Supreme Court of Canada

SUPREME COURT OF CANADA

Appeal Heard: November 9, 2021 Judgment Rendered: May 13, 2022 Docket: 39781

Between:

Matthew Winston Brown Appellant

and

Her Majesty The Queen Respondent

— and —

Attorney General of Canada, Attorney General of Ontario, Attorney General of Manitoba, Attorney General of British Columbia, Attorney General of Saskatchewan, Canadian Civil Liberties Association, Empowerment Council, Criminal Lawyers' Association and Women's Legal Education and Action Fund Inc. Interveners

Coram: Wagner C.J. and Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin, Kasirer and Jamal JJ.

Reasons for Judgment: (paras. 1 to 168)

Kasirer J. (Wagner C.J. and Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin and Jamal JJ. concurring)

Indexed as: R. v. Brown

2022 SCC 18

File No.: 39781.

2021: November 9; 2022: May 13.

Present: Wagner C.J. and Moldaver, Karakatsanis, Côté, Brown, Rowe, Martin, Kasirer and Jamal JJ.

on appeal from the court of appeal for alberta

Headnote

Constitutional law — Charter of Rights — Fundamental justice — Presumption of innocence — Reasonable limits — Section 33.1 of Criminal Code preventing accused from raising common law defence of self‑induced intoxication akin to automatism — Whether s. 33.1 violates principles of fundamental justice or presumption of innocence — If so, whether infringement justified — Canadian Charter of Rights and Freedoms, ss. 1, 7, 11(d) — Criminal Code, R.S.C. 1985, c. C‑46, s. 33.1.

At a house party, B consumed alcohol and magic mushrooms. Magic mushrooms contain psilocybin, an illegal drug that can bring about hallucinations. B lost his grip on reality and left the house. B was not simply drunk or high: while capable of physical movement, he was in a psychotic state and had no willed control over his actions. He broke into the nearby house of a stranger and attacked the occupant, causing permanent injuries. He then broke into another residence and the occupants called the police. B was charged with break and enter and aggravated assault, and with break and enter and mischief to property over $5,000.

At trial, B argued that he was not guilty of the offences by reason of automatism caused by the consumption of psilocybin. Expert evidence adduced at trial confirmed that B had no voluntary control over his conduct at the time. The Crown invoked s. 33.1 of the Criminal Code as a means of precluding B from relying on self‑induced intoxication akin to automatism as a defence to the charge of aggravated assault. Parliament added s. 33.1 to the Criminal Code in response to R. v. Daviault, [1994] 3 S.C.R. 63. The Court in Daviault confirmed the common law rule that intoxication is not a defence to crimes of general intent, but a majority recognized that the Charter mandated an exception where intoxication is so extreme that an accused falls into a condition akin to automatism and is incapable of voluntarily committing a guilty act or of having a guilty mind. Section 33.1 was enacted to address the constitutional failings identified by the majority in Daviault in a manner that would properly reflect the blameworthiness of the extremely self-intoxicated accused identified by the dissent. Section 33.1 blocks the defence of automatism for general intent crimes designated in s. 33.1(3), including aggravated assault and sexual assault.

B challenged the constitutionality of s. 33.1. The voir dire judge concluded that s. 33.1 violates the principles of fundamental justice and the presumption of innocence guaranteed by ss. 7 and 11(d) of the Charter and that the violations are not justified pursuant to s. 1 of the Charter. He declared s. 33.1 to be of no force and effect pursuant to s. 52(1) of the Constitution Act, 1982. As a result, B was entitled to raise the defence of extreme intoxication akin to automatism at trial. The trial judge found that the defence was an answer to both charges and entered acquittals. The Court of Appeal reversed the declaration that s. 33.1 was of no force or effect, set aside the acquittal on the count of break and enter and aggravated assault, and entered a conviction for that offence. The acquittal on the mischief charge was unaffected by s. 33.1 and not appealed.

Held: The appeal should be allowed. Section 33.1 of the Criminal Code should be declared unconstitutional and of no force or effect pursuant to s. 52(1) of the Constitution Act, 1982. The acquittal on the count of unlawful break and enter of a dwelling house and committing aggravated assault therein should be restored.

This is not a drunkenness case. B consumed a drug which, taken in combination with alcohol, provoked psychotic, delusional and involuntary conduct. Criminal liability for violent conduct produced by alcohol alone, short of the psychotic state akin to automatism experienced by B, is not in issue. The outcome of the declaration of unconstitutionality with respect to s. 33.1 has no impact on the rule that intoxication short of automatism is not a defence to violent crimes of general intent, such as assault or sexual assault. While s. 33.1 is unconstitutional, there may well have been other paths for Parliament to achieve its legitimate aims connected to combatting extreme intoxicated violence. The sense that an accused who acts violently in a state of extreme self‑induced intoxication is morally blameworthy is by no means beyond the proper reach of the criminal law. Protecting the victims of violent crime — particularly in light of the equality and dignity interests of women and children who are vulnerable to intoxicated sexual and domestic violence — is a pressing and substantial social purpose. And it was not impermissible for Parliament to enact legislation seeking to hold an extremely intoxicated person accountable for a violent crime when they chose to create the risk of harm by ingesting intoxicants.

Section 33.1(1) of the Criminal Code eliminates the defence of self‑induced intoxication akin to automatism applied to the violent offences identified in s. 33.1(3) where the accused departs markedly from the standard of care described in s. 33.1(2). Section 33.1 does not create a new predicate act offence of self-induced extreme intoxication or a new criminal negligence offence. The accused faces the full stigma of conviction and the full brunt of punishment for the general intent offence pointed to in s. 33.1(3). Section 33.1 applies when three conditions are met: the accused was intoxicated at the material time, the intoxication was self‑induced, and the accused departed markedly from the standard of reasonable care generally recognized in Canadian society by interfering or threatening to interfere with the bodily integrity of another person. When these three things are proved, it is not a defence that the accused lacked the general intent or the voluntariness required to commit the offence named in s. 33.1(3). The requirements of s. 33.1 are not, together or separately, a measure of fault; they are conditions of liability, as the use of the word "while" in s. 33.1(2) confirms. The marked departure described in s. 33.1(2) depends on proof of two facts: that the person was in a state of self‑induced intoxication that rendered them unaware of, or incapable of controlling, their behaviour, and that the violent act occurred while they were in that state. These facts are conditions of liability and not measures of fault because neither of them import a criminal negligence standard. Thus s. 33.1 deems criminal fault for the violent offence to be present based on the accused's choice to become intoxicated. What Parliament sought was to impose liability for the charged offence, and not the act of self‑induced intoxication itself.

The rights of victims of intoxicated violence, in particular the rights of women and children, should be considered at the justification stage under s. 1 of the Charter rather than informing the analysis of a possible breach of the accused's rights under s. 7. Balancing competing Charter rights under the breach analysis should occur where the rights of the accused and another party conflict and are directly implicated by state action. The equality, dignity and security interests of vulnerable groups informed the overarching public policy goals of Parliament but they are best considered under s. 1.

Section 33.1 breaches s. 7 of the Charter by allowing a conviction without proof of mens rea or proof of voluntariness. It is a principle of fundamental justice that proof of penal negligence, in the form of a marked departure from the standard of a reasonable person, is minimally required for a criminal conviction, unless the specific nature of the crime demands subjective fault. Section 33.1 requires an intention to become intoxicated but intention to become intoxicated to any degree suffices — it matters little that a person did not foresee their loss of awareness or control, and nothing is said about the licit or illicit nature of the intoxicant or its known properties. For this reason, while s. 33.1 applies to those who recklessly invite their loss of control, it also captures unexpected involuntariness, for example an unexpected reaction to a prescribed pain medication. It also imposes criminal liability where a person's intoxication carries no objective foreseeability of harm. Furthermore, instead of asking whether a reasonable person would have foreseen the risk and taken steps to avoid it and whether the failure to do so amounted to a marked departure from the standard of care expected in the circumstances, s. 33.1 deems a marked departure to be present whenever a violent act occurs while the person is in a state of extreme voluntary intoxication akin to automatism. Since s. 33.1 allows the court to convict an accused without proof of the constitutionally required mens rea, it violates s. 7 of the Charter. Section 33.1 also directs that an accused person is criminally responsible for their involuntary conduct. Because involuntariness negates the actus reus of the offence, involuntary conduct is not criminal, and the law recognizes that voluntariness for the conviction of a crime is a principle of fundamental justice.

Section 33.1 also breaches the right to be presumed innocent until proven guilty guaranteed by s. 11(d) of the Charter. To convict the accused, the Crown must prove all the essential elements of an offence beyond a reasonable doubt. A direction from Parliament that proof of one fact is presumed to satisfy proof of one of the essential elements of an offence can only comply with s. 11(d) if, in all cases, proof of the substituted fact leads inexorably to the conclusion that the essential element it replaces exists. Otherwise, the substitution may result in the accused being convicted, based on proof of the substituted fact, despite the existence of a reasonable doubt as to the essential element of the offence that it replaces. Section 33.1 improperly substitutes proof of self‑induced intoxication for proof of the essential elements of an offence. The fault and voluntariness of intoxication are substituted for the fault and voluntariness of the violent offence. This amounts to a constitutionally improper substitution. It cannot be said that in all cases under s. 33.1, the intention to become intoxicated can be substituted for the intention to commit a violent offence.

Parliament had before it a record that highlighted the strong correlation between alcohol and drug use and violent offences, in particular against women, and brought to the fore of Parliament's attention the equality, dignity, and security rights of all victims of intoxicated violence. Parliament's protective public goals cannot be understated: these interests bear meaningful attention at both principal steps in the s. 1 analysis. But the Crown must show on a balance of probabilities that the limits of ss. 7 and 11(d) brought by s. 33.1 are reasonable and demonstrably justified under s. 1 of the Charter. Given the patent risk that s. 33.1 may result in the conviction of an accused person who had no reason to believe that their voluntary intoxication would lead to a violent consequence, s. 33.1 fails at the proportionality step and thus cannot be saved under s. 1.

With respect to pressing and substantial purpose, the purpose of a provision must be properly identified with a view to justifying the infringement of the Charter, otherwise the exercise is not helpful for the balancing mandated by s. 1. In enacting s. 33.1, Parliament blocked the defence of automatism for the extremely intoxicated offender for two legitimate purposes: to protect the victims of extremely intoxicated violence, with particular attention to women and children whose equal place in society is compromised by sexual assault and other violent crimes of general intent in such circumstances; and to call offenders to answer for their choice to voluntarily ingest intoxicants where that choice creates a risk of violent crime. The protective purpose is sufficiently pressing and substantial to warrant limiting Charter rights — the protection of the public from intoxicated offenders is of sufficient importance to warrant overriding a constitutionally protected right or freedom. As for the accountability objective, it rests on a philosophical idea that one should not be able to create the conditions of one's own criminal defence to block liability for the crime committed. An individual is responsible for their involuntary state because that person's choice to ingest intoxicants and become extremely intoxicated ultimately creates a risk of violence. Stated in this manner, accountability in this context is pressing and substantial and fits appropriately within the Oakes analysis.

The deterrent and denunciating effects of s. 33.1 provide a rational connection to Parliament's protective objective. While it is true that s. 33.1 applies to an accused who could not have foreseen the risk of a loss of control or of bodily harm, it also extends to situations in which there was a foreseeable risk of a loss of control and harm. Thus, an individual who consumes an intoxicant with psychosis‑inducing effects, including those who know they lost control of their conduct while in a drug-induced psychosis in the past, will be caught by s. 33.1. It is reasonable that Parliament would expect the provision to hold some modest deterrent effect for such individuals. This deterrent effect dissuades those contemplating this kind of intoxication and, as such, s. 33.1 is rationally connected to its protective purpose. In addition, s. 33.1 is rationally connected to the objective of holding individuals accountable, in as full a manner as possible, for the choice to become extremely intoxicated and the violence committed while in that state. It is obvious that a person foreclosed from advancing a defence that could result in an acquittal is held accountable.

