Article body

Notwithstanding anything in this Charter, the rights and freedoms referred to in it are guaranteed equally to male and female persons.

Indépendamment des autres dispositions de la présente charte, les droits et libertés qui y sont mentionnés sont garantis également aux personnes des deux sexes.

Section 28 of the Canadian Charter of Rights and Freedoms

The growing use of s. 33 of the Canadian Charter of Rights and Freedoms,[1] the so-called “notwithstanding clause,” by provincial legislatures has led to debates about little known or often forgotten aspects of the Canadian Constitution.[2] In this context, the debate surrounding s. 28 of the Canadian Charter has intensified in recent years, perhaps because the Supreme Court of Canada has yet to provide an account of the specific relationship between these provisions.[3] For example, in the context of the constitutional challenge to Quebec’s Bill 21,[4] which bans the wearing of religious symbols in certain contexts to advance an understanding of laïcité,[5] it is notably argued that s. 28, which is not itself subject to s. 33, can be used to guarantee gender equality through the courts. While the facts necessary to ground a violation of supra-legislative equality rights are debated in that case,[6] the legal argument is that if the law is found to be indirectly discriminatory against Muslim women, s. 28 could be used by courts to declare it unconstitutional and without legal effect. This is the case even though the Quebec National Assembly has invoked s. 33 to ensure the law operates notwithstanding, among other things, the supra-legislative equality guarantee in s. 15 and the freedom of religion guarantee in s. 2 (a) of the Canadian Charter.

Kerri Froc’s work arguing that s. 28 is a substantive provision that blocks the effect of a s. 33 declaration with respect to gender equality questions can be credited for making the argument particularly compelling. Of course, the provision is not unknown to scholars, particularly feminist scholars who have written about it over the years. However, it is fair to state that the provision has received little judicial attention and, to the extent that it has, is generally perceived as “interpretive” rather than substantive, contrary to the position advanced by Froc. In addition, some scholars continue to defend this understanding of the provision as being “interpretive.”

The purpose of this article is to suggest that neither of these camps are entirely correct. It is important to distinguish between the nature of the provision and how it applies in relation to other constitutional provisions. This distinction has been blurred by the debate so far. However, as I will explain, a key feature of this article is that the distinction between these questions, while important, is not watertight. For the sake of conceptual clarity, I consider them separately before tackling the relationship between the two. Accordingly, the debate about the nature of s. 28 requires the consideration of two distinct issues : (i) whether it is substantive or interpretive and (ii) a related question regarding the kind of substantive provision. The debate about the application of s. 28 involves determining its relationship with other provisions of the Constitution, including s. 33. This entails considering two distinct issues in relation to s. 28 : (i) whether it can itself be targeted by s. 33, and (ii) whether it is absolute. I contend that there is a strong argument to be made that the answer to the latter question is informed by what kind of substantive provision s. 28 creates.

I share Froc’s view that s. 28 is a substantive guarantee and not merely an interpretive provision. In practice, that means that it is a guarantee with its own work to do and is concerned with both wrongful direct and indirect discrimination. In my view, this conclusion is inescapable when one considers the provision’s language and history, although I acknowledge that there are contested accounts of the historical context. That said, Froc and I differ somewhat in our understanding of the nature of the provision and its application in practice. More specifically, I propose to offer an alternative account of the interrelationship between these two issues. On the account offered by Froc, it is sometimes said that s. 28 is an “independent” guarantee and/or that it is “absolute” in its application. On this account, s. 28 guarantees gender equality independently and absolutely or guarantees the equal exercise of the supra-legislative rights and freedoms enshrined in the Canadian Charter, irrespective of whether the underlying provision is targeted by a legislature under s. 33. Froc has deployed similar reasoning in relation to s. 1 of the Canadian Charter, which allows legislatures to justify reasonable limitations on supra-legislative rights and freedoms. In addition, while Froc herself has not joined them, some scholars argue that s. 28’s absolute status applies with respect to s. 32 as well, which concerns the application of the Canadian Charter.

In my view, while s. 28 is indeed a substantive provision, it is not independent. The difficult question is whether it is absolute or not in relation to s. 33. My view is that there is a strong argument to be made that s. 28 is not absolute in this context, particularly when one understands it in harmony with the scheme of the Canadian Charter and the Canadian Constitution generally. Accordingly, my view is that s. 28 must always be tied to an underlying right or freedom. I would make clear that it is an autonomous—but not independent—substantive guarantee, a distinction which I explain in this article. Moreover, and relatedly, there is a compelling argument to make that s. 28 does not eviscerate the purpose of s. 33 of the Canadian Charter with respect to the underlying right or freedom with which s. 28 must be tied. In highlighting this argument, I build on the work of scholars who argue that ss. 1 and 32 remain relevant in this context. They essentially argue that s. 28 is substantive, but not absolute. I agree and, furthermore, suggest that the argument can be extended to s. 33. I therefore share the view advanced by Froc and other scholars, namely that the words “Notwithstanding anything in this Charter” in the opening to s. 28 entail that the substantive gender equality guarantee will continue to apply and cannot itself be targeted by s. 33. However, it is not clear to me that recognizing this fact necessarily shatters the purpose of ss. 1, 32 and 33, with respect to the underlying provisions, thereby making s. 28 absolute.

