Blog of the International Journal of Constitutional Law and
Home Analysis 150 Years On: What is the Constitution of Canada?–Part 1 of 3–The Problem of Identification

150 Years On: What is the Constitution of Canada?–Part 1 of 3–The Problem of Identification

Editor’s Note: Today we begin a three-day series to mark the 150th anniversary of Confederation in Canada. In three separate posts, Maxime St-Hilaire, Patrick Baud and Éléna S. Drouin offer critical reflections on a provocative question: What is the Constitution of Canada? We thank them for sharing their views in this forum.

Maxime St-Hilaire, Université de Sherbrooke; Patrick F. Baud, McGill University; and Éléna S. Drouin, McGill University

Canada’s constitution was for many decades perhaps best known for combining British-style constitutional monarchy and parliamentary democracy with American-inspired federalism, albeit in a colonial form. Such a mixed constitution would have been considered improbable by the constitutional scholars of that age, such as A.V. Dicey (chap. III).

The Canadian federation’s constitution, which was initially set out in Canada’s Constitution Act, 1867 (CA 1867) served as a model for the Commonwealth of Australia Constitution Act in 1901. Over three decades ago, the Canada Act 1982 and its Schedule B, the English and French versions of the Constitution Act, 1982 (CA 1982), transferred full and exclusive constituent competency to Canada. This final step in Canada’s peaceful transition from British colony to independent country continues to serve as a model throughout the Commonwealth. Yet the constitutional reforms brought about in 1982 continue to raise the following essential, theoretical question: how, absent a revolution, can the full constituent competency be truly transferred rather than simply delegated to the former colony by the imperial legislator?

A basic question with no clear answer (yet)

The CA 1982 also brought major reforms to Canada’s constitution. In the past 35 years, Canada has become renown for its Charter of Rights and Freedoms, which served as one of the models for South Africa’s Bill of Rights, as well as the recognition and affirmation of Aboriginal and treaty rights in section 35 of the CA 1982, section that currently informs ongoing Australian debates on the constitutional recognition of Aboriginal and Torres Strait Islander peoples. But these oft-debated features of Canada’s constitution mask a more fundamental problem facing Canadian constitutionalists—they cannot answer what should be a simple question: what is the supreme law of Canada composed of? This question is one without a clear answer in Canadian law. For one, there exists no reliable list of the supreme law’s components and no such list could be established. Rather, it seems that the supreme law of Canada includes any possible provision that, based on the nature of its contents, falls under one of the (real) constitutional amendment procedures that the supreme law provides in certain of its key components. However, one of the amendment procedures does not specify which contents it applies to, yet it presents itself as the “normal,” that is to say, residual or default procedure. This draws a vicious circle. In our view, this problem is one that should be of interest not only to Canadian constitutionalists. This problem also particularly highlights the issues that can arise from perfectible constitutional drafting and the interaction among the constitutional amendment procedure and the rest of a constitution, notably a federative one.

A clue in the enshrining of the concept of “supreme law”

Subsection 52(1) of the CA 1982 “enshrines” the concept of supreme law into Canadian law, that is, of the “Constitution of Canada” as supreme law of this country, providing that “any law that is inconsistent with the provisions of the Constitution is, to the extent of the inconsistency, of no force or effect.” The French version, which possesses “equally authoritative” legal value under section 54 of the CA 1982, uses the word “inopérantes”. This is very unfortunate, because in Canadian federalism and distribution of powers jurisprudence—we do not use the single word “power” to refer to a legal power, just like we care much about the distinction between the ideal of a (political) constituent power and that of a (legal) constituent competency—the term “inopérant” does not mean “invalid”. To be sure, no later than 1985, the Supreme Court of Canada made clear, in an opinion concluding that “manner and form” requirements prescribed by the supreme law had not been met, that “[t]he words ‘of no force or effet’ [or “inopérantes”] in section 52 of the CA 1982, “mean that a law thus inconsistent with the Constitution has no force or effect because it is invalid”.

A provision that is misleadingly underinclusive

Subsection 52(2) of the CA 1982 indicates that the “Constitution of Canada”, so understood as Canada’s supreme law, “includes”: “(a) the Canada Act 1982, including [the Constitution Act, 1982]; (b) the Acts and orders referred to in the schedule [to the Constitution Act, 1982, including the Constitution Act, 1867]; and (c) any amendment to any Act or order referred to in paragraph (a) or (b).” Does subsection 52(2) mean that the legal, but supra-legislative constitution of Canada is to be exhaustively found in these Acts and orders, as amended? No, it does not.