Section 33.1 is, however, not minimally impairing of an accused's ss. 7 and 11(d) rights. There are less harmful means of achieving Parliament's objectives in a real and substantial manner. Options have been advanced that would trench less on the rights of the accused, including a stand-alone offence of criminal intoxication. Alternatively, a path to liability for the underlying violent offence might be based on a criminal negligence standard that would allow the trier of fact to consider whether a loss of control and bodily harm were both reasonably foreseeable at the time of intoxication. This latter option could allow an accused to be convicted for the underlying violent act and not simply negligent or dangerous intoxication while achieving the minimum objective fault standard required by the Constitution.

Section 33.1 also fails on an assessment of the relative benefits and negative effects of the law under the Oakes test. At the final stage under s. 1, the question is whether there is proportionality between the overall effects of the Charter‑infringing measure and the legislative objectives. This invites the broadest assessment of the benefits of s. 33.1 to society, weighed against the cost of the limitations to ss. 7 and 11(d) of the Charter. With respect to its salutary effects, s. 33.1 gives expression to the close and harmful association between extreme self-induced intoxication and violence and affirms society's commitment to the equality and security rights of victims vulnerable to intoxicated crime. It responds meaningfully to inequality by recognizing that women and children deserve the full protection of the law and by condemning intoxicated gendered and family violence. It includes in its reach the irresponsible use and mixing of intoxicants that could lead to automatism and violence which discourages such behaviour and raises awareness about the link between extreme intoxication and violence. It contributes to public confidence in the criminal justice system, although this benefit must be balanced against recognizing society's interests in a system of law governed by the principles of fundamental justice. As well, it fosters personal responsibility in respect of voluntary intoxication, which Parliament saw as one of the root sources of violent crime.

However, s. 33.1's deleterious effects are serious and troubling. Its fundamental flaw is the risk of wrongful convictions it presents. It contravenes virtually all the criminal law principles that the law relies upon to protect the morally innocent. It enables conviction where the accused acted involuntarily, where the accused did not possess the minimum level of fault required, and where the Crown has not proven beyond a reasonable doubt the essential elements of the offence for which an accused is charged. Because s. 33.1 does not build in a criterion of objective foreseeability, it is impossible to say who, among those who voluntarily ingest intoxicants, has the degree of blameworthiness that would justify the stigma and punishment associated with the underlying offence with which they are charged. Where the intoxicant is licit, or where no reasonable person would anticipate the risk of automatism, whatever blameworthiness that comes from voluntary intoxication is relatively low and likely disproportionate to the punishment the individual would face if convicted for an offence committed in a state akin to automatism. It cannot be concluded that the morally innocent will not be punished. This is an extremely serious deleterious effect. Additionally, s. 33.1 disproportionately punishes for unintentional harm, contrary to the principle that punishment must be proportionate to the gravity of the offence.

The Crown has not discharged its burden of showing that the benefits suggested by the evidence are fairly realized by s. 33.1. There are socially and constitutionally acceptable alternatives to the Daviault exception that achieve the legitimate objectives of the law more fairly than in s. 33.1. In the absence of s. 33.1, the benefits tied to accountability and protection will continue to be met through the application of common law rules which prevent the defence of intoxication including to general intent crimes of violence. Parliament can further advance these goals with respect to self‑induced extreme intoxication akin to automatism through other means. The weight to be accorded to the principles of fundamental justice and the presumption of innocence cannot be ignored. Section 33.1 trenches on fundamental principles at the core of Canada's criminal law system, creates a liability regime that disregards principles meant to protect the innocent, and communicates the message that securing a conviction is more important than respecting the basic principles of justice. Its impact on the principles of fundamental justice is disproportionate to its overarching public benefits. It should therefore be declared unconstitutional and of no force or effect.

Cases Cited

Applied: R. v. Oakes, [1986] 1 S.C.R. 103; considered: R. v. Sullivan, 2020 ONCA 333, 151 O.R. (3d) 353; R. v. Daviault, [1994] 3 S.C.R. 63; referred to: R. v. Stone, [1999] 2 S.C.R. 290; Rabey v. The Queen, [1980] 2 S.C.R. 513; R. v. Luedecke, 2008 ONCA 716, 93 O.R. (3d) 89; R. v. Sullivan, 2022 SCC 19, [2022] 1 S.C.R. 460; Re B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486; R. v. Creighton, [1993] 3 S.C.R. 3; Leary v. The Queen, [1978] 1 S.C.R. 29; R. v. Bernard, [1988] 2 S.C.R. 833; Director of Public Prosecutions v. Beard, [1920] A.C. 479; R. v. Chaulk, [1990] 3 S.C.R. 1303; R. v. Parks, [1992] 2 S.C.R. 871; Bratty v. Attorney‑General for Northern Ireland, [1963] A.C. 386; R. v. Ruzic, 2001 SCC 24, [2001] 1 S.C.R. 687; R. v. Bouchard‑Lebrun, 2011 SCC 58, [2011] 3 S.C.R. 575; R. v. Théroux, [1993] 2 S.C.R. 5; R. v. Vaillancourt, [1987] 2 S.C.R. 636; R. v. Hundal, [1993] 1 S.C.R. 867; R. v. Roy, 2012 SCC 26, [2012] 2 S.C.R. 60; R. v. Penno, [1990] 2 S.C.R. 865; R. v. DeSousa, [1992] 2 S.C.R. 944; Canada (Attorney General) v. Bedford, 2013 SCC 72, [2013] 3 S.C.R. 1101; Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331; Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835; R. v. Mills, [1999] 3 S.C.R. 668; R. v. Malmo‑Levine, 2003 SCC 74, [2003] 3 S.C.R. 571; R. v. George, 2017 SCC 38, [2017] 1 S.C.R. 1021; R. v. Levigne, 2010 SCC 25, [2010] 2 S.C.R. 3; R. v. Chaulk, 2007 NSCA 84, 257 N.S.R. (2d) 99; R. v. Morrison, 2019 SCC 15, [2019] 2 S.C.R. 3; R. v. Vickberg (1998), 16 C.R. (5th) 164; Ontario v. Canadian Pacific Ltd., [1995] 2 S.C.R. 1031; R. v. Beatty, 2008 SCC 5, [2008] 1 S.C.R. 49; R. v. Cooper, [1993] 1 S.C.R. 146; R. v. K.R.J., 2016 SCC 31, [2016] 1 S.C.R. 906; Frank v. Canada (Attorney General), 2019 SCC 1, [2019] 1 S.C.R. 3; Alberta v. Hutterian Brethren of Wilson Colony, 2009 SCC 37, [2009] 2 S.C.R. 567; R. v. Robinson, [1996] 1 S.C.R. 683; RJR‑MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; Quebec (Attorney General) v. A, 2013 SCC 5, [2013] 1 S.C.R. 61; R. v. Martineau, [1990] 2 S.C.R. 633; R. v. Dunn (1999), 28 C.R. (5th) 295; Rodriguez v. British Columbia (Attorney General), [1993] 3 S.C.R. 519; R. v. Brenton (1999), 180 D.L.R. (4th) 314; R. v. Chan, 2018 ONSC 3849, 365 C.C.C. (3d) 376; R. v. Stevens, [1988] 1 S.C.R. 1153; R. v. Hess, [1990] 2 S.C.R. 906.

Statutes and Regulations Cited

Act to amend the Criminal Code (self‑induced intoxication), S.C. 1995, c. 32, preamble.

Canadian Charter of Rights and Freedoms, ss. 1, 7, 11, 15, 28.

Constitution Act, 1982, s. 52.

Criminal Code, R.S.C. 1985, c. C‑46, Part I, s. 33.1.

Authors Cited

Baker, Dennis, and Rainer Knopff. "Daviault Dialogue: The Strange Journey of Canada's Intoxication Defence" (2014), 19 Rev. Const. Stud. 35.

Canada. Department of Justice. Self‑Induced Intoxication as Criminal Fault: Information Note. Ottawa, 1995.

Canada. House of Commons. House of Commons Debates, vol. 133, No. 177, 1st Sess., 35th Parl., March 27, 1995, pp. 11037‑39.

Canada. House of Commons. House of Commons Debates, vol. 133, No. 224, 1st Sess., 35th Parl., June 22, 1995, p. 14470.

Canada. House of Commons. Standing Committee on Justice and Legal Affairs. Evidence, No. 161, 1st Sess., 35th Parl., June 13, 1995, pp. 22‑25.

Canada. House of Commons. Standing Committee on Justice and Legal Affairs. Minutes of Proceedings and Evidence of the Standing Committee on Justice and Legal Affairs, No. 98, 1st Sess., 35th Parl., April 6, 1995, pp. 6, 17.

Chapman, Frances E. "Sullivan. Specific and General Intent be Damned: Volition Missing and Mens Rea Incomplete" (2020), 63 C.R. (7th) 164.

Coughlan, Steve. "Sullivan: Can a Section 7 Violation Ever be Saved Under Section 1?" (2020), 63 C.R. (7th) 157.

Dimock, Susan. "Actio Libera in Causa" (2013), 7 Crim. Law and Philos. 549.

Ferguson, Gerry. "The Intoxication Defence: Constitutionally Impaired and in Need of Rehabilitation" (2012), 57 S.C.L.R. (2d) 111.

Gardner, John. Offences and Defences: Selected Essays in the Philosophy of Criminal Law. New York: Oxford University Press, 2007.

Grant, Isabel. "Second Chances: Bill C‑72 and the Charter" (1995), 33 Osgoode Hall L.J. 379.

Grant, Isabel. "The Limits of Daviault" (1995), 33 C.R. (4th) 277.

Healy, Patrick. "Criminal Reports Forum on Daviault: Extreme Intoxication Akin to Automatism Defence to Sexual Assault — Another Round on Intoxication" (1995), 33 C.R. (4th) 269.

Healy, Patrick. "Intoxication in the Codification of Canadian Criminal Law" (1994), 73 Can. Bar Rev. 515.

Hogg, Peter W., and Allison A. Bushell. "The Charter Dialogue Between Courts and Legislatures (Or Perhaps The Charter of Rights Isn't Such A Bad Thing After All)" (1997), 35 Osgoode Hall L.J. 75.

Kelly, Lisa M., and Nadya Gill. The punishing response to the defence of extreme intoxication, October 13, 2020.

Lawrence, Michelle S. "Voluntary Intoxication and the Charter: Revisiting the Constitutionality of Section 33.1 of the Criminal Code" (2017), 40:3 Man. L.J. 391.

Parent, Hugues. "La constitutionnalité de l'article 33.1 du Code criminel: analyse et commentaires" (2022), 26 Can. Crim. L.R. 175.

Plaxton, Michael, and Carissima Mathen. "What's Right With Section 33.1" (2021), 25 Can. Crim. L.R. 255.

Quigley, Tim. "A Time for Parliament to Enact an Offence of Dangerous Incapacitation" (1995), 33 C.R. (4th) 283.

Roy, Simon. "Intoxication", dans JurisClasseur Québec — Collection Droit pénal — Droit pénal général, par Marie‑Pierre Robert et Simon Roy, dir. Montréal: LexisNexis, 2013, fascicule 13 (mis à jour 7 juillet 2020).

