ) as I will discuss shortly.

[ 163 ] An accused can avoid the operation of the first presumption of identity in two ways. First, he or she can raise a doubt about the accuracy of the initial result, as described above. Obviously if there is a doubt about the accuracy of the initial result, neither the presumption of accuracy nor the first presumption of identity can operate. Secondly, the accused can attempt to rebut the second presumption of identity by presenting Carter evidence, provided that it is consistent with both the accuracy of the initial result and with a BAC at the time of driving under .08. As I explained earlier, this permits evidence of bolus or intervening drinking. If the second presumption of identity does not operate, neither does the first.

[ 164 ] To sum up, the true nature and effect of the provisions is this. If the result of the analysis shows a BAC over .08, the BAC of the accused is presumed to be over .08 in the absence of evidence raising a doubt that, but for malfunction or improper operation, the result should in fact have been under .08. If there is no doubt about the accuracy of the initial result, it is presumed that the accused’s BAC at the time of driving was not lower than the BAC at the time of testing, absent evidence that is both consistent with the test result and a BAC below .08 at the time of driving.

[ 165 ] Understood in this way, the provisions are simply rules about the burden of proof. They are not presumptions with basic facts in the traditional sense. The presumption of accuracy does not depend simply on the “deduction that may logically and reasonably be drawn” from the test result with regard to the actual BAC of the accused at the time of driving. The presumption rather reflects the demonstrated accuracy of the breathalyzer analysis conducted under the statutory conditions. Similarly, the first presumption of identity does not depend on logical deduction that the BAC at the time of driving is identical to that at the time of testing. There of course could be no such logical deduction because it is generally known that this cannot be the case. Rather, the presumption depends on the demonstrated fact that the accused’s BAC does not go up between the time of driving and the time of testing, absent the unusual situations of bolus or intervening drinking. My view, therefore, is that these presumptions cannot be analyzed for constitutional purposes as if they were mere factual presumptions that depend only on the force of the logical deduction from the proved facts that the presumed fact exists.

4. Do These Provisions Limit the Right to Be Presumed Innocent?

[ 166 ] In light of the true nature and effect of the provisions, my view is that the questions in relation to whether these provisions limit the right to be presumed innocent are these:

1. If the results of an analysis which complied with the statutory requirements show that an accused’s BAC at the time of testing was over .08, would it be reasonable for a trier of fact to have a doubt that the accused’s BAC was in fact over .08 at that time, in the absence of evidence raising a doubt about whether improper operation or malfunctioning of the device produced a result above point .08?

2. If an accused’s BAC was shown to be above .08 at the time of analysis, would it be reasonable for a trier of fact to have a doubt that the accused’s BAC was in fact over .08 at the time of driving in the absence of evidence to the contrary?

[ 167 ] In my view, the answer to both of these questions is “no”.

[ 168 ] I turn first to the presumption of accuracy. The first two matters set out as necessary to raise a reasonable doubt are that the approved instrument was malfunctioning or was operated improperly and that this resulted in the determination that the accused’s BAC exceeded the legal limit. There is strong support for the view that a breathalyzer test conducted according to the statutory requirements will yield a reliable and accurate result. Of course, this is not to say that the devices are infallible; but that is not what the presumption assumes. It seems to me that, given the evidence about the accuracy of the results, it would be unreasonable to have a doubt on this subject unless there is some basis in the evidence or the absence of evidence to ground a doubt about the proper functioning or operation of the device. In other words, absent some basis to suggest that something may have gone wrong with the breathalyzer or during the testing process, it would be unreasonable for a trier of fact not to consider the breathalyzer results to be reliable. In requiring some evidence tending to show improper functioning or operation, s. 258(1)( c ) simply enacts common sense in light of accepted scientific fact.

[ 169 ] In my view, Parliament is entitled to legislate this rather than require the evidence to be called in every “blowing over” prosecution. There is no risk of conviction in the presence of a reasonable doubt about guilt.

[ 170 ] The third aspect of s. 258(1)( c ) requires some evidence tending to raise a doubt that the accused’s BAC in fact did not exceed .08 at the time of the offence. I note to begin that as I have discussed earlier, if there is doubt about the first two matters, it is hard to conceive of why there would not also be a doubt about this third aspect. In any event, this Court has decided that in order to constitute evidence to the contrary as a matter of logic and relevance, that evidence must tend to raise a doubt that the BAC in fact did not exceed .08 at the time of testing: Crosthwait . It follows that this third aspect of the provision simply translates that requirement for materiality into the consideration of whether the device functioned or was operated improperly. The provision does not create any risk of conviction in the presence of a reasonable doubt about guilt. The provision simply requires that in order for there to be a reasonable doubt about whether the BAC reading was over .08, there must be some basis in the evidence to think that the device either malfunctioned or was improperly operated in a material respect.

[ 171 ] I therefore conclude that the presumption of accuracy in s. 258(1)( c ) does not create a risk of conviction in the presence of a reasonable doubt about guilt. It therefore does not limit the right to be presumed innocent and there is no need to consider whether any limitation is justified under s. 1.