Let us first recall that the “supreme law” is only the “written” (as opposed to the “unwritten” jurisprudential) element of the legal constitution in the formal, that is, supra-legislative sense. But then it must be recognized that the supreme law so properly—and strictly—understood does not correspond with what subsection 52(2) of the CA 1982 says it includes. For more than 20 years, it seemed well understood that the Supreme Court of Canada did not clearly rule out the possibility that the supreme law might include provisions other than those referred to by subsection 52(2) of the CA 1982 (through its Schedule). But in its 2014 opinion on sections 5 and 6 of the Supreme Court Act, the Court confirmed that provisions of this very Act that relate to the Court’s “essential features” are part of the supreme law. While the Supreme Court Act is not referred to by subsection 52(2) of or in the Schedule to the CA 1982, the Court reached its conclusion by virtue of the constitutional amendment procedure prescribed by the CA 1982. This procedure deals specifically with the “composition of the Supreme Court of Canada” —a category held by the SCC to include “the continued existence of the Court” —and more generally, “the Supreme Court of Canada” —a category held by the SCC to include, “at the very least, the Court’s jurisdiction as the final general court of appeal for Canada, including in matters of constitutional interpretation, and its independence” —at paragraphs 41(d) and 42(1)(d), respectively. This means that it is not subsection 52(2), but the constitutional amendment procedure, in the CA 1982, that determines whether or not a given provision is part of the supreme law of Canada.

A provision that is also misleadingly overinclusive

Indeed, subsection 52(2) of the CA 1982 is all the less useful in identifying the supreme law’s contents when we consider how many provisions included in the instruments it refers to do not belong to the supreme law properly understood—that is, in a way that makes sense given different legal effects. This is because many such provisions may be amended or repealed by ordinary legislation enacted by a regular legislator, notably under sections 44 or 45 of the CA 1982. It is contingent, if not unfortunate, that the latter sections have been put in Part V of the CA 1982, titled “Procedure for Amending the Constitution of Canada”. Their predecessor provisions, subsections 91(1)—its 1949-introduced version—and 92(1) of the CA 1867 rightly appeared in Part VI of that Act, titled “Distribution of Legislative Powers. Unlike most other sections of Part V of the CA 1982, sections 44 and 45 do not provide, or participate in the provision, of a special, heavier, and ad hoc procedure. Instead, they simply confer jurisdiction to those regular legislators that are the federal parliament and the provincial legislatures. They plainly read “44. Subject to sections 41 and 42, Parliament may exclusively make laws amending the Constitution of Canada in relation to the executive government of Canada or Senate and House of Commons. 45. Subject to section 41, the legislature of each province may exclusively make laws amending the constitution of the province.”

Although the Supreme Court of Canada is perfectly aware that some provisions referred to by subsection 52(2) of the CA 1982 can be amended or repealed by ordinary legislation under section 44 or 45 of the same British-imperial/supreme-law-belonging Act, it still refers to them as though they were part of the “Constitution of Canada” for the purposes of subsection 52(1), that is, the supreme law of Canada.

As a result of its blind fidelity to the wording of the CA 1982 and its particularly misleading subsection 52(2), the Supreme Court’s lack of both a theoretical and a practical understanding of what makes the difference between ordinary law and supreme law has it uselessly struggling in the New Brunswick Broadcasting case with the trap, for instance, of whether provisions that are not referred to in subsection 52(2), but are part of the “constitution of the province” for the purpose of section 45, are part of the “Constitution of Canada” and therefore the supreme law of Canada under subsection 52(1). But even when dealing with provisions that both fall under section 44 or 45 of the CA 1982 and that are undoubtedly part of an instrument referred to by subsection 52(2), the Court falls into the trap. This is what happened in Eurig Estate (Re), where Justice Major wrote the majority reasons—and was not contradicted by the dissenting or concurring opinions on this particular point. Justice Major took for granted that because the provincial legislatures’ competency over the “constitution of the province” had been moved from the federative distribution of competencies in the CA 1867 to a part of the CA 1982 titled “Procedure for Amending the Constitution of Canada”, it had to be read “in association with [sub]section 52(1) of the [CA 1982]”, as if it had changed from being a merely substantive constitutional competency to being a formal one. This led Justice Major to “read in” an (express) “form requirement” into section 45. It is symptomatic of his falling into the trap that Justice Major explained the existence of contradicting comments by Justice Pigeon in the Reference re Agricultural Products Marketing Act by mentioning the fact that they “were made before the passage of the 1982 amendments to the Constitution”. Yet, while what is covered by section 45 may be considered “constitutional law” in the substantial sense of the expression, it belongs, in (formal) essence, to ordinary law, which, by definition, cannot be part of the supreme law.

We submit Eurig Estate is incorrect on this point and should not be followed. No particular form beyond passage by the Senate and House of Commons or the legislative assembly of the province and royal assent by the Governor General or the Lieutenant Governor, as the case may be, is required. That was the case before 1982, as neither subsection 91(1) nor 92(1) required the use of a particular form to amend the constitutional provisions to which they applied. As the Supreme Court noted in its 2014 opinion on the reform of the Senate, sections 44 and 45 are the successors of those provisions and there was no intention in 1982 to do anything more than preserve the powers that subsections 91(1) and 92(1) conferred. It is notable, though not decisive, that Eurig Estate was not cited in the Supreme Court’s discussion of sections 44 and 45 in its opinion on the reform of the Senate. If Eurig Estate was good law, and set out a rule such that if the federal parliament or a provincial legislature failed to expressly amend (or repeal) a provision, the amendment (or repeal) would be invalid, we expect that the Supreme Court would have taken the opportunity to remind legislators.