Shaffer, Martha. "R. v. Daviault: A Principled Approach To Drunkenness or A Lapse of Common Sense?" (1996), 3 Rev. Const. Stud. 311.

Sheehy, Elizabeth. "The intoxication defense in Canada: why women should care" (1996), 23 Contemp. Drug Probs. 595.

Silver, Lisa. Who is Responsible for Extreme Intoxication?, October 7, 2021.

Stuart, Don. "Parliament Should Declare a New Responsibility for Drunkenness Based on Criminal Negligence" (1995), 33 C.R. (4th) 289.

Tremblay, Mathilde. "Charte canadienne et intoxication volontaire: l'article 33.1 du Code criminel et ses solutions de rechange" (2020), 79 R. du B. 67.

United Kingdom. Law Commission. Criminal Liability: Insanity and Automatism — A Discussion Paper. London, 2013.

Appeal

APPEAL from a judgment of the Alberta Court of Appeal (Slatter, Khullar and Hughes JJ.A.), 2021 ABCA 273, 30 Alta L.R. (7th) 1, 404 C.C.C. (3d) 311, [2021] 11 W.W.R. 191, [2021] A.J. No. 1028 (QL), 2021 CarswellAlta 1808 (WL), setting aside a decision of Hollins J., 2020 ABQB 166, 9 Alta. L.R. (7th) 375, [2020] A.J. No. 294 (QL), 2020 CarswellAlta 442 (WL). Appeal allowed.

Counsel

Sean Fagan and Michelle Biddulph, for the appellant.

Deborah J. Alford, for the respondent.

Michael H. Morris, Roy Lee and Rebecca Sewell, for the intervener the Attorney General of Canada.

Michael Perlin and Jeffrey Wyngaarden, for the intervener the Attorney General of Ontario.

Ami Kotler, for the intervener the Attorney General of Manitoba.

Lara Vizsolyi, for the intervener the Attorney General of British Columbia.

Noah Wernikowski, for the intervener the Attorney General of Saskatchewan.

Anil K. Kapoor and Dana Achtemichuk, for the intervener the Canadian Civil Liberties Association.

Carter Martell, Anita Szigeti, Sarah Rankin and Maya Kotob, for the intervener the Empowerment Council.

Lindsay Daviau and Eric Neubauer, for the intervener the Criminal Lawyers' Association.

Lara Kinkartz and Megan Stephens, for the intervener the Women's Legal Education and Action Fund Inc.

Reasons for Judgment

The judgment of the Court was delivered by

Kasirer J. —

I. Overview

[ 1 ] Following a party at which he had consumed alcohol and "magic mushrooms", Matthew Winston Brown violently attacked Janet Hamnett, a person he did not know and who had done nothing to invite the assault. At the time, Mr. Brown was in what the trial judge described as a "substance intoxication delirium" that was so extreme as to be "akin to automatism" (2020 ABQB 166, 9 Alta. L.R. (7th) 375, at para. 87). While capable of physical movement, he was in a delusional state and had no willed control over his actions. Mr. Brown's extreme intoxication akin to automatism was brought about by his voluntary ingestion of the magic mushrooms which contained a drug called psilocybin. Mr. Brown was acquitted at trial. The Alberta Court of Appeal set aside that verdict and convicted him of the general intent offence of aggravated assault.

[ 2 ] At common law, automatism is "a state of impaired consciousness, rather than unconsciousness, in which an individual, though capable of action, has no voluntary control over that action" (R. v. Stone, [1999] 2 S.C.R. 290, at para. 156). It is sometimes said that the effect of automatism is to provoke physical involuntariness whereby there is no connection between mind and body (see Rabey v. The Queen, [1980] 2 S.C.R. 513, at p. 518). Examples often given include the involuntary physical movement of an individual who has suffered a heart attack or seizure. Conduct that is involuntary in this sense cannot be criminal (see R. v. Luedecke, 2008 ONCA 716, 93 O.R. (3d) 89, at paras. 53‑56, relying in particular on Rabey, at p. 519, per Ritchie J., and at p. 545, per Dickson J., as he then was, dissenting but not on this point).

[ 3 ] Mr. Brown's appeal before this Court turns on the circumstances in which persons accused of certain violent crimes can invoke self‑induced extreme intoxication to show that they lacked the general intent or voluntariness ordinarily required to justify a conviction and punishment. Similar matters are at the heart of the Crown appeals in R. v. Sullivan and R. v. Chan, for which judgments are rendered simultaneously with this case (R. v. Sullivan, 2022 SCC 19, [2022] 1 S.C.R. 460) (the "Sullivan and Chan appeals"). The Court is asked in all three cases to decide upon the constitutionality of An Act to amend the Criminal Code (self‑induced intoxication), S.C. 1995, c. 32 ("Bill C‑72"), in light of, on the one hand, the principles of fundamental justice and the presumption of innocence guaranteed to the accused by ss. 7 and 11(d) of the Canadian Charter of Rights and Freedoms and, on the other, Parliament's aims to protect victims of intoxicated violence, in particular women and children, and hold perpetrators to account.

[ 4 ] These are not drunkenness cases. The accused in each of these appeals consumed drugs which, they argued, taken alone or in combination with alcohol, provoked psychotic, delusional and involuntary conduct, which are reactions not generally associated with drunkenness. As I note below, there is good reason to believe Parliament understood that alcohol alone is unlikely to bring about the delusional state akin to automatism it sought to regulate in enacting s. 33.1 of the Criminal Code, R.S.C. 1985, c. C‑46. As Lauwers J.A. wrote in R. v. Sullivan, 2020 ONCA 333, 151 O.R. (3d) 353, "it is not clear that extreme alcohol intoxication causes non-mental disorder automatism as a matter of basic science" (para. 288). In any event, these reasons say nothing about criminal liability for violent conduct produced by alcohol alone short of the psychotic state akin to automatism experienced by Mr. Brown and spoken to by the trial judge. I specifically leave intact the common law rule that drunkenness, absent clear scientific evidence of automatism, is not a defence to general intent crimes, including crimes of violence such as sexual assault.

[ 5 ] It thus bears emphasizing that Mr. Brown was not simply drunk or high. To be plain: it is the law in Canada that intoxication short of automatism is not a defence to the kind of violent crime at issue here. The outcome of the constitutional questions in these appeals has no impact on the rule that intoxication short of automatism is not a defence to violent crimes of general intent in this country.

[ 6 ] Parliament added s. 33.1 largely in response to R. v. Daviault, [1994] 3 S.C.R. 63. In that case, the Court confirmed the common law rule that intoxication is not a defence to crimes of general intent. The majority in Daviault recognized, however, that the Charter mandated an exception to the common law rule: where intoxication is so extreme that an accused falls into a condition akin to automatism, a conviction for the offence charged would violate ss. 7 and 11(d) of the Charter. It would be unfair, reasoned the Court, to hold an individual responsible for crimes committed while in a state of automatism, as they are incapable of voluntarily committing a guilty act or of having a guilty mind.

[ 7 ] Crown counsel in this appeal and the Sullivan and Chan appeals recall that the Daviault exception was met with public incomprehension and disapproval. In dissent, Sopinka J. anticipated this grievance when he wrote that those who voluntarily render themselves intoxicated and then violently cause bodily harm to others are "far from blameless" (p. 128). In order to address the constitutional failings identified by the majority of the Court in a manner that would properly reflect the blameworthiness of the extremely self‑intoxicated accused identified by the dissent, Parliament enacted s. 33.1. The new provision purported to remove the defence of automatism for the extremely self-intoxicated accused and put in place a constitutionally‑compliant measure of criminal fault for the underlying violent offence. The Crown and the intervening attorneys general urge us to interpret s. 33.1 as validly imposing liability for violent crimes based on a standard of criminal negligence that would answer the violations of the Charter pointed to in Daviault.

[ 8 ] But the impugned provision of the Criminal Code does not establish a proper measure of criminal fault by reason of intoxication. Instead, s. 33.1 imposes liability for the violent offence if an accused interferes with the bodily integrity of another "while" in a state of self‑induced intoxication rendering them incapable of consciously controlling their behaviour. Section 33.1 treats extreme voluntary intoxication, foreseeable or otherwise, as a condition of liability for the underlying violent offence and not as a measure of fault based on criminal negligence.

[ 9 ] Accordingly, the accused risks conviction for the relevant general intent offence — in Mr. Brown's case, for aggravated assault — based on conduct that occurred while they are incapable of committing the guilty act (the actus reus) or of having the guilty mind (mens rea) required to justify conviction and punishment. They are not being held to account for their conduct undertaken as free agents, including the choice to ingest an intoxicant undertaken when neither the risk of automatism nor the risk of harm was necessarily foreseeable. Instead, the accused is called to answer for the general intent crime that they cannot voluntarily or wilfully commit, an offence for which the whole weight of the criminal law and ss. 7 and 11(d) say they may be morally innocent. To deprive a person of their liberty for that involuntary conduct committed in a state akin to automatism — conduct that cannot be criminal — violates the principles of fundamental justice in a system of criminal justice based on personal responsibility for one's actions. On its face, not only does the text of s. 33.1 fail to provide a constitutionally compliant fault for the underlying offence set out in its third paragraph, it creates what amounts to a crime of absolute liability.

[ 10 ] I hasten to say that there may well have been other paths for Parliament to achieve its legitimate aims connected to combatting extreme intoxicated violence. The sense that an accused who acts violently in a state of extreme self‑induced intoxication is morally blameworthy is by no means beyond the proper reach of the criminal law. Protecting the victims of violent crime — particularly in light of the equality and dignity interests of women and children who are vulnerable to intoxicated sexual and domestic violence — is a pressing and substantial social purpose. And as I shall endeavour to show, it was not impermissible for Parliament to enact legislation seeking to hold an extremely intoxicated person accountable for a violent crime when they chose to create the risk of harm by ingesting intoxicants.

[ 11 ] The alternatives to the constitutionally fragile s. 33.1 strike different balances between individual rights and societal interests and, no doubt, each has advantages and shortcomings as a matter of social policy. Some of these options would be manifestly fairer to the accused while achieving some, if not all, of Parliament's objectives. I am mindful that it is not the role of the courts to set social policy, much less draft legislation for Parliament, as courts are not institutionally designed for these tasks. But it is relevant to the analysis that follows that, as noted by the majority in Daviault itself (p. 100) and by the majority of the Court of Appeal in Sullivan (para. 132), it would likely be open to Parliament to establish a stand‑alone offence of criminal intoxication. Others, including the voir dire judge in this very case (2019 ABQB 770, at para. 80), have suggested liability for the underlying offence would be possible if the legal standard of criminal negligence required proof that both of the risks of a loss of control and of the harm that follows were reasonably foreseeable. In either of these ways, Parliament would be enacting a law rooted in a "moral instinct" that says a person who chooses to become extremely intoxicated does so at their own peril.

[ 12 ] Parliament did not enact a new offence of dangerous intoxication, nor did it adopt a new mode of liability for existing violent offences based on a proper standard of criminal negligence. With respect, Parliament fell between the two stools, enacting a law that does not respect the minimum fault standard required by the Constitution.

[ 13 ] The violations of the rights of the accused in respect of the principles of fundamental justice and the presumption of innocence occasioned by s. 33.1 are grave. Notwithstanding Parliament's laudable aims, s. 33.1 fails at the proportionality stage of the s. 1 Oakes analysis and accordingly cannot be justified in a free and democratic society.