[ 172 ] I turn then to the presumptions of identity in ss. 258(1)( c ) and 258(1)( d.1 ). In my view, there is overwhelming evidence that a breathalyzer test administered in accordance with the statutory requirements and which reveals an over .08 result is a reliable indication that the accused had a BAC which was equal to or higher than that at the time of driving: A.R., vol. 21, at pp. 100-102.

[ 173 ] As the report of Jacques Tremblay, filed in this case, stipulates, an individual’s BAC rises steadily as one consumes alcohol ([ TRANSLATION ] “absorption stage”), stabilizing itself rapidly after the last consumption. The BAC thereafter remains more or less stable for a period estimated at anywhere between 0 and two hours ([ TRANSLATION ] “plateau stage”). This stability is due to the fact that, in the presence of a regular rhythm of alcohol intake, the body’s rate of alcohol elimination is more or less equal to its rate of alcohol absorption during this period, such that the alcohol still being digested by the body after the last consumption has little to no effect on the individual’s BAC. It is only after that plateau period that the BAC decreases steadily until completely eliminated by the body ([ TRANSLATION ] “elimination stage”). As such, in cases where individuals, as usual, consume alcohol at a regular rate, a test taken within two hours of their last consumption will usually reveal a BAC that is either identical or lower than that which he had at the time of that consumption. This also leads to the conclusion that a person arrested and asked to submit himself or herself to a breathalyzer test will, when having consumed alcohol at a normal rate of consumption, either be in the plateau stage or the elimination stage at the time of the test, and thus, have a BAC that is either identical to or lower than when he was driving. Bearing in mind that the offence is having any BAC above .08 and given that I have concluded above that the presumption of accuracy is not objectionable, there is no infringement of the right to be presumed innocent by deeming that the BAC at the time of testing is the same as at the time of driving. Once again, Parliament has simply legislated well-established facts so that they do not have to be proved in every case. There is no risk of conviction on the basis of a reasonable doubt that has a basis in common sense and logic in the evidence or the absence of evidence.

[ 174 ] In my opinion, a doubt about the presumptions of identity based on bolus or intervening drinking would be speculative, absent evidence supporting the fact that one or the other of those scenarios had actually occurred. I note that the Ontario Court of Appeal has ruled in several cases that triers of fact are entitled to draw an inference that “normal people do not consume large quantities of alcohol shortly before, or while, driving”, in the absence of evidence putting in doubt the soundness of drawing this inference in a particular case: R. v. Bulman , 2007 ONCA 169 , 221 O.A.C. 210, at para. 13 . See also R. v. Grosse (1996), 1996 6643 (ON CA) , 29 O.R. (3d) 785; R. v. Hall , 2007 ONCA 8 , 83 O.R. (3d) 641; R. v. Paszczenko , 2010 ONCA 615 , 103 O.R. (3d) 424. I would apply the same principle to drinking after driving but before the breathalyzer test is administered. Even if it might be said that a reasonable doubt could exist about drinking between the time of driving and testing, placing an evidentiary burden on the accused to simply point to evidence capable of raising a doubt on this issue would readily pass the justification requirement under s. 1 of the Charter . The fact of post‑driving drinking is peculiarly in the knowledge of the accused and it would be unduly onerous to require the prosecution to negate this rather unusual possibility in every case even when it had no foundation in the evidence.

[ 175 ] I am also of the view that the challenged provisions do not relieve the Crown of its obligation to present a case to meet before the accused is called on to answer. Where an over .08 breathalyzer test result is obtained in accordance with the statutory requirements, a trial judge cannot, in my opinion, conclude that there is no evidence upon which he could reasonably convict an accused.

5. Section 1 of the Charter

[ 176 ] I turn to the s. 1 justification on the assumption that my colleague, Deschamps J. is correct that the second and third requirements under s. 258(1)( c ) limit the right to be presumed innocent. These are the requirements to point to evidence raising a doubt that the malfunctioning or the improper operation of the device resulted in an over .08 reading and about whether the accused’s BAC at the time of driving was under .08. My view is that any limitation is reasonable and demonstrably justified.

[ 177 ] As I explained above, I read the requirements in s. 258(1)( c ) differently than does my colleague. The second requirement — to point to evidence raising a doubt that the malfunctioning or the improper operation of the device resulted in an over .08 reading — simply requires the evidence to be material to the issue of whether the accused had a BAC over the legal limit. Requiring evidence to be material to that issue, to my way of thinking, easily passes all of the steps of the Oakes analysis. The provision simply prevents evidence being adduced that is not logically probative of the accused’s innocence. Without this second requirement, evidence that the device consistently underestimates the true BAC would rebut the presumption even though there is no realistic possibility that the accused’s BAC was lower than the results obtained from the device.

[ 178 ] As for the third requirement — that the accused point to evidence raising a doubt that the accused’s BAC at the time of driving was under .08 — I find it does not impose any significant burden that is not already discharged by meeting the second requirement. As I explained earlier, I find it hard to imagine how there could be a doubt about whether the device produced a result over .08 as a result of a malfunction or improper operation and yet there would not inevitably also be a doubt that the accused’s BAC was in fact under .08.

