The Notwithstanding Clause as a Constitutional Safety Valve

This week, the Supreme Court of Canada heard argument in the appeal from the decision of the Quebec Court of Appeal in Organisation mondiale sikhe du Canada c. Procureur général du Québec, 2024 QCCA 254. One of the issues in this sprawling and complex case, is whether the Quebec legislature’s use of the notwithstanding clause in relation to Bill 21, which outlaws the wearing of religious symbols by public employees (and contractors engaged by the government), is constitutionally valid.

In this post, I want to comment on the general role of the notwithstanding clause within Canada’s constitutional structure and suggest that it may function as a safety valve not just to respond to problematic individual judicial decisions but rather systemically, to prevent Canada’s legal system from sliding into a vicious cycle of political disputes over constitutional adjudication. My goal here is not to assess the rightness or wrongness of any particular decision, but to adopt a broader perspective on this week’s debate over the notwithstanding clause. In particular, the question animating my discussion is “How should a legal system correct judicial decisions thought by the population to be ‘wrong’?”

My jumping-off point is the recent decision in Quebec (Attorney General) v. Senneville, 2025 SCC 33, where the judges once again invalidated a mandatory minimum sentence — this time for child-pornography offences — on the basis that such a sentence, imposed on a hypothetical offender in a reasonably foreseeable scenario, would be grossly disproportionate and thus constitute cruel and unusual treatment or punishment contrary to s. 12 of the Charter of Rights and Freedoms. The facts of the case before the Court were horrendous, involving adult men who had accessed images of children as young as three years old being sodomized. Yet, according to the Court’s jurisprudence, the actual circumstances of the offender are not relevant to the ultimate determination of whether a mandatory sentence would amount to cruel and unusual punishment. The backlash was swift. Numerous provincial politicians, as well as the federal Conservatives in opposition, called for use of the notwithstanding clause to reverse the effects of the judgment.

This is not the only recent call for — or use of — the notwithstanding clause. Last week, Alberta legislated a mandatory collective agreement on its striking teachers, invoking the clause to ensure the law could not be challenged before the courts. Ontario did something similar in 2022, though the legislation failed to achieve its intended effect, as unions relied on labour-relations mechanisms and the threat of civil disobedience to persuade the province to roll it back. Ontario has, however, floated the idea of using the notwithstanding clause to overturn the effects of a recent judgment declaring it unconstitutional for the province to legislate the removal of bike lanes in the City of Toronto: Cycle Toronto et al. v. Attorney General of Ontario et al., 2025 ONSC 4397. Schabas J held that such a measure would interfere with the security of the person of cyclists, particularly those who rely on bike lanes for work. In this, Ontario would be following in the footsteps of Saskatchewan, which invoked the clause in an attempt to shield legislation relating to gender issues from judicial scrutiny.

Before going any further, I will note that there is a ready legislative ‘fix’ for the mandatory minimums issue — including a provision that allows a sentencing judge to depart from the mandatory minimum where its effect would be grossly disproportionate would defang any invocation of s. 12. But is striking that each of the invocations of — or calls to use — the notwithstanding clause, has its origins in Supreme Court jurisprudence from the first couple of decades of this century. The mandatory-minimum jurisprudence kicked off by R. v. Smith (Edward Dewey), [1987] 1 SCR 1045 began in earnest with R. v. Ferguson, 2008 SCC 6, [2008] 1 SCR 96, and the use of hypothetical offenders was reaffirmed strongly in R. v. Nur, 2015 SCC 15, [2015] 1 SCR 773, both interpreting section 12 of the Charter, which protects against “cruel and unusual treatment or punishment.” The right to strike was recognized in Saskatchewan Federation of Labour v. Saskatchewan, 2015 SCC 4, [2015] 1 SCR 245, building on the recognition of a limited right to collective bargaining in Health Services and Support – Facilities Subsector Bargaining Assn. v. British Columbia, 2007 SCC 27 . The case law on section 7—which guarantees the right to “life, liberty and security of the person” and the right not to be deprived thereof except in accordance with the principles of fundamental justice—can be traced to Canada (Attorney General) v. PHS Community Services Society, 2011 SCC 44, [2011] 3 SCR 134 on safe-injection sites, Canada (Attorney General) v. Bedford, 2013 SCC 72, [2013] 3 SCR 1101, on sex work, and Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 SCR 331, on assisted suicide.

Across all of these areas, the Supreme Court set out legal standards that are now being hotly contested. In each, the Supreme Court ventured into extremely sensitive terrain. The underlying issues touched on the balance of interests in the political community (the right to strike), society’s capacity to express moral condemnation through the criminal law (mandatory minimums), and deeply contested moral questions (safe-injection sites, sex work, assisted suicide).