[ 14 ] In the case of Mr. Brown, and on the strength of the findings of fact at trial, the conclusion may be plainly stated. Mr. Brown might well be reproached for choosing to drink alcohol and ingest magic mushrooms, the harmful and illegal nature of which he was aware, but he does not answer in criminal law for the violence he committed while in a state of extreme self‑induced intoxication akin to automatism. I would allow his appeal.

II. Background

[ 15 ] At a friend's house party on a January night in Calgary, Mr. Brown had six or seven mixed drinks, a few beers and consumed several one‑half gram or smaller portions of magic mushrooms. He was 26 years old and in his last year of university and was aware that psilocybin in magic mushrooms is an illegal drug that can bring about hallucinations. He had tried magic mushrooms once before and believed that they generally gave a "fuzzy but positive feeling" (trial reasons, at para. 38).

[ 16 ] As Mr. Brown would testify at his trial, at around 1:30 a.m. he felt "wonky" and began to "los[e] [his] grip on reality" (A.R., vol. V, at p. 13). Without any memory of having done so, Mr. Brown removed his clothing and left the house in an agitated state at around 3:45 a.m., running naked and barefoot into the cold winter night. His friends searched for him for about 10 to 15 minutes and then called the police.

[ 17 ] In a nearby house, Janet Hamnett was awoken around 4:00 a.m. by a loud noise. When she went to investigate, Ms. Hamnett was attacked by someone she did not know who she later described as a huge presence screaming at the top of his lungs. The intruder was Mr. Brown. Ms. Hamnett fell to the ground and put her arms up as he beat her repeatedly with a broken broom handle. With her head, face and arms covered in blood, she managed to get to a bathroom and lock the door. Mr. Brown left the house and continued into the street. When all appeared quiet Ms. Hamnett sought refuge at a neighbour's house, at which time the police were called. The attack left her with cuts and contusions, as well as broken bones in her right hand which resulted in permanent injuries. She also suffered psychological harm from the incident.

[ 18 ] At about 5:00 a.m., Mr. Brown broke into the Varshney residence a kilometer away by throwing a heavy object through the front door window. Mr. and Mrs. Varshney, who did not know Mr. Brown, heard screaming and the sound of breaking glass. They were able to take shelter in their bedroom and call the police. The police found Mr. Brown lying naked on the floor of a bathroom. He was whispering and appeared confused by his surroundings; his feet were visibly bruised and bloodied. Mr. Brown complied with police instructions and was taken for medical care. He recalled coming to in hospital then waking later in a jail cell. Mr. Brown later said he had no memory of what transpired at either of the two homes.

[ 19 ] Mr. Brown had no previous criminal record and no history of mental illness. He was charged with one count of breaking and entering Ms. Hamnett's home and committing the indictable offence of aggravated assault and one count of breaking and entering the Varshney home and committing the indictable offence of mischief to property over $5,000.

[ 20 ] At trial, Mr. Brown argued that he was not guilty of the offences charged by reason of automatism. He claimed to have been so impaired by the consumption of psilocybin that his actions were involuntary and that he did not have the necessary mens rea for conviction of aggravated assault or mischief to property. Expert evidence adduced at trial confirmed that that the psilocybin was the "clear causative factor" for what was described as the accused's delirium (trial reasons, at para. 73). On the basis of this evidence, Mr. Brown was said to have no voluntary control over his conduct at the time.

[ 21 ] The Crown invoked s. 33.1 as a means of precluding Mr. Brown from relying on self‑induced intoxication akin to automatism as a defence to the charge of aggravated assault. Mr. Brown answered that, insofar as it prevented him from raising automatism as a defence, s. 33.1 violated ss. 7 and 11(d) of the Charter and could not be saved by s. 1. He said that the defence should be available to him against both charges, including the offence relating to the aggravated assault to which s. 33.1 purportedly applied.

III. Proceedings Below

A. Alberta Court of Queen's Bench

(1) The Constitutional Ruling, 2019 ABQB 770 (deWit J.)

[ 22 ] In a judgment rendered following a voir dire, deWit J. concluded that s. 33.1 violated the principles of fundamental justice and the presumption of innocence guaranteed by the Charter and was not saved by s. 1. He declared it to be of no force and effect.

[ 23 ] The voir dire judge observed that s. 33.1 "does not deal with the consequence of criminal acts" but "simply eliminates any evidence and argument regarding the mens rea and voluntariness of the accused's actions" (2019 ABQB 770, at para. 44). He found breaches of ss. 7 and 11(d) of the Charter, specifically the principles of fundamental justice that the accused must have the required mens rea for conviction and that an accused must voluntarily commit the actus reus.

[ 24 ] He then held that these limits cannot be reasonably justified in a free and democratic society. The judge did recognize that the provision had pressing and substantial objectives relevant to s. 1, namely protection of victims of intoxicated violence, particularly women and children, and holding those who get severely intoxicated and commit crimes accountable for those crimes. He then found, however, that it did not satisfy the requirements of minimal impairment or proportionality.

[ 25 ] The voir dire judge declared s. 33.1 to be of no force and effect pursuant to s. 52(1) of the Constitution Act, 1982. As a result, Mr. Brown was entitled to raise the defence of extreme intoxication akin to automatism at trial.

(2) Reasons for Judgment on the Merits, 2020 ABQB 166, 9 Alta. L.R. (7th) 375 (Hollins J.)

[ 26 ] At trial, Mr. Brown led evidence in support of the defence of extreme intoxication akin to automatism. Hollins J. held that every material piece of evidence supported a finding of automatism.

[ 27 ] The trial judge found that Mr. Brown was in a state of delirium due to his consumption of psilocybin, "which meant that he was not acting voluntarily in the commission of these offences nor with the requisite intent" (para. 91). Expert evidence confirmed the psilocybin was the clear causative factor for the delirium.

[ 28 ] Hollins J. accepted these conclusions as well as the evidence of other witnesses who testified, including the victims, who were all credible in her view. The defence was an answer to the properly laid charges and both acquittals were entered.

B. Court of Appeal of Alberta, 2021 ABCA 273, 30 Alta. L.R. (7th) 1 (Slatter, Khullar and Hughes JJ.A.)

[ 29 ] In separate opinions written by Slatter, Khullar and Hughes JJ.A., the Court of Appeal reversed deWit J.'s declaration that s. 33.1 was of no force or effect. The court set aside the acquittal on the count of break and enter and aggravated assault and entered a conviction for that count.

[ 30 ] Slatter and Hughes JJ.A. both held that the voir dire judge had erred in concluding that s. 33.1 violated ss. 7 and 11(d) of the Charter.

[ 31 ] For Slatter J.A., the Supreme Court had "expressly invited Parliament to do exactly what it did, namely legislate to fill the gap created by Daviault, contemplating that the result would comply with the Charter" (para. 26). He wrote that it would be "anomalous" to insist that s. 33.1 provide a criminal negligence standard when Parliament was directed by this Court to legislate in such a way as to address the rights of victims of violent crime. In his view, s. 33.1 was consistent with ss. 7 and 11(d) of the Charter.

[ 32 ] In the alternative, wrote Slatter J.A., s. 33.1 would be saved under s. 1. Protecting citizens from violent crimes and holding violent citizens accountable, as specific purposes, can be acknowledged as pressing and substantial. He held as well that s. 33.1 is proportionate to those purposes, in part because the risk of a wrongful conviction is low given the rarity of the extreme intoxication defence.

[ 33 ] In concurring reasons, Hughes J.A. wrote that s. 33.1 requires a measure of fault that reflects a marked departure from the standard of care of a reasonable person. No substitution breach arose because the subjective intention to become intoxicated is a sufficient substitute for the subjective intention in the underlying offence.

[ 34 ] In her reasons concurring in the result, Khullar J.A. found breaches of ss. 7 and 11(d) but decided that the provision could be upheld under s. 1. In respect of the prima facie breach, she relied on and agreed with the majority of the Court of Appeal in Sullivan (Lauwers and Fairburn JJ.A.), who found that s. 33.1 violates ss. 7 and 11(d) of the Charter.

[ 35 ] Khullar J.A. concluded however that the provision could be saved under s. 1. She stated that the justification analysis was a "hard and close case" but that Parliament's choice was a defensible one. The provision was minimally impairing because there was no less harmful means available to achieve Parliament's objectives of protection and accountability.

[ 36 ] For Khullar J.A., Parliament had pressing and substantial objectives relating to accountability and protection in enacting s. 33.1. She disagreed with the view, expressed by the majority in Sullivan, that accountability for the choice to self-intoxicate is not a pressing and substantial objective. Rather, it was Parliament's aim to hold offenders answerable in a full measure for the choice to create the conditions for this specific type of intoxicated violence.

[ 37 ] Khullar J.A. observed that the provision might have been drafted using the modified objective test described by the voir dire judge which would have been "less problematic" (para. 197). Parliament was entitled, however, to make reasonable choices about how to balance its objectives with rights-respecting protections, and it was not for the courts to require Parliament to adopt the least intrusive means.

[ 38 ] At the final balancing stage, Khullar J.A. acknowledged that there are "serious and troubling negative effects of s. 33.1" in that it allows conviction for violent crimes even where the conduct was involuntary and the accused lacked even a minimal level of fault (para. 207). However, for Khullar J.A., these deleterious effects were outweighed by the beneficial effects of the provision.

[ 39 ] In conclusion, Khullar J.A. adopted Slatter J.A.'s disposition of the matter. The Court entered a conviction for aggravated assault.

IV. Issue

[ 40 ] The only issue is whether s. 33.1 violates ss. 7 and 11(d) of the Charter and, if so, whether it can be saved under s. 1.

[ 41 ] Section 33.1 provides:

33.1 (1) It is not a defence to an offence referred to in subsection (3) that the accused, by reason of self-induced intoxication, lacked the general intent or the voluntariness required to commit the offence, where the accused departed markedly from the standard of care as described in subsection (2).

(2) For the purposes of this section, a person departs markedly from the standard of reasonable care generally recognized in Canadian society and is thereby criminally at fault where the person, while in a state of self-induced intoxication that renders the person unaware of, or incapable of consciously controlling, their behaviour, voluntarily or involuntarily interferes or threatens to interfere with the bodily integrity of another person.

(3) This section applies in respect of an offence under this Act or any other Act of Parliament that includes as an element an assault or any other interference or threat of interference by a person with the bodily integrity of another person.

V. Analysis

A. Intoxication and Criminal Liability

[ 42 ] While Mr. Brown was convicted of aggravated assault, it is of central importance to recall that s. 33.1 blocks the defence of automatism for all general intent crimes designated in s. 33.1(3), including sexual assault. The violent and non-consensual character of the crime in cases of intoxicated violence against women is evident and deserves emphasis. It is for this reason that the equality interests of vulnerable groups who are the victims of intoxicated sexual and domestic violence are implicated in the constitutional analysis.

[ 43 ] The common law has developed an unsympathetic view towards offenders who argue that their intoxication rendered them incapable of forming the necessary guilty mind. In principle, intoxication does not negate the mens rea of a general intent crime because self-intoxication is considered a reckless course of action that can be substituted for the fault required (see Leary v. The Queen, [1978] 1 S.C.R. 29; Director of Public Prosecutions v. Beard, [1920] A.C. 479). This is the Leary rule.

[ 44 ] The constitutionality of the Leary rule was upheld in Bernard. In concurring reasons, Wilson J. upheld the rule on the basis that the mens rea for sexual assault could be inferred from the commission of an obviously dangerous act. Intoxication to the point of automatism, however, was considered to be a qualitatively different state — tantamount to a state in which the mind was not operating — and was recognized by Wilson J. as one in which the general intent should not be substituted.