[ 179 ] My conclusion is that the provisions which my colleague would hold to limit the right to be presumed innocent readily pass the s. 1 justification and should be upheld as reasonable limits demonstrably justified in a free and democratic society.

D. Conclusion

[ 180 ] I would uphold the conviction entered by Judge Chapdelaine against the respondent. I would however allow the appeal with respect to the constitutional questions and answer them as follows:

1. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C-46, infringe s. 7 of the Canadian Charter of Rights and Freedoms ?

Answer: No.

2. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: It is not necessary to answer this question.

3. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C-46, infringe s. 11( c ) of the Canadian Charter of Rights and Freedoms ?

Answer: No.

4. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: It is not necessary to answer this question.

5. Do ss. 258(1)( c ), 258(1)( d.01 ) and 258(1)( d.1 ) of the Criminal Code , R.S.C. 1985, c. C-46, infringe s. 11( d ) of the Canadian Charter of Rights and Freedoms ?

Answer: No.

6. If so, is the infringement a reasonable limit prescribed by law that can be demonstrably justified in a free and democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?

Answer: It is not necessary to answer this question.

APPENDIX

Criminal Code , R.S.C. 1985, c. C‑46

258. (1) In any proceedings under subsection 255(1) in respect of an offence committed under section 253 or subsection 254(5) or in any proceedings under any of subsections 255(2) to (3.2),

( c ) where samples of the breath of the accused have been taken pursuant to a demand made under subsection 254(3), if

(i) [Repealed before coming into force, 2008, c. 20, s. 3 ]

(ii) each sample was taken as soon as practicable after the time when the offence was alleged to have been committed and, in the case of the first sample, not later than two hours after that time, with an interval of at least fifteen minutes between the times when the samples were taken,

(iii) each sample was received from the accused directly into an approved container or into an approved instrument operated by a qualified technician, and

(iv) an analysis of each sample was made by means of an approved instrument operated by a qualified technician,

evidence of the results of the analyses so made is conclusive proof that the concentration of alcohol in the accused’s blood both at the time when the analyses were made and at the time when the offence was alleged to have been committed was, if the results of the analyses are the same, the concentration determined by the analyses and, if the results of the analyses are different, the lowest of the concentrations determined by the analyses, in the absence of evidence tending to show all of the following three things — that the approved instrument was malfunctioning or was operated improperly, that the malfunction or improper operation resulted in the determination that the concentration of alcohol in the accused’s blood exceeded 80 mg of alcohol in 100 mL of blood, and that the concentration of alcohol in the accused’s blood would not in fact have exceeded 80 mg of alcohol in 100 mL of blood at the time when the offence was alleged to have been committed;

( d.01 ) for greater certainty, evidence tending to show that an approved instrument was malfunctioning or was operated improperly, or that an analysis of a sample of the accused’s blood was performed improperly, does not include evidence of

(i) the amount of alcohol that the accused consumed,

(ii) the rate at which the alcohol that the accused consumed would have been absorbed and eliminated by the accused’s body, or

(iii) a calculation based on that evidence of what the concentration of alcohol in the accused’s blood would have been at the time when the offence was alleged to have been committed;

( d.1 ) if samples of the accused’s breath or a sample of the accused’s blood have been taken as described in paragraph ( c ) or ( d ) under the conditions described in that paragraph and the results of the analyses show a concentration of alcohol in blood exceeding 80 mg of alcohol in 100 mL of blood, evidence of the results of the analyses is proof that the concentration of alcohol in the accused’s blood at the time when the offence was alleged to have been committed exceeded 80 mg of alcohol in 100 mL of blood, in the absence of evidence tending to show that the accused’s consumption of alcohol was consistent with both

(i) a concentration of alcohol in the accused’s blood that did not exceed 80 mg of alcohol in 100 mL of blood at the time when the offence was alleged to have been committed, and

(ii) the concentration of alcohol in the accused’s blood as determined under paragraph ( c ) or ( d ), as the case may be, at the time when the sample or samples were taken;

Appeal allowed in part, Rothstein and Cromwell JJ. dissenting in part.

Solicitor for the appellants: Attorney general of Québec, Montréal.

Solicitors for the respondent: Fréchette, Blanchette, Sherbrooke.

Solicitor for the intervener the Attorney General of Canada: Attorney general of Canada, Montréal.

Solicitor for the intervener the Attorney General of Ontario: Attorney General of Ontario, Toronto.

Solicitor for the intervener the Attorney General of Manitoba: Attorney General of Manitoba, Winnipeg.

Solicitor for the intervener the Attorney General of British Columbia: Attorney General of British Columbia, Victoria.

Solicitor for the intervener the Attorney General of Alberta: Attorney General of Alberta, Edmonton.

Solicitors for the intervener Barreau du Québec: LaBrie, Gariépy & Associés, Longueuil.

Solicitors for the intervener Association québécoise des avocats et avocates de la défense: Downs Lepage, Montréal.

Solicitors for the intervener the Criminal Lawyers’ Association of Ontario: Ducharme Fox, Windsor; Burstein Bryant, Toronto.

Solicitors for the intervener the Criminal Trial Lawyers’ Association: Gunn Law Group, Edmonton.