The Supreme Court of Canada (“SCC”) decided R v Brown, 2022 SCC 18 [Brown] and R v Sullivan, 2022 SCC 19 [Sullivan] together on May 13, 2022. Much attention was on the Court’s decision in Brown, which declared s. 33.1 of the Criminal Code, RSC 1985, c C-46 [“the Code”] unconstitutional under s. 52(1) of the Constitution Act, 1982. The struck-down provision prohibited an accused from invoking “automatism resulting from self-induced intoxication” as a legal defence in crimes requiring general intent. General intent offences require the prosecution to prove that the accused intended to bring about the criminal action, but not that they intended a specific outcome related to the action.
Following the SCC’s decision in Brown, in June 2022, Parliament enacted amendments to the Code to address the Court’s declaration that s. 33.1, as it stood at the time of its decision, was unconstitutional.
Parliament had initially enacted s. 33.1 to protect victims of violent crimes perpetrated by assailants under extreme states of intoxication, following the Court’s decision in R v Daviault,  3 SCR 63 [Daviault]. In Daviault, the majority of the Court held that to permit the mens rea requirement to be satisfied by substituting the accused’s intention to become intoxicated, for an intention to commit either specific or general intent crimes would violate their rights under ss. 7 and 11(d) of the Canadian Charter of Rights and Freedoms. Parliament responded by bringing s. 33.1 into force, which limited the scope of the Daviault defence, such that it would be unavailable for persons accused of crimes of general intent. This amendment to the Code generated a bevy of constitutional challenges, with courts across Canada divided in their conclusions, culminating in the SCC’s decision in Brown.
In the companion appeal in Sullivan, the SCC clarifies the legal effect of s. 52(1) declarations by a superior court. The Court stated that, with limited exceptions, a superior court decision will be binding on judges of the same court in the same province, following the principles of judicial comity and horizontal stare decisis. According to commentators, this unanimous decision is long overdue, given that, from 1999 onwards, four superior courts had struck down s. 33.1 in Ontario alone (Sullivan SCC, para 37).
Background facts and procedural history
The Court in Sullivan heard appeals of two accused persons, David Sullivan and Thomas Chan. Both had been charged and convicted for committing violent crimes while under the influence of intoxicants in unrelated circumstances. The respondents both argued that they could not be held criminally responsible for the violent offences requiring general intent due to their extreme state of intoxication.
Mr. Chan was charged with manslaughter and aggravated assault following an unexpected bad reaction to the “magic mushrooms” he had taken earlier with his friends. In his state of intoxication, Mr. Chan broke into his father’s home and assaulted him and his partner with a knife. Mr. Chan’s father was killed and his partner was critically injured. In a similarly tragic turn of events, Mr. Sullivan was charged with aggravated assault and assault with a weapon, after seemingly experiencing the effects of an episode of psychosis from an overdose of Wellbutrin, a drug he had been prescribed to help him quit smoking. In both cases, the trial judges rejected the defence of “involuntariness akin to automatism.”
Both cases were heard together by the Court of Appeal for Ontario, which declared s. 33.1 unconstitutional under s. 52(1) of the Constitution and permitted the use of an automatism defence, a defence available to the accused under common law. Since the trial judge had found that Mr. Sullivan’s actions resulted from non-mental disorder automatism, Justices Paciocco and Lauwers acquitted Mr. Sullivan. The same was not true for Mr. Chan, and so the Court of Appeal ordered a retrial in his case, during which he will be free to use the defence of non-mental disorder automatism.
Of note for the SCC’s discussion in Sullivan, counsel for Mr. Chan had unsuccessfully challenged the constitutionality of s. 33.1 at trial. They argued that the superior court judge was bound by the prior s. 33.1 decision of the same court, which had declared the provision unconstitutional. The Court of Appeal ruled that ordinary rules of stare decisis applied when superior courts consider whether to follow a previous s. 52(1) declaration of unconstitutionality and that the trial judge was correct to consider the question anew.
In its decision in Sullivan, the Court sought to clarify the following question: On what basis can a superior court issue a declaration under s. 52(1) of the Constitution be considered binding on courts of coordinate jurisdiction?
The Court’s decision
In its decision penned by Justice Kasirer, the SCC held that when a superior court of the same province has made a declaration of unconstitutionality, the ordinary rules of horizontal stare decisis and judicial comity apply. As such, judges of the same court are bound to follow the earlier decision, unless the earlier decision is factually distinct from the case in question or if the court had no practical way of knowing an earlier decision existed. The Court also noted that if there are conflicting binding decisions on the constitutionality of a provision, courts must follow the most recent authority.