[ 45 ] It bears recalling, then, that most degrees of intoxication do not provide a defence to crimes of general intent like the offence of aggravated assault from which Mr. Brown was convicted on appeal. Only the highest form of intoxication — that which results in a person losing voluntary control of their actions — is at issue here: extreme intoxication akin to automatism as a defence to violent crimes of general intent and, then again only intoxication that is self-induced.

[ 46 ] The defence of automatism denies the element of voluntariness and therefore negates the actus reus of the offence (R. v. Chaulk, [1990] 3 S.C.R. 1303, at p. 1321; R. v. Parks, [1992] 2 S.C.R. 871, at p. 896). Involuntary conduct is understood to be genuinely exculpatory because, while the prohibited act was harmful, the accused lacks the capacity to answer for what they did (J. Gardner, Offences and Defences: Selected Essays in the Philosophy of Criminal Law (2007), at p. 142). A physically involuntary act, however wrongful in outward appearance, is not a guilty act that can be imputed to an accused.

[ 47 ] Automatism is reflected in involuntary movements that may be associated with heart attacks, seizures or "external" shock, or conditions such as sleepwalking or delirium, where the body moves but the mind is not directing the movements. In these situations, there is a disconnect between the mind and body.

[ 48 ] In addition, an automaton cannot form the mens rea, or guilty mind, if their actions are involuntary. Where an accused has no conscious awareness of their movements, they necessarily cannot intentionally commit a prohibited act. In most cases of automatism, there will be a "double acquittal" because the accused will simultaneously lack both the actus reus and the mens rea of the offence.

[ 49 ] I recall that, in Bernard, Wilson J. wrote that in a case of true intoxication akin to automatism, it may be improper to substitute proof of intention to become intoxicated for proof of intent to commit the offence: "[I]f the accused's intoxication has reached the point at which his mind no longer accompanies his physical movements, it is quite another matter. One must then consider whether the law can properly hold him responsible for those movements" (p. 880).

[ 50 ] I note that the defence has been referred to as "rare" in the case law (Daviault, at pp. 92‑93; Sullivan, at para. 118). The Attorney General of Manitoba disputes that and points to instances that have arisen since Daviault in which the defence was raised. The fact that the defence may rarely succeed does not diminish the importance of the constitutional questions at issue in this appeal.

(1) Daviault

[ 51 ] Disagreements concerning the blameworthiness of voluntary intoxication animated the majority and minority opinions in Daviault, which addressed squarely the impact of the Leary rule and the presumption against intoxication as a defence in the context of the Charter.

[ 52 ] Mr. Daviault sexually assaulted an acquaintance after drinking 7 or 8 bottles of beer and 35 ounces of brandy, an amount that would likely cause death or a coma in an ordinary person. The issue was whether he could avail himself of the defence of extreme intoxication akin to automatism.

[ 53 ] The majority was concerned that an accused in a state of extreme intoxication akin to automatism could be convicted for conduct that, by reason of its involuntary character, cannot amount to the actus reus or mens rea of a general intent crime. The majority concluded that the Charter requires an exception to the intoxication rule: an accused must be able to raise the defence of extreme intoxication akin to automatism when this would otherwise negate the existence of the general intent or the voluntariness required for conviction.

[ 54 ] The majority in Daviault thus modified the rule that intoxication is not a defence to crimes of general intent. Exceptionally, a defence of extreme intoxication akin to automatism, including self-induced extreme intoxication, was available. The accused bears the burden of establishing this on a balance of probabilities by expert evidence.

[ 55 ] In dissent, Sopinka J. said that self‑induced extreme intoxication cannot be a defence for general intent crimes such as sexual assault. In his view, the principles of fundamental justice could be satisfied by substituting for the mens rea of a general intent crime, the intent to become intoxicated and thereby create the risk to others of the harm that resulted.

[ 56 ] Thus, following Daviault, at common law, an accused was entitled to an acquittal for a general intent offence if they could prove, on a balance of probabilities, that they committed the acts in a state of extreme intoxication akin to automatism.

[ 57 ] In its aftermath, the majority opinion in Daviault was criticized for its "alarming lack of consideration of the social context of sexual assault particularly for women and children" (I. Grant, "The Limits of Daviault" (1995), 33 C.R. (4th) 277, at p. 277). The decision was harshly received by the public, particularly by women's advocacy groups.

(2) Bill C-72

[ 58 ] Within a relatively short period following the judgment of this Court in Daviault, Parliament added s. 33.1 to the general part (Part I — General) of the Criminal Code, which sought to abolish the defence of extreme intoxication akin to automatism for the general intent crimes listed in s. 33.1(3).

[ 59 ] Before the Standing Committee on Justice and Legal Affairs in advance of the Bill's enactment, the Minister of Justice stated that in Daviault, Sopinka J. "wrote a strong judgment for the dissent — one that was in keeping with Canadians' common sense notions of justice" (Evidence, No. 161, 1st Sess., 35th Parl., June 13, 1995, at p. 23).

[ 60 ] Even after Daviault, there is general agreement that a person acting in a state of automatism deserves an acquittal where their loss of control is truly morally blameless. But, to borrow a phrase from Don Stuart — one of the academic critics of Daviault — Parliament's thinking was that "[s]omeone who knowingly gets so drunk or stoned that he loses conscious control over his behaviour should be held responsible for his actions" ("Parliament Should Declare a New Responsibility for Drunkenness Based on Criminal Negligence" (1995), 33 C.R. (4th) 289, at p. 289). Parliament was motivated by the sense that extreme self-induced intoxication is morally blameworthy, particularly as it relates to violent crime.

[ 61 ] I take due note that the preamble to Bill C‑72 provides that, in most cases, extreme intoxication akin to automatism is brought about not through the consumption of alcohol alone, but by the combined consumption of alcohol and drugs.

[ 62 ] Although both Daviault and Parliament were focussed on "drunkenness", the parliamentary record and facts of this appeal and the Sullivan and Chan appeals suggest that the defence of extreme intoxication akin to automatism is not principally a "drunkenness" defence at all. As noted, alcohol alone is unlikely to induce the kind of psychosis-related state akin to automatism at issue in these appeals (Sullivan (2020), at para. 288). It is in this light that the preamble to Bill C‑72 is telling.

[ 63 ] I now turn to the question of whether s. 33.1 infringes ss. 7 and 11(d) of the Charter, as alleged by Mr. Brown.

B. Section 33.1 Infringes Sections 7 and 11(d)

(1) Principal Arguments

[ 64 ] Mr. Brown says s. 33.1 violates ss. 7 and 11(d) of the Charter. In its reading of the provision, the Court of Appeal erred in departing from the principles set forth in Daviault. He says that s. 33.1 does not displace the fault element required for conviction of the violent offence — as the majority of the Court of Appeal appears to have held — but eliminates the voluntariness and fault elements required to establish the actus reus and mens rea of the designated general intent crimes.

[ 65 ] The Crown responds that, when properly interpreted, s. 33.1 complies with the Charter. Under the Crown's reading, s. 33.1 only punishes the intentional and voluntary consumption of drugs to an extreme level, and a conviction under s. 33.1 requires a criminal negligence standard of fault because the provision requires a "marked departure" from the standard of care.

[ 66 ] The intervening attorneys general say that s. 33.1 has cast a unique mode of liability that provides the necessary ingredients for a constitutionally valid offence. The mental element inherent in s. 33.1 is the subjective choice to become self-intoxicated to the extreme degree specified in s. 33.1(2) and, they say, Parliament was entitled to make this choice a sufficient and proper substitute for the subjective intent required for the underlying violent offence.

(2) Threshold Issue: Internal Balancing Under Section 7

[ 67 ] As a preliminary matter, the Court must first decide whether the rights of victims of intoxicated violence, in particular the rights of women and children under ss. 7 and 15 of the Charter and the principle of substantive equality that informs the Charter as a whole, should be factored into the analysis of a possible breach of the accused's rights under s. 7.

[ 68 ] The intervener Women's Legal Education and Action Fund Inc. (LEAF) invites this Court to balance the rights of the accused against the rights of women and children in the s. 7 analysis. It says that the social harm caused by extreme intoxication should be considered at the s. 7 stage in determining whether the principles of fundamental justice have been violated.

[ 69 ] LEAF invokes Dagenais v. Canadian Broadcasting Corp., [1994] 3 S.C.R. 835, and R. v. Mills, [1999] 3 S.C.R. 668, in which this Court balanced competing Charter rights where the rights of the accused and another party conflict and are directly implicated by state action.

[ 70 ] In my view, the Dagenais and Mills mode of analysis does not apply and does not support the argument that balancing between the rights and interests of alleged perpetrators and victims of crime should occur at the breach stage under s. 7 of the Charter. This Court's balancing in Dagenais and Mills was engaged under different circumstances where the rights of the accused and the rights of an identified individual were directly implicated by state action. The equality, dignity and security interests of vulnerable groups, however important, are not rights engaged directly by the operation of s. 33.1 in the same way.

[ 71 ] Considering these as societal concerns under s. 1 does not "relegate" the equality, security and dignity interests of women and children to second order importance. LEAF is correct to say that courts must be attentive to the social context in which constitutional rights operate and be alive to the impact of gender discrimination, as the Court has said since Andrews (see, e.g., R. v. Malmo‑Levine, 2003 SCC 74, [2003] 3 S.C.R. 571, at para. 98; R. v. George, 2017 SCC 38, [2017] 1 S.C.R. 1021, at para. 44). These interests are indeed weighty and will be given their full measure in the s. 1 analysis below.

[ 72 ] Finally, and with due respect for other views, the basic values against arbitrariness, overbreadth and gross disproportionality are unrelated to the analysis of the Charter rights engaged in the equality and security of women and children. They are about the accused's liberty and security interests and the principles that protect against criminal liability in the absence of constitutionally required fault.

(3) Interpretation of Section 33.1

[ 73 ] Much of the argument in this appeal, as well as in the Sullivan and Chan appeals, turns on the proper interpretation of s. 33.1. Crown counsel and interveners in these appeals offer somewhat different readings of the provision.

[ 74 ] The Attorney General of Canada stated that s. 33.1 must be read in keeping with the presumption of constitutionality. The Attorney General argued that the fault targeted by s. 33.1(1) and (2) is not the accused's fault for committing the act of violence, but the fault of voluntarily becoming intoxicated to the extreme degree described in s. 33.1(2). The Crown says it follows from this that it is proper to use the intention to become extremely intoxicated as a substitute for the general intent required for conviction of the underlying violent offence.

[ 75 ] I see no ambiguity in s. 33.1 and disagree with the interpretation proposed by the Crown and the attorneys general which, with due respect for those who hold other views, falls afoul of the ordinary meaning of the text, the principles of statutory interpretation, and the settled jurisprudence of this Court on the provision.

[ 76 ] What does the impugned provision in fact say? Section 33.1(1) eliminates the defence of self‑induced intoxication akin to automatism applied to the violent offences identified in s. 33.1(3) where the accused departs markedly from the standard of care described in s. 33.1(2). Section 33.1 does not create a new predicate act offence of self-induced extreme intoxication or a new criminal negligence offence. The accused faces the full stigma of conviction and the full brunt of punishment for the general intent offence pointed to in s. 33.1(3).

[ 77 ] I agree with LeBel J. in Bouchard-Lebrun when he said that s. 33.1 applies when three conditions are met: (1) that the accused was intoxicated at the material time; (2) the intoxication was self‑induced; and (3) the accused departed markedly from the standard of reasonable care generally recognized in Canadian society by interfering or threatening to interfere with the bodily integrity of another person. When these three things are proved, it is not a defence that the accused lacked the general intent or the voluntariness required to commit the offence named in s. 33.1(3).

