Preventing Constitutional Clauses from Disappearing: Keeping Proper Understandings of Section 33

Dwight Newman, KC, FRSC is Professor of Law and Tier 1 Canada Research Chair in Rights, Communities, and Constitutional Law at the University of Saskatchewan.

The four days of the Supreme Court of Canada hearing in the EMSB v. Québec case concerning Bill 21 saw surprisingly limited reference to the purposes of section 33 of the Canadian Charter of Rights and Freedoms (the ‘notwithstanding clause’) and some surprising attempts to make it subservient to other sections of the Canadian Charter.

Notably, there were various arguments that seemed to suggest subjecting uses to part or all of the section 1 limitations clause, from which the proportionality test expounded in R. v. Oakes and developed over the years has been derived. It is not simple to understand what precisely some of these arguments suggested because the section 1 test itself has been the subject of scholarly and jurisprudential discussions and changes.

The Attorney General of Alberta, in oral submissions made by the Deputy Minister of Justice himself, did one of the best jobs in articulating a purpose for the notwithstanding clause. As put by Malcolm Lavoie KC,

Section 33 was the pillar of the constitutional settlement of 1982 because it deals with the most fundamental question in any constitutional system, and that question is, who decides? Section 33 is fundamentally about who decides the scope and limits of the rights enumerated within it. (just after 1:03:07 of Day 3)

But such moments of clarity were rare. And we also had moments of counsel arguing for a judicial testing of parliaments’ and legislatures’ purposes in using section 33, calling for something much like the first branch of the Oakes test. On this issue, Justice Malcolm Rowe had an informal rejoinder that is very much on point. He explained that “if you impose the Oakes analysis on the exercise of 33, 33 goes poof,” in that it disappears, which means “all you have left is section 1” (just around 2:23:00 of Day 1).

In this short post, I want to highlight the three key purposes of section 33 that I have referenced in my past writing and how renewed attention to them bears on the hearing and the issues that arose.

I have referenced these three key purposes most recently and thoroughly in my chapter in the collection on section 33 edited by Peter Biro. There, I identify these purposes as making space for a legislative role in the interpretation of rights, permitting diversity in understandings of balances between different rights across the different units within Canadian federalism, and making it possible to respond to urgent circumstances with a use of section 33.

First, the most central of these purposes has been adopted by the Supreme Court of Canada in its City of Toronto decision in 2021, where the majority spoke to the role of section 33 in the course of explaining why unwritten constitutional principles cannot be used to strike down legislation (at para 60):

[…] s. 33 preserves a limited right of legislative override. Where, therefore, a court invalidates legislation using s. 2(b) of the Charter, the legislature may give continued effect to its understanding of what the Constitution requires by invoking s. 33 and by meeting its stated conditions (D. Newman, “Canada’s Notwithstanding Clause, Dialogue, and Constitutional Identities”, in G. Sigalet, G. Webber and R. Dixon, eds., Constitutional Dialogue: Rights, Democracy, Institutions (2019), 209, at p. 232).

In other words, section 33 permits legislatures to articulate and enforce their conception of rights and their implications, preserving a balance with aspects of the Westminster system of government in the face of calls by some to fully embrace American-style strong-form review.

The majority said this in 2021. It is not clear why anyone would suppose that the purpose of section 33 changed between 2021 and 2026. If the Court is to operate as a court of law rather than a sandbar of shifting political tides, there is strong reason for it to stand by its decision of just five years ago.

And if we say that section 33 permits legislative interpretation of the scope of rights, then it follows that there cannot be a section 1-style analysis of uses of section 33. To do so would be to subject a legislative decision once again to judicial scrutiny in a manner that denies the very legislative role at issue.

Second, implicit within the idea of provincial legislatures being able to use section 33 is an idea of permitted diversity within the federation. The Canadian Charter can bring Canadians together without being a homogenizing monster. Different provinces will take different stances on how to approach some rights questions based on different histories, different traditions, and different societal contexts. And that is a good thing within a place as large and diverse as Canada. The submissions for Droits collectifs Québec referencing the distinctive Québec experience with the notwithstanding clause must be understood as one exemplar of a differing use of the clause within the federation. Provinces made clear the importance of the notwithstanding clause even in who they sent to the hearings. As mentioned, Alberta sent not just a lawyer but its Deputy Attorney General. Ontario sent its Attorney General himself. Subjecting uses of section 33 to one homogenized section 1 analysis would take away from the federal nature of the clause.

Third, section 33 also functions to allow governments to respond to urgent circumstances. Ironically, some of those who sometimes speak of wanting to restrict section 33 to only those circumstances of ‘emergency’ and ‘last resort’ are the very first to call for imposing additional tests on section 33 and imposing forms of judicial scrutiny to uses of the clause that can perpetuate legal uncertainties through the courts for years upon years. Recognizing that the clause serves as a way for the federal Parliament or provincial legislatures to interpret rights differently, including in interpreting how to respond quickly to an issue, makes clear that it cannot be subjected to different layers of legal tests and judicial inquisition.

Consider Saskatchewan’s announcement of a readiness to use the clause in 2017 right after an April 2017 court decision that might otherwise have led children and parents to make longer-term decisions on where they would go to school based on the risk that the trial court decision would stand and force them to move partway through high school. The readiness to use the notwithstanding clause (later made unnecessary by an appellate decision) functioned to restore clarity when courts disrupted it. It is also exactly the sort of decision that those wanting to put additional restrictions and tests on section 33 would render unviable.

There is much at stake here, including the matter of whether a key structural clause that enabled the very adoption of the Canadian Charter is to disappear. If it does, there may be related consequences for trust in the courts and for national unity itself. For the Court, there is good reason to try to preserve the constitution, courts, and country.

Suggested citation: Dwight Newman, “Preventing Constitutional Clauses from Disappearing: Keeping Proper Understandings of Section 33,” Constitutional Theory Monitor (18 May 2026), online: https://constitutionaltheorymonitor.substack.com/p/preventing-constitutional-clauses

This post is part of a Symposium on the Notwithstanding Clause Appeals.