A shift from a substantive to a procedural criterion (and to another provision)

As a result of the foregoing, we must not only discard subsection 52(2) of the CA 1982 in favour of a procedural criterion for determining whether a given provision is part of the supreme law of Canada, but also reject from this criterion sections 44 and 45, even though they appear in Part V, whose title suggests that its provisions are entirely devoted to the constitutional amendment procedure.

By the same rationale, we have to include section 35.1 of the CA 1982 in the procedural criterion, even though is it not only likely non-justiciable, but it is also found in a part other than Part V, that is, Part II, titled “Rights of Aboriginal Peoples of Canada”. Section 35.1 provides that “[t]he government of Canada and the provincial governments are committed to the principle that, before any amendment is made to Class 24 of section 91 of the “Constitution Act, 1867”, to section 25 of this Act or to this Part, (a) a constitutional conference that includes in its agenda an item relating to the proposed amendment, composed of the Prime Minister of Canada and the first ministers of the provinces, will be convened by the Prime Minister of Canada; and (b) the Prime Minister of Canada will invite representatives of the aboriginal peoples of Canada to participate in the discussions on that item”. Section 35.1 was added to Part II of the CA 1982 in 1983 by an amendment made under the general procedure, which is mostly provided for by section 38 of the CA 1982 and requires, among other things, the approval of two-third of the provinces representing 50 percent of the total population of the provinces.

Although section 35.1 undoubtedly concerns the rights of Aboriginal peoples, it was likely added to Part II in an attempt to avoid the application of paragraph 41(e), which requires, among other things, the unanimity of the provinces to amend Part V. In its opinion on the Senate reform, the Supreme Court concluded that, given the Senate’s role in the constitutional amendment procedure set out in Part V, its abolition, even if it left the text of Part V intact, would (more than incidentally) alter the functioning of the procedure and therefore falls under paragraph 41(e). Generalizing from the Supreme Court’s conclusion, paragraph 41(e)’s purpose is to protect the functioning of the constitutional amendment procedure and, absent the unanimity of the provinces, prevents the addition or removal of requirements for making constitutional amendments. Since section 35.1 seems to add a requirement for the making of a constitutional amendment to certain specified provisions relating to Aboriginal peoples and was not added with the unanimous approval of the provinces, we have reason to doubt its validity—it appears to be an unconstitutional constitutional amendment.

Section 35.1 is a further example, in a pool of many, of the careless drafting of the CA 1982. Our point here is that, because the defining criterion of the supreme law of Canada has to be procedural, its general principle is to be found in subsection 52(3) of the CA 1982, which provides that “[a]mendments to the Constitution of Canada [so understood] shall be made only in accordance with the authority contained in the Constitution of Canada”. This principle, and by extension, the declaration that the Constitution is the “supreme law of Canada” in subsection 52(1), are integral to the constitutional amendment procedure and likely fall under paragraph 41(e). In some sense, paragraph 41(e) is the ultimate constituent competency for Canada—the power to amend the constitutional amendment procedure—and thus the de jure sovereign in Canada. In the next two posts, we will examine the procedure in greater detail to see what it can tell us about what is and is not part of the supreme law of Canada.

Suggested Citation: Maxime St-Hilaire, Patrick F. Baud and Éléna S. Drouin, 150 Years On: What is the Constitution of Canada?–Part 1 of 3: The Problem of Identification, Int’l J. Const. L. Blog, Sept. 12, 2017, at:

Maxime St-Hilaire is an assistant professor of law at the Université de Sherbrooke. He wishes to thank the students in his constitutional law courses contributing to the development of the arguments made in this series of posts, as well as Laurence Bich-Carrière for helping to develop and present many of these arguments in more fragmentary form in “La constitution juridique et politique du Canada : notions, sources et principes”, in JurisClasseur Québec, Droit constitutionnel, Droit public collection (Montreal: LexisNexis, original ed.: 2011; latest update: 2015), booklet 1.

Patrick F. Baud is an honours law student at McGill University. He serves as senior policy advisor to Ontario’s Financial Accountability Officer and research assistant to the Parliamentary Budget Officer. These posts do not necessarily reflect the views of either of his employers.

Éléna S. Drouin is a law student at McGill University. She currently works as an Editor for the McGill Law Journal and as a research assistant for the Peter MacKell Chair in Federalism. These posts do not necessarily reflect the views of either of her employers

Print Friendly
Published on September 12, 2017
Author:          Filed under: Analysis

2 Responses

  1. Christa Scholtz

    The open question of s.35.1’s constitutional validity is what I explore in my chapter “Part II and Part V: Aboriginal Peoples and Constitutional Amendment”, in Emmett Macfarlane (ed), Constitutional Amendment in Canada, University of Toronto Press, 2016.

    • Patrick Baud

      Professor Scholtz, thank you for your comment. I greatly enjoyed your chapter on section 35.1. You are, to my knowledge, the only person other than Bryan Schwartz to have taken seriously the idea that section 35.1 might be invalid. I would be curious to know whether you think of our approach to s. 41(e) adds to the debate about section 35.1. Thanks again.

Leave a Reply

Your email address will not be published. Required fields are marked *