[ 78 ] The Crown is mistaken when it draws an analogy between impaired driving offences and s. 33.1. The gravamen of the offence faced by Mr. Brown does not include intoxication, unlike criminal offences such as operation of a motor vehicle while impaired by alcohol or drug, where the crime itself involves the condition of intoxication. Mr. Brown faces conviction for aggravated assault under s. 33.1 — not for an offence of criminal intoxication.

[ 79 ] The requirements of s. 33.1 — that the accused be intoxicated at the material time and the intoxication be self‑induced — are not, together or separately, a measure of fault. They are, as Bouchard-Lebrun describes them, "conditions of liability" (para. 44). This is confirmed by the use of the word "while" in s. 33.1(2): the violent act is committed while the accused is in a state of extreme intoxication, not because of the accused's prior intention to become intoxicated.

[ 80 ] Section 33.1(1) blocks the defence of automatism to general intent offences where the automatism was the result of self‑induced intoxication and the accused departed markedly from the standard of care described in s. 33.1(2). That standard is:

. . . the person, while in a state of self-induced intoxication that renders the person unaware of, or incapable of consciously controlling, their behaviour, voluntarily or involuntarily interferes or threatens to interfere with the bodily integrity of another person.

[ 81 ] In 1995, the Minister of Justice said in the House that "[w]e are stating in Bill C‑72 conclusively that intoxicating yourself to the point at which you lose conscious control and harm others is a marked departure from the standard of care which Canadians expect of themselves and of each other" (House of Commons Debates, vol. 133, No. 177, 1st Sess., 35th Parl., March 27, 1995, at p. 11037). This confirms what the text of s. 33.1 says: that the marked departure arises from being in a state of extreme intoxication while causing harm, not from a voluntary and foreseeable decision to become extremely intoxicated.

[ 82 ] I disagree with the view advanced by the Attorney General of Saskatchewan and others that the adjective "self-induced" must be read so that s. 33.1 carries with it a proper criminal negligence standard of fault. By definition, the term "self-induced" intoxication does not import a foreseeable risk of violence, a foreseeable loss of control, or even a foreseeable risk that one will become extremely intoxicated.

[ 83 ] Moreover, no plausible reading of the text suggests that self‑induced intoxication brings with it a reasonable foreseeability of bodily harm, as the voir dire judge rightly wrote in this case, "the marked departure standard is deemed present because of the violent act committed while the accused is in the state of extreme intoxication" (2019 ABQB 770, at para. 48).

[ 84 ] In oral argument in the Sullivan and Chan appeals, the Crown asserted that "nobody says that this is a pure deeming provision such that when this state of intoxication coincides with violence it is automatically met". The Crown says the word "while" in s. 33.1(2) properly creates a test of marked departure by requiring that the conditions in s. 33.1(2) be present at the same time as the violent act. But this reading of "while" amounts to exactly the kind of deeming provision the Crown says it is not.

[ 85 ] Contrary to the Crown's position, the "marked departure" standard of fault in s. 33.1(2) clearly attaches then to the violent offence, not the act of self‑induced intoxication. Neither can the Crown's argument that s. 33.1 creates a criminal negligence standard be reconciled with the words that directly follow "marked departure": the provision requires that the "person, while in a state of self-induced intoxication . . . voluntarily or involuntarily interferes or threatens to interfere with the bodily integrity of another person." The "marked departure" is the state of extreme intoxication during which violence occurs.

[ 86 ] The whole of the text confirms this. Section 33.1(1) distinguishes self‑induced intoxication from the prohibited offence, meaning the two cannot be the same. It provides that no defence is available for the prohibited offence on the basis of self‑induced intoxication. This is the logical opposite of making intoxication an element of the prohibited offence itself.

[ 87 ] Counsel was unable to cite a single case in the 25‑year history of s. 33.1 that adopted the interpretation proposed here by the Crown apart from the majority view of the Court of Appeal in this case. This is precisely because the text of s. 33.1 does not support it.

[ 88 ] Neither can this Court "read in" that interpretation, as the Crown suggested, by relying on the marginal notes accompanying the legislation and the presumption of constitutionality. To do so would be to rewrite the provision, not interpret it.

(4) Breaches of Sections 7 and 11(d) of the Charter

[ 89 ] Section 33.1 thus applies whenever a person interferes with the bodily integrity of another while in a state of extreme intoxication akin to automatism, regardless of whether a loss of awareness or control of their conduct was reasonably foreseeable when they voluntarily became intoxicated. This is the rule that Parliament wrote, and I must now determine whether this rule is consistent with the Charter.

(a) Mens Rea as Required by Section 7

[ 90 ] It is a principle of fundamental justice that proof of penal negligence, in the form of a marked departure from the standard of a reasonable person, is minimally required for a criminal conviction, unless the specific nature of the crime demands subjective fault (Re B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486; R. v. Creighton, [1993] 3 S.C.R. 3). Section 33.1 fails to meet this standard.

[ 91 ] Section 33.1 requires, for its proper application, an intention to become intoxicated. As noted, the term "self‑induced intoxication" has been interpreted to mean voluntarily ingesting a substance the accused knows or ought to know is intoxicating (R. v. Bouchard‑Lebrun, 2011 SCC 58, [2011] 3 S.C.R. 575, at para. 83). But intention to become intoxicated to any degree suffices. It matters little that a person did not foresee their loss of awareness or control, and nothing is said in s. 33.1 about the licit or illicit nature of the intoxicant or its known properties.

[ 92 ] For this reason, while the provision applies to those who recklessly invite their loss of control, it also captures the sudden and unexpected onset of involuntariness produced by "self-induced intoxication", such as an unexpected reaction to a prescribed pain medication or an unknown allergy. As Paciocco J.A. wrote in Sullivan, "you can be charged with no more fault than choosing to take a prescribed pain medication" (para. 92).

[ 93 ] Section 33.1 also imposes criminal liability where a person's intoxication carries no objective foreseeability of harm. Just as it draws no distinction based on the seriousness of the effects of the intoxicant, s. 33.1 makes no distinction based on whether, at the time of becoming intoxicated, a reasonable person would have foreseen the risk of harm to others.

[ 94 ] Additionally, even where an offence criminalizes an inherently dangerous activity, the trier of fact must not simply infer a marked departure from the standard of care (see, e.g., R. v. Beatty, 2008 SCC 5, [2008] 1 S.C.R. 49; R. v. Roy, 2012 SCC 26, [2012] 2 S.C.R. 60). A proper negligence-based fault standard requires asking whether a reasonable person would have foreseen the risk and taken steps to avoid it, and whether failure to do so represented a marked departure from the standard of care expected in the circumstances.

[ 95 ] Instead, s. 33.1 deems a person to have departed markedly from the standard of care expected in Canadian society whenever a violent act occurs while the person is in a state of extreme voluntary intoxication akin to automatism. Since s. 33.1 allows the court to convict an accused without proof of the constitutionally required mens rea, it violates s. 7 of the Charter.

(b) Voluntariness as Required by Section 7

[ 96 ] Section 33.1 also directs that an accused person is criminally responsible for their involuntary conduct. Because involuntariness negates the actus reus of the offence, involuntary conduct is not criminal, and the law recognizes that voluntariness for the conviction of a crime is a principle of fundamental justice (R. v. Ruzic, 2001 SCC 24, [2001] 1 S.C.R. 687, at paras. 40-46; R. v. Bouchard‑Lebrun, at paras. 54-56).

[ 97 ] There may be situations in which an accused should be answerable for their involuntary actions where they are to blame for the conditions that led to their involuntariness. In terms of physical involuntariness, the doctrine of actio libera in causa applies where the accused voluntarily placed themselves in a position of involuntariness with the intention of committing an offence. But as the voir dire judge rightly noted, this doctrine does not assist the Crown here because s. 33.1 does not require that the accused intended, or even foresaw, that they would commit the specific violent offence that resulted.

[ 98 ] It may be that the voluntariness problem could be avoided if Parliament legislated an offence of dangerous intoxication or intoxication causing harm that incorporates voluntary intoxication as the actus reus. But this is not what Parliament enacted in s. 33.1. Instead, Parliament tried to hold the accused responsible for the violent offence they committed while in a state of automatism.

(c) Substitution as Prohibited by Section 11(d)

[ 99 ] Section 11(d) of the Charter guarantees the accused's right to be presumed innocent until proven guilty. To convict the accused, the Crown must prove all the essential elements, including the requisite mens rea, beyond a reasonable doubt. A parliamentary direction that proof of one fact substitutes for proof of an essential element of the offence violates s. 11(d) unless, in all cases, proof of the substituted fact leads inexorably to proof of the element it replaces (R. v. Morrison, 2019 SCC 15, [2019] 2 S.C.R. 3, at paras. 130-140).

[ 100 ] Many critics of Daviault take the view that intoxication should aggravate, not excuse, the liability of someone who self‑intoxicates and causes injury to another, even if their actions are involuntary. Parliament took this view in enacting s. 33.1 and I have no doubt that many persons share it. But the question is whether Parliament properly implemented this policy choice in a manner consistent with the Charter.

[ 101 ] The Crown argues that s. 33.1 is not an instance of improper substitution, but instead a choice by Parliament to redefine the fault and voluntariness required for conviction of the underlying violent offence. I disagree.

[ 102 ] I disagree with the Crown. Mr. Brown is right to say that s. 33.1 improperly substitutes proof of self‑induced intoxication for proof of the essential elements of an offence, contrary to s. 11(d) of the Charter.

[ 103 ] As noted, s. 33.1 unequivocally removes a defence that the accused lacked the general intent or voluntariness to commit the offence. Accordingly, the fault and voluntariness of intoxication are substituted for the fault and voluntariness of the violent offence. This amounts to a constitutionally improper substitution. As the majority of the Court of Appeal in Sullivan held, it cannot be said that in all cases under s. 33.1, the intention to become intoxicated can be substituted for the intention to commit a violent offence.

[ 104 ] Section 33.1 fails the test in Morrison and amounts to a constitutionally improper substitution. While an accused who loses conscious control and assaults another person after a night of substance abuse could be described as intending to become intoxicated, that does not mean that the intention to become intoxicated necessarily leads to the further conclusion that the accused intended to commit — or that the accused possessed the minimum fault required to commit — the violent offence. The two elements are not inexorably linked.

[ 105 ] In sum, the effect of s. 33.1 is to invite conviction even where a reasonable doubt remains about the voluntariness or the fault required to prove the violent offence, contrary to the presumption of innocence guaranteed by s. 11(d) of the Charter.

(d) Contemporaneity

[ 106 ] As a final point, Mr. Brown asserts that s. 33.1 infringes s. 7 of the Charter because the violent offence occurs later in time than the intention to become intoxicated. Mr. Brown says this creates an absence of symmetry required by the fundamental justice principle that the fault element coincide with the act element of the offence.

[ 107 ] Symmetry differs from contemporaneity. Symmetry refers to knowledge or foreseeability of the precise consequences of the actus reus. For example, in Creighton, McLachlin J., as she then was, explained that while the Crown need not prove that the accused foresaw the precise consequence of death, the Crown must show that the accused could have foreseen the risk of bodily harm. The principle that the accused's fault must align with the prohibited conduct does not require that intent coincide precisely in time with the prohibited act.

[ 108 ] I thus agree with the conclusion of the voir dire judge and with Khullar J.A., who relied on the reasons of the majority of the Court of Appeal in Sullivan, that s. 33.1 violates ss. 7 and 11(d) of the Charter.