My position can ultimately be broken down into five key points : (i) s. 28 is a substantive rather than a merely interpretive provision ; (ii) s. 28 is not an independent substantive guarantee like s. 15 ; (iii) s. 28 is an autonomous substantive guarantee of the equal exercise of the supra-legislative rights and freedoms guaranteed under the Canadian Charter for male and female persons ; (iv) if s. 33 is invoked by a legislature, such that a law continues in operation, there is a strong argument that the underlying provision to which s. 28 is tied cannot be guaranteed equally and assist in invalidating legislation ; and (v) however, since s. 33 cannot target s. 28 itself, the latter provision will continue to guarantee the equal exercise of the supra-legislative rights and freedoms that are not or cannot be targeted by s. 33 to male and female persons.

In this article, I begin by outlining the predominant jurisprudential view of s. 28 (1) before considering the alternative account of this provision advanced by scholars such as Froc (2). I then seek to chart a middle ground between these positions, by first recognizing and affirming the substantive nature of s. 28, thus rejecting the accounts of those who argue that it is merely an interpretive provision. I distinguish an autonomous substantive guarantee from an independent substantive guarantee, with a view to clearly reject the position that s. 28 could be an independent guarantee (3). I also consider the application of s. 28 in relation to other constitutional provisions and observe that, while two views are possible, there is a strong argument to make that s. 28 is not absolute in relation to s. 33 (4). In coming to this conclusion, I consider the debate regarding the relationship between s. 28 and ss. 1 and 32 of the Canadian Charter, which serves to bolster my basic argument. In both parts, I turn to jurisprudence and scholarship on s. 10 of Quebec’s Charter of Human Rights and Freedoms,[7] which, in my view, provides a compelling legal framework for understanding the nature and application of s. 28 going forward, particularly with respect to indirect discrimination. This then leads me to discuss the other constitutional principles that bear on the questions raised by s. 28 and its relationship with s. 33 (5). Finally, I discuss the trial decision in Hak v. Procureur général du Québec, which appears to be consistent, at least in part, with this middle ground position (6). I therefore do not propose an exhaustive review of the scholarly literature on s. 28 and a discussion of gender equality itself or the Supreme Court’s record on this front. The equality jurisprudence has been through a series of changes, particularly in recent years.[8] I do not enter the debate here. Instead, I propose to bring to the fore an alternative understanding of s. 28 and its relationship with s. 33 in the light of recent debates.

1 Section 28 : An Interpretive and Non-Absolute Provision

Section 28 of the Canadian Charter provides as follows :

  • 28. Notwithstanding anything in this Charter, the rights and freedoms referred to in it are guaranteed equally to male and female persons.

    28. Indépendamment des autres dispositions de la présente charte, les droits et libertés qui y sont mentionnés sont garantis également aux personnes des deux sexes.

The predominant jurisprudential view is that s. 28 plays an interpretive or confirmatory function, i.e., that it has no substantive content of its own. As the British Columbia Court of Appeal explained in McIvor v. Canada (Registrar of Indian and Northern Affairs), s. 28 is “a provision dealing with the interpretation of the Charter. It does not, by itself, purport to confer any rights, and therefore cannot be ‘contravened’.”[9] Indeed, even some scholars who advocate the contrary position have sometimes indicated that to interpret s. 28 literally might lead one to conclude that it is “little more than a guide to the application of the Charter.”[10] This has led to observations that the provision has “to date played a limited role in litigation.”[11]

On this view, therefore, s. 28 can—at most—inform the interpretation of other provisions, and its existence could be relevant to a limitations analysis pursuant to s. 1 of the Canadian Charter.[12] Moreover, invoking s. 33 of the Canadian Charter with respect to the guarantee of equality provided in s. 15 generally forecloses the possibility of courts declaring laws unconstitutional on gender equality grounds. While these scholars do not always explain the practical consequences of their view as applied to individual cases, the effect of their position appears to be that s. 28 does not have as much work to do or will be concerned primarily with direct discrimination.[13]

Some authors have continued to defend a similar view in recent years. For instance, Maxime St-Hilaire has argued that s. 28 is “more of an interpretive provision of rights otherwise guaranteed […], not a provision conferring a gender equality right separate from section 15, which expressly includes such right.”[14] He also grounds his analysis in the historical context by pointing to Barry Strayer’s article on s. 15 of the Canadian Charter recounting his participation in the drafting. At the time, ss. 25 and 27 of the Canadian Charter were added because some Indigenous groups and minorities did not want s. 15 to limit their traditional practices. Strayer therefore contends that s. 28 was meant to counteract the addition of these provisions.[15]