More notably still, in setting out these standards the Court arguably strayed far from the text of the Charter of Rights and Freedoms and, at times, departed from precedent. Only a few decades earlier, the same Court had refused to recognize a constitutional right to strike (Reference re Public Service Employee Relations Act (Alta.), [1987] 1 SCR 313), to assisted suicide (Rodriguez v. British Columbia (Attorney General), [1993] 3 S.C.R. 519), or to engage in sex work (Reference re ss. 193 and 195.1(1)(c) of the Criminal Code (Man.), [1990] 1 SCR 1123). And, to be charitable, the concepts of arbitrariness, overbreadth, and gross disproportionality that animate the jurisprudence under sections 7 and 12 of the Charter have no obvious link to the text of those provisions. They are, again being charitable, what Americans would call examples of constitutional construction rather than constitutional interpretation — that is, the judicial elaboration of principles and tests to fill gaps where constitutional text runs out.

To be sure, these judicial interventions did not occur in a vacuum. They often responded to pressing social realities and to perceived legislative inertia on questions of moral and empirical complexity. Whether this was legitimate constitutional construction or overreach into the political sphere is, in truth, the central question animating the modern debate over Charter interpretation and the limits of judicial creativity. And, as I say, the judges in that era were not (always) fashioning new limitations on state action from whole cloth — indeed, some of the developments (especially in relation to s. 7, which has been a notoriously wild horse) might be justifiable or even salutary.

Nonetheless, the recent political backlash has been a long time coming, as evidenced by coordinated calls for the use of s. 33 across multiple provinces. It has coincided with the emergence of more conservative legal organizations that question the Court’s incursions into politically and morally sensitive territory, its reversals of precedent, and the development of legal standards untethered from constitutional text. Yet these organizations — such as the Runnymede Society, Advocates for the Rule of Law, and the Canadian Constitution Foundation — are not simply partisan actors. They represent an intellectual response to what they perceive as judicial innovation without sufficient democratic warrant, seeking to restore textual, historical, and institutional constraints on judicial power.

Here, the parallels to the Warren Court in the United States — and the backlash it provoked, ultimately leading to the emergence of originalism as a constitutional theory — are difficult to ignore. In the 1950s and 1960s, culminating in (admittedly, post-Warren) Roe v. Wade (1973), the U.S. Supreme Court rewrote legal standards relating to criminal procedure and individual rights. There, too, the judges ventured beyond the constitutional text into areas of moral and political controversy. This prompted the rise of organizations like the Federalist Society, which brought together conservative lawyers who developed theories claiming to respect the original intent and meaning of the constitutional document, as well as theories redefining the balance of executive, legislative, and judicial power.

We are now seeing the results of those developments in real time, in a system without an equivalent to s. 33 and in which the resulting tensions had to be worked out within the judicial sphere. In the United States, there has been a collapse in public respect for the judicial system, due in no small part to the advances of the Warren Court and the subsequent backlash. When courts undo precedent and depart from constitutional text, they create a fundamental tension: should the legal community respect flawed precedent, or return to first principles? A return to first principles is often appealing, but it risks setting in motion the dynamic we now observe in the United States — successive generations of lawyers, some conservative and some progressive, vying for control of the courts to reverse the precedents of their ideological opponents. This is not, in my view, a positive development.

The line that can be drawn between the Supreme Court of Canada’s jurisprudence and the backlash it has provoked — first within the legal community and now within the political community—should inform any mature discussion of the rights and wrongs of resorting to the notwithstanding clause. The parallel with the United States should also be borne in mind. Whilst I entirely appreciate that the Warren Court was working in a very different context, the dynamics are similar enough that the American experience may be instructive. In Canada, access to the notwithstanding clause can alleviate the risk of such ideological cycles. It may serve as a useful safety valve for the Canadian legal system, avoiding the cyclical pressures causing damage in the United States.

This is one perspective from which to consider the Supreme Court’s foray into high constitutional politics in the English School Board case. The Court is being asked to overrule or modify its 1988 decision in Ford v. Quebec (Attorney General), [1988] 2 SCR 712, which legitimated a prophylactic use of the clause in Quebec, but it is now being asked to rule out “pre-emptive” uses of the clause and to preserve the judiciary’s ability to adjudicate on a Charter violation even where the clause has been invoked.

In my view, quite apart from Ford, the use of the notwithstanding clause cannot plausibly be confined to cases in which a judge has previously held a law to be unconstitutional. It defies common sense to require Quebec to legislate, await litigation, and then adopt the notwithstanding clause even where the inconsistency between the proposed legislation and the jurisprudence is clear at the outset.

Meanwhile, I think the argument for the preservation of a declaratory jurisdiction is too clever by half, especially because it assumes that which it needs to prove: namely, that the courts’ views on the Charter are relevant considerations in political and public debate about legislation that seeks to achieve policy goals notwithstanding jurisprudential constraints. Declarations that the Charter has been violated could and no doubt would play into debates about whether the clause should be invoked, renewed, or reversed — but the question is whether they should. And there is a very good argument that the whole point of s. 33, as evidenced by its text, is to keep the courts at one remove from invocations of the clause.

In all events, the safety-valve feature could usefully weigh in the Supreme Court’s analysis. Personally, I find most uses of the notwithstanding clause to be objectionable, not least because the political justifications offered for its use and the parliamentary scrutiny applied are deficient. Nonetheless, I find it hard to shake the thought that the notwithstanding clause could function as a constitutional safety valve that allows us to avoid spiralling into a vicious cycle that degrades both law and politics.

This content has been updated on March 28, 2026 at 21:58.