[ 109 ] I turn to a consideration of whether s. 33.1 can be saved under s. 1.

C. Justification Analysis

[ 110 ] The Crown must show on a balance of probabilities that the limits on ss. 7 and 11(d) brought by s. 33.1 are reasonable and demonstrably justified under s. 1 of the Charter (Oakes, at pp. 136‑37). The Oakes test is the framework for this analysis.

[ 111 ] Unlike in the appeal in Daviault, Parliament had before it a record related to the social problems associated with extreme intoxication and violence when enacting s. 33.1. The evidence highlighted the strong correlation between alcohol and drug use and violent offences, in particular against women, and brought to the fore of Parliament's attention the equality, dignity, and security rights of all victims of intoxicated violence.

[ 112 ] While s. 33.1 reflects broad Parliamentary pursuits relating to the common good, it also engages the traditional confrontation between the individual accused and the state in the context of a criminal prosecution. The burden of showing justification on a balance of probabilities rests with the Crown.

[ 113 ] Invoking the goals served by s. 33.1, the Crown submits that the Alberta Court of Appeal was right to decide the provision imposes reasonable limits and is justified under s. 1. As for Mr. Brown, he says the violations of ss. 7 and 11(d) of the Charter occasioned by s. 33.1 cannot be justified under s. 1.

[ 114 ] In my respectful view, Mr. Brown significantly understates the important goals pursued by Parliament in enacting s. 33.1. That said, given the patent risk that s. 33.1 may result in the conviction of an accused person who had no reason to believe that their voluntary intoxication would lead to a violent consequence, s. 33.1 fails at the proportionality step and thus cannot be saved under s. 1.

(1) Pressing and Substantial Purpose

[ 115 ] The parliamentary record, the preamble and, of course, s. 33.1 itself, all point to the two broad reasons why s. 33.1 was enacted in the period following Daviault: the protection of the victims of extremely intoxicated violence and a sense that the law should hold offenders accountable for the choice to create the conditions for that violence. These goals are easy enough to identify but must be stated with greater precision in a justification analysis.

[ 116 ] While these broad aspirations are easy enough to identify, the accused in this appeal and in the Sullivan and Chan appeals argue that the purposes of a provision must be described with greater precision when one examines whether a law that breaches the Charter is justified under s. 1. It is difficult for courts to evaluate whether measures that limit Charter rights are suitable to achieve their aims unless the aims are stated with some specificity.

[ 117 ] It is not therefore enough to say, as the Attorney General of Canada does here, that "protection for victims of intoxicated crime" and "holding perpetrators of intoxicated violence accountable" defines Parliament's purpose in enacting s. 33.1 in such a sufficiently precise a manner so as to allow for proper proportionality analysis.

[ 118 ] What, then, are the objectives of s. 33.1 and are they properly identified for conducting the justification exercise under s. 1?

[ 119 ] It is plain that s. 33.1, above all things, blocks the defence of automatism for the extremely intoxicated offender that was recognized in Daviault as an exception to the intoxication rules. Parliament did so with two specific purposes in mind. First, it sought to protect the victims of extremely intoxicated violence, with particular attention to women and children whose equal place in society is compromised by sexual assault and other violent crimes of general intent in such circumstances. Second, Parliament sought to call offenders to answer for their choice to voluntarily ingest intoxicants where that choice creates a risk of violent crime.

[ 120 ] I share the view that the protective purpose is sufficiently pressing and substantial to warrant limiting Charter rights. As stated by Lamer C.J. in R. v. Robinson, [1996] 1 S.C.R. 683, at para. 43: "There is no question that the protection of the public from intoxicated offenders is of sufficient importance to warrant overriding a constitutionally protected right or freedom."

[ 121 ] Paciocco J.A. explained in Sullivan why he viewed the accountability objective as impermissible and, as a result, cannot be considered as a "pressing and substantial objective" under the Oakes analysis. He wrote in part: "It cannot be that a preference for other values over constitutionally protected rights constitutes a pressing and substantial objective since that would risk laundering unconstitutional laws into constitutional ones . . . " (para. 118).

[ 122 ] Respectfully, I disagree. First, as I note below, Parliament did not "reject" the constitutional values spoken to in Daviault when it enacted s. 33.1. As the Minister stated in the House and as evidenced by the preamble to Bill C‑72, the law sought to respect both the rights of the accused and the rights of victims of violent crime.

[ 123 ] The objectives that moved Parliament to enact s. 33.1 in the wake of Daviault are not completely encapsulated by the single goal of protecting the victims of extremely intoxicated violent crime. For Parliament, s. 33.1 also sought to express a moral view, stated in the preamble, that a person who voluntarily takes an intoxicant to a dangerous level should be accountable for the acts of violence that result.

[ 124 ] The objective that Parliament sought to act upon is distinct from the protective purpose of the law. In fact, it rests on a philosophical idea that one should not be able to create the conditions of one's own criminal defence to block liability for the crime committed (see S. Dimock, "Actio Libera in Causa" (2013), 7 Crim. Law and Philos. 549).

[ 125 ] Perhaps the plainest demonstration that Parliament's objective cannot be limited to the protective goal is found in the explanation, by the Minister, as to why the stand‑alone offence was rejected as not meeting his goals. He accepted the view that the option of a new stand-alone offence of dangerous intoxication would go "part of the way" but that Bill C‑72 was necessary because it was thought there was a need to hold the perpetrators of intoxicated violence to account for the "real crime they committed, not just a lesser crime."

[ 126 ] This distinct and particularized accountability goal can serve as an objective for the purpose of the Oakes test in the unusual circumstances of this case. Here, the objective concerns the choice to create a risk, and this choice is not the conduct Parliament aims to criminalize. In other words, accountability is not Parliament's way of justifying liability without fault. It is rather Parliament's statement that fault in the choice to self-intoxicate is, in the unusual circumstances of s. 33.1, a sufficient basis on which to build criminal liability.

[ 127 ] To be clear, this conclusion rests on the specific concerns Parliament had when enacting s. 33.1. This case engages unusual issues and should not be seen as allowing governments to justify attempts to expand criminal liability as a routine matter. The accountability objective must, as here, rest on a choice to engage in conduct with a recognized inherent risk of causing harm.

(2) Proportionality

(a) Rational Connection

[ 128 ] At this stage, the Crown must show, first, that s. 33.1 is rationally connected to holding individuals accountable, in as full a manner as possible, for the choice to become extremely intoxicated and the violence committed while in that state and, second, that it is rationally connected to protecting victims of intoxicated violence.

[ 129 ] Mr. Brown focussed his criticism on the connection between s. 33.1 and the protection of vulnerable groups, as did Mr. Chan and Mr. Sullivan. In Mr. Brown's submission, the protective purpose relies on an unrealizable deterrent effect of s. 33.1 for its value. Since a person cannot necessarily foresee that voluntary intoxication will lead to extreme involuntariness, there is no meaningful deterrent value in s. 33.1.

[ 130 ] I agree with the Court of Appeal in this case that the deterrent and denunciating effects of s. 33.1 provide a rational connection to Parliament's protective objective.

[ 131 ] I recognize Paciocco J.A.'s criticism that s. 33.1 fails to offer meaningful deterrence in support of Parliament's protective purpose. In Sullivan, he wrote that "[e]ffective deterrence requires foresight . . . of the penal consequence" (para. 121). "I am not persuaded", he continued, "that s. 33.1 deters anyone" (ibid.).

[ 132 ] It is no doubt true that the deterrent effect of the provision would be more immediately felt if one could be sure that the accused understood the risk of violence associated with extreme self-induced intoxication and that this element is lacking in s. 33.1. However, in keeping with its moral purpose, Parliament could reasonably have expected that s. 33.1 would broadly deter individuals from becoming extremely intoxicated in circumstances where there is a risk of violence. Parliament could reasonably expect that the provision would have some deterrent effect for individuals who know they are susceptible to the loss of control associated with certain drugs.

[ 133 ] Courts should exercise caution before concluding that a measure in like circumstances is ineffective simply because they could imagine a way in which it would be more effective. To my mind, the Crown has met the rational connection test for Parliament's protective purpose.

[ 134 ] In addition, s. 33.1 is rationally connected to the objective of holding individuals accountable, in as full a manner as possible, for the choice to become extremely intoxicated and the violence committed while in that state. It is obvious that where a person is foreclosed from advancing a defence that could result in an acquittal, they are held accountable for the violent offence with which they are charged.

(b) Minimal Impairment

[ 135 ] The state must show that the impugned provision impairs rights as little as reasonably possible in furtherance of the legislative objective (RJR-MacDonald, at para. 160; Oakes, at p. 139). The law should only fail minimal impairment when there are less harmful means of achieving the objective that Parliament could have reasonably chosen (R. v. K.R.J., 2016 SCC 31, [2016] 1 S.C.R. 906, at para. 68).

[ 136 ] I have no hesitation imagining less impairing options. Many scholars have advanced options that would trench less on the rights of the accused (see, e.g., D. Stuart, "Parliament Should Declare a New Responsibility for Drunkenness Based on Criminal Negligence" (1995), 33 C.R. (4th) 289; T. Quigley, "A Time for Parliament to Enact an Offence of Dangerous Incapacitation" (1995), 33 C.R. (4th) 283; M. S. Lawrence, "Voluntary Intoxication and the Charter: Revisiting the Constitutionality of Section 33.1 of the Criminal Code" (2017), 40:3 Man. L.J. 391). The majority in Daviault itself identified the possibility of a stand-alone offence as a less intrusive means of achieving Parliament's aims.

[ 137 ] Apart from the stand‑alone offence, others have proposed alternative paths to liability for the underlying violent offence based on a criminal negligence standard more carefully crafted than that advanced by s. 33.1. One example is that proposed by the voir dire judge. He accepted Parliament's interest in holding accountable those who voluntarily ingest intoxicants but proposed a modified objective test that would require proof of objective foreseeability of both a loss of control and of the harm that resulted.

[ 138 ] In terms of the minimal impairment analysis, the stand-alone offence fails to meet Parliament's full objective and thus was not a viable alternative. It would have labelled Mr. Brown's offence as one of negligent or dangerous intoxication, rather than stigmatize him for the aggravated assault he committed. To my mind, Parliament was entitled to pursue the goal of making Mr. Brown answerable for the aggravated assault with which he was charged and not merely a lesser offence of criminal intoxication, and could have pursued this goal by other less impairing means.

[ 139 ] The alternative proposed by the voir dire judge could however allow an accused to be convicted for the underlying violent act and not simply negligent or dangerous intoxication. Incorporating a true marked departure standard into s. 33.1 would allow it to achieve the minimum objective fault standard required by the Constitution. This alternative, which Parliament was aware of through the voir dire judge's decision, was not incorporated into s. 33.1.

[ 140 ] I am aware that Parliament is entitled to deference in this analysis. Indeed, in crafting a new legislative response to the problem of intoxicated violence, it is up to Parliament to decide how to balance its objectives while also respecting Charter rights as much as possible (see, e.g., P. W. Hogg and A. A. Bushell, "The Charter Dialogue Between Courts and Legislatures (Or Perhaps The Charter of Rights Isn't Such A Bad Thing After All)" (1997), 35 Osgoode Hall L.J. 75). But the alternatives identified above show that s. 33.1 does not represent the least impairing option of achieving Parliament's aims.

[ 141 ] In light of these alternatives, in particular that proposed by the voir dire judge, which would have achieved Parliament's accountability objective in a real and substantial manner, I conclude that s. 33.1 is not minimally impairing. Parliament's objectives of protection and accountability were achievable by an approach that would have impaired the accused's Charter rights less.