However, even if a decision is binding based on the criteria stated above, the Court further held that a trial judge may still consider the question afresh if at least one of the exceptions stated in Re Hansard Spruce Mills,  4 D.L.R. 590 [Spruce Mills] applies. These exceptions are:
- “The rationale of the earlier decision has been undermined by subsequent appellate decisions;
- The earlier decision was reached per incuriam, i.e., ‘through carelessness’ or ‘by inadvertence’; or
- The earlier decision was not fully considered, e.g., taken in exigent circumstances” (Sullivan SCC, para 75).
The Court then clarified the Spruce Mills standard should be applied to determine when departure from precedent is appropriate. The Court noted the lack of uniformity in its current application, with courts in the past using qualitative standards like “plainly wrong,” “when there is good reason,” or “in extraordinary circumstances” (Sullivan SCC, para 74). The lack of precision and utility of these terms to guide judges in their decision-making led the Court to state that only the narrow Spruce Mills criteria should be applied, to avoid inconsistencies associated with such qualitative descriptors.
In its discussion on how the principles of stare decisis and judicial comity inform its analysis, the Court stated that “[t]he legal effect of a s. 52(1) declaration by a superior court must be defined with reference to constitutional supremacy, the rule of law, and federalism” (Sullivan SCC, para 60). The Court went on to say that when a declaration of unconstitutionality is considered with a holistic view of the fundamental principles of Canadian constitutionalism, it becomes clear that horizontal stare decisis cannot bind courts across provinces. Instead, absent narrow exceptions, the rules of horizontal stare decisis and judicial comity must bind courts of coordinate jurisdiction within a province, with vertical stare decisis binding lower courts after a constitutional ruling is made by an appellate court in the same province.
The arguments presented to the Court indicate a significant lack of consensus on the effect of s.52(1) rulings by superior courts in the absence of an appellate decision. This confusion about the precise effect of a court’s declaration of unconstitutionality seems to have been further exacerbated by specific phrases in the SCC’s decision in R v Ferguson,  1 SCR 96. This decision characterized such declarations as rendering the law “null and void” as the law is “effectively removed from the statute books” (Sullivan SCC, para 38). The several interveners in this case also pointed to rule of law values which are undermined, when a provision that has been declared of no force or effect under the Constitution is allowed to apply inconsistently both across and within provinces in the absence of an appellate decision on the matter. Given this state of affairs, the Court’s willingness to clarify the issue is welcome, and evidently, much needed.
The Court’s analysis in Sullivan is unsurprisingly driven by the need to balance “stability and predictability against correctness and the orderly development of the law” (Sullivan SCC, para 66). The discussion on how the Court’s decision derives its force from the need to take federalism seriously as a fundamental constitutional principle, along with the rule of law is deftly argued. Given the potentially confusing nature of the Court’s own language in prior decisions that characterize the effect of s. 52(1) declarations as “striking out” or “severing” the impugned provision from the statutory text, Sullivan serves to elucidate that such language is not to be interpreted literally. Instead, such declarations by courts can only render the provision in question inoperative—it is only Parliament that can enact legislative amendments to address a court’s s. 52(1) declaration of unconstitutionality.
The SCC also states in no uncertain terms that appealing s. 52(1) declarations for a final word on the question of constitutionality is at the discretion of attorneys general. This is despite the fact that, as the Court notes, the constitutional status of s. 33.1 has been unsettled since it was enacted over twenty years ago, partly due to the prosecution’s consistent failure to appeal. It is difficult to disagree with those who argue that federal legislation should not be left unsettled within a province, because the government refuses to appeal cases when it comes up against conflicting decisions. Nor is it hard to see how the same law being constitutional in one province but not in another could lead to uniformity issues.
Having said that, however undesirable such outcomes may be, the Court rightly insists that in the absence of abuse of authority, the relevant attorney general cannot be held to account by the courts on how they exercise discretion. While this leaves the possibility of the same controversy and lack of clarity that mired the constitutional status of s. 33.1 to reoccur with respect to a different provision in the future, the judiciary can only extend its reach so far before over-exerting its legitimacy and authority. As such, in Sullivan, the SCC may have provided the most effective holding at its disposal.