[ 142 ] While I conclude that s. 33.1 is not minimally impairing of an accused's ss. 7 and 11(d) rights, I recognize that Parliament is entitled to a degree of deference in measuring the reasonable character of policy alternatives. But even if those who defend the law as minimally impairing were right, the analysis of the benefits and costs of s. 33.1 would confirm that it cannot be saved under s. 1.

(c) Proportionality Between Effects and Objectives

[ 143 ] At the final stage under s. 1, the question is whether there is proportionality between the overall effects of the Charter-infringing measure and the legislative objectives (Oakes, at p. 139; Hutterian Brethren, at paras. 72‑73). This invites the broadest assessment of the benefits of s. 33.1 to society, weighed against the cost of the limitations to ss. 7 and 11(d) of the Charter.

[ 144 ] Mr. Brown says that the Court of Appeal erred in minimizing the deleterious effects of s. 33.1 in that the provision would allow for wrongful convictions based on involuntary acts and the absence of even reasonable foreseeability of the consequences of those acts. In response, the Crown asks the Court to "give weight to Parliament's record which demonstrated a link between violence and extreme intoxication."

[ 145 ] In my view, Mr. Brown is right on this point. In the result, the meaningful benefits of s. 33.1 do not outweigh the costs, in particular to what the voir dire judge describes as the "sacrosanct principles" integral to our criminal justice system, including the presumption of innocence (para. 89).

(i) Salutary Effects

[ 146 ] By including general intent crimes of sexual and domestic violence within s. 33.1(3), the amendments to the Criminal Code do help to ensure the rights of women and children to equal protection and benefit of the law as guaranteed by the Charter, as promised in the preamble. The provision gives expression to the close and harmful association between extreme self-induced intoxication and violence and affirms society's commitment to the equality and security rights of victims vulnerable to intoxicated crime.

[ 147 ] The "recognition of the dignity and self‑worth of women and children" was well described by Khullar J.A., who wrote that this "breathes some meaning" into equality rights in the Charter, and properly took notice of this as a beneficial effect of s. 33.1 in the justification analysis (para. 207).

[ 148 ] Section 33.1 also provides societal benefit through its communicative and deterrent effects. Khullar J.A. recognized not only that s. 33.1 contributes to a social ethos of disapproval of extreme self‑induced intoxication, she wrote that it "serves the important role of signalling to people the risk that they undertake by choosing to become extremely intoxicated" (para. 207).

[ 149 ] In addition, as the Attorney General of Manitoba submits, s. 33.1 contributes to public confidence in the criminal justice system. In Creighton, McLachlin J. observed that it would "offend common notions of justice to acquit a person who has killed another of manslaughter and find him guilty of negligently handling a firearm" (p. 46). By analogy, it might offend common notions of justice to acquit a person who has violently attacked and injured another person because they were too intoxicated to have acted voluntarily.

[ 150 ] There is a manifest benefit to be connected with the accountability purpose of Parliament described above which, in my respectful view, the majority of the Court of Appeal in Sullivan should have considered as part of the full measure of the beneficial effects of s. 33.1 (see, on this point, Khullar J.A.'s reasons in the present case).

[ 151 ] At the end of the day, Parliament's own accountability objective was undone by the very means it chose to pursue it. In holding the extreme self‑intoxicated offender to account, s. 33.1 does not require objective foreseeability of the risk of falling into a state of automatism, much less the risk of violence. And when the objective is to hold people accountable for choices that create risks, the risk of the harm that follows must, at a minimum, be objectively foreseeable at the time of the choice.

(ii) Deleterious Effects

[ 152 ] The fundamental flaw of s. 33.1 is the risk of wrongful convictions it presents. By denying even a small fraction of accused persons the ability to raise a reasonable doubt as to the voluntariness or mens rea elements of the offence charged, s. 33.1 permits an individual to be convicted, and face full criminal punishment, for a violent offence they committed involuntarily, without the required fault, and without the Crown having proven beyond a reasonable doubt the essential elements of the offence.

[ 153 ] I agree with the courts below that the deleterious effects of s. 33.1 are serious and troubling. As Vertes J. put it in R. v. Brenton (1999), 180 D.L.R. (4th) 314 (N.W.T.S.C.), in denying the defence of automatism, s. 33.1 "casts aside the fundamental principle of absolute liability" (para. 68, emphasis added). As Wilson J. wrote in Bernard, there is something troubling about a legal system that punishes a person for conduct they could not help and could not have foreseen.

[ 154 ] It is not unfair to say that the narrow compass of s. 33.1 limits these negative effects. Section 33.1 only applies, as we have seen, to the violent offences of general intent named in s. 33.1(3). It is also not unfair to say that the burden of showing automatism presents a high hurdle for the accused. The defence must be supported by expert evidence — which may be difficult to obtain — and must establish the extreme state of automatism on a balance of probabilities.

[ 155 ] But even if its compass is narrow and its application rare, s. 33.1 limits not just one but three fundamental rights of the accused. It enables conviction where the accused acted involuntarily, where the accused did not possess the minimum level of fault required, and where the Crown has not proven beyond a reasonable doubt the essential elements of the offence for which an accused is charged. These are the very principles the criminal law uses to protect against punishing morally innocent people.

[ 156 ] I acknowledge that some, pointing in particular to Sopinka J.'s dissent in Daviault, have argued that s. 33.1 does not create a true offence of absolute liability since extreme voluntary intoxication can bring with it a degree of fault. Contrary to the person who falls into a state of involuntariness due to natural causes, the accused who voluntarily consumes intoxicants is morally blameworthy to some degree for the consequences that follow.

[ 157 ] Where the intoxicant is licit, or where no reasonable person would anticipate the risk of automatism, whatever blameworthiness that comes from voluntary intoxication is relatively low and likely disproportionate to the punishment the individual would face if convicted for an offence committed in a state akin to automatism. It cannot be concluded that the morally innocent will not be punished under s. 33.1. This is an extremely serious deleterious effect.

[ 158 ] One is hard pressed to disagree with the trial judge in R. v. Chan, 2018 ONSC 3849, 365 C.C.C. (3d) 376, who observed that "Parliament is entitled to express the view that extreme self‑intoxication is morally blameworthy behaviour" (para. 152). But I respectfully disagree with him where he concluded that this moral view could sustain the constitutional validity of s. 33.1 under s. 7.

[ 159 ] The voir dire judge in Mr. Brown's case was right to speak of the principles breached by s. 33.1 as "sacrosanct". In the Motor Vehicle Reference, Lamer J., as he then was, wrote that the principle that the innocent must not be punished "has long been recognized as an essential element of a system for the administration of criminal justice that is founded upon a belief in the dignity and worth of the human person" (p. 513).

[ 160 ] The idea that a conviction in the absence of the required proof of blameworthiness threatens a system of criminal justice founded on the dignity and worth of the human person has been a recurring theme in our jurisprudence (see, e.g., Oakes, at p. 136; R. v. Stevens, [1988] 1 S.C.R. 1153, at p. 1175; R. v. Vaillancourt, [1987] 2 S.C.R. 636, at pp. 654‑55; R. v. Martineau, [1990] 2 S.C.R. 633, at p. 645; Re B.C. Motor Vehicle Act, at p. 513).

[ 161 ] The voir dire judge rightly recognized that these principles exist to ensure that the morally innocent are not convicted (para. 89). Section 33.1 creates the risk that a person will be convicted of the underlying offence on the basis of proof of the blameworthiness associated with extreme intoxication, which blameworthiness may be low or non-existent in the circumstances of a given case.

[ 162 ] An additional deleterious effect is that s. 33.1 disproportionately punishes for unintentional harm, contrary to the principle that punishment be proportionate to the gravity of the offence. Section 33.1 requires the offender to face full responsibility of the underlying crime even though the mens rea for that crime was not proven. While s. 33.1 does not set a minimum punishment for the underlying offence, it exposes the accused to the full range of punishment, including maximum sentences, for violent offences they committed involuntarily and without the required fault.

[ 163 ] In saying this, I have not lost sight of the often-cited public outcry that followed Daviault which the Minister of Justice of the day said was one of the driving forces behind the enactment of s. 33.1. But public outcries do not in themselves justify unconstitutional laws. And to my mind, the "public confidence" benefit of s. 33.1 is weaker than the Crown asserts, because a genuinely innocent person ought not to be convicted in order to maintain public confidence in the justice system.

(iii) Weighing the Salutary and Deleterious Effects

[ 164 ] As this Court held in Bedford, at the final stage of the s. 1 analysis, the negative impact of the law is weighed against the beneficial impact of the law in terms of achieving its goal for the greater public good. The impacts are weighed both qualitatively and quantitatively. As with the proportionality analysis as a whole, the court must look at the actual public good achieved and not rely on speculation.

[ 165 ] In my respectful view, the Crown has not discharged its burden of showing that the benefits suggested by the evidence are fairly realized by s. 33.1. The Crown warns of widespread sexual and intimate partner violence, with the implication that such gendered violence will go undeterred in the absence of s. 33.1. But the Crown has not established that s. 33.1 operates as an effective deterrent to intoxicated violence against vulnerable groups, beyond a modest deterrent effect. Nor has the Crown established that, in the absence of s. 33.1, the victims of intoxicated violence would be deprived of legal protection, since it is settled law that intoxication short of automatism is not a defence to general intent crimes.

[ 166 ] The limits imposed on the most fundamental Charter rights in our system of criminal justice outweigh societal benefits that are already in part realized, and which Parliament can advance through other means. The weight to be accorded to the principles of fundamental justice and the presumption of innocence cannot be ignored. Section 33.1 trenches on fundamental principles at the core of Canada's criminal law system, creates a liability regime that disregards principles meant to protect the innocent, and communicates the message that securing a conviction is more important than respecting the basic principles of justice. Its impact on the principles of fundamental justice is disproportionate to its overarching public benefits. It should therefore be declared unconstitutional and of no force or effect.

VI. Conclusion

[ 167 ] I would answer the constitutional questions as follows: s. 33.1 of the Criminal Code infringes ss. 7 and 11(d) of the Charter and the infringements are not justified under s. 1 of the Charter. I would allow Mr. Brown's appeal. Section 33.1 should be declared unconstitutional and of no force or effect.

[ 168 ] The judgment of the Court of Appeal should be set aside. The acquittal entered by Hollins J. on the count of unlawful break and enter of a dwelling house and committing aggravated assault therein should be restored; the acquittal for unlawful break and enter of a dwelling house and committing mischief to property over $5,000 was unaffected by s. 33.1, was not appealed by the Crown, and need not be addressed.

Appeal allowed.

Solicitors

Solicitors for the appellant: Evans & Fagan, Calgary; Greenspan Humphrey Weinstein, Toronto.

Solicitor for the respondent: Justice and Solicitor General, Appeals, Education & Prosecution Policy Branch, Edmonton.

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

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 Saskatchewan: Attorney General of Saskatchewan, Regina.

Solicitors for the intervener the Canadian Civil Liberties Association: Kapoor Barristers, Toronto.

Solicitors for the intervener the Empowerment Council: Martell Defence, Toronto; Anita Szigeti Advocates, Toronto; McKay Ferg, Calgary.

Solicitors for the intervener the Criminal Lawyers' Association: Rosen & Company Barristers, Toronto; Neubauer Law, Oshawa.

Solicitors for the intervener the Women's Legal Education and Action Fund Inc.: WeirFoulds, Toronto; Megan Stephens Law, Toronto.