In addition, St-Hilaire observes that this interpretive function “applies only until such time as the right or rights in question have been suspended in respect of particular legislative provisions by means of the overriding provision adopted under section 33.”[16] This leads him to conclude that s. 33 “allows for the overriding of Charter rights, but not the principles of interpretation it sets out.”[17] St-Hilaire also considers the relevance of international law, namely the International Covenant on Civil and Political Rights,[18] and notes that it would support Froc’s argument. In the end, however, his view is that s. 28 does not go “that far.”[19]In other words, the Canadian Constitution is not reflective of international law on this point. Instead, the drafters “made gender equality a component of a general right to equality as a protection against discrimination, from which it has authorized the ordinary legislator to derogate, so that the principle of gender equality in the exercise of fundamental rights is only a principle of interpretation (even if it cannot be derogated from).”[20]

Asher Honickman has advanced a similar position.[21] He argues that “while s. 28 is an important interpretive provision, it does not contain an independent and justiciable right.” Honickman finds support for his conclusion in the fact that s. 28 “was placed in the part of the Charter entitled ‘General’.” He observes that “[i]t seems unlikely, to say the least, that the most far-reaching and absolute right enumerated in the Charter would be placed smack dab in the middle of the part dealing with interpretation.” He similarly points to Strayer’s article to anchor his view in the context and concludes from this that “feminist groups […] wanted to ensure that the protections afforded to aboriginal groups and minorities were not prejudicial to women.”

In addition, Honickman agrees with Froc that, to the extent of any conflict between s. 28 and s. 33, the former must prevail. However, he argues that these provisions do not actually conflict with each other. Section 28 deals “with the rights Canadians possess and particularly how to interpret those rights. Section 33, by contrast, deals with the operation of legislation that is contrary to those rights.” In this respect, he adopts Grégoire Webber’s view that “the invocation of the notwithstanding clause does not act as a bar to judicial review since the issue of whether a right has been violated is separate and distinct from whether government legislation may continue to operate in spite of the violation.”[22] Finally, he adds that “the necessary implication” of Froc’s position, described in further detail below, is that “s. 28 would also supersede, and therefore not be subject to, s. 1.” He notes that this aspect of her reasoning is “plausible,” but that “a Charter right guaranteed to one sex but not the other would arguably never constitute a reasonable limit” in any event.

Overall, while St-Hilaire and Honickman differ in some respects, they agree that s. 28 is interpretive in nature and does not confer a substantive gender equality right distinct from s. 15. As mentioned above, the result is that s. 28 has a more limited role in Canadian Charter adjudication and certainly cannot prevent the legislature from using s. 33 to maintain the operation of its laws notwithstanding ss. 2 and 7–15. While these authors do not present their arguments in this way, I will come to explain that their position appears to be that the nature of s. 28 is interpretive. When they pronounce on s. 33, they should be understood to be making a distinct argument that s. 28 is not absolute in its application.

2 Section 28 : An Independent and/or Absolute Substantive Provision

This predominant jurisprudential view has been contested by many scholars, most notably Kerri Froc, who wrote a compelling doctoral thesis on this topic.[23] She recognizes the predominant jurisprudential view and takes aim at it by looking to the provision’s language and the historical context surrounding its adoption. In addition, she argues that the use of s. 33 of the Canadian Charter may foreclose challenges that rely solely on ss. 2 and 7–15, but it cannot bar a legal challenge based on gender equality because s. 28, a substantive guarantee with “its own independent work to do,”[24] is not subject to the notwithstanding clause. Indeed, its opening words expressly specify that s. 28 applies “Notwithstanding anything in this Charter.”[25]

Froc explains that s. 28 “was intended to transform judicial understandings of rights (particularly equality) to ensure that they were accessible to women in practice, and to protect gender equality from being undermined by other provisions of the Charter or judges themselves.”[26] She highlights that “[w]omen had been troubled by years of narrow interpretations of equality that trivialized sex discrimination claims under the statutory Canadian Bill of Rights,”[27] and points to the holding in Bliss v. Attorney General of Canada as an egregious case of the narrow approach to equality that characterized the pre-Canadian Charter era.[28]

As a result, women and the organizations that represent them came out strongly to bolster what would become s. 15 of the Canadian Charter. The changes announced by the federal government in January 1981 accordingly revised the equality guarantee to protect equality “before and under the law and […] equal protection and equal benefit of the law.” It was thought that this would preclude the interpretations that prevailed with the narrower wording outlined in the Canadian Bill of Rights.[29]

At the same time, that version included what would become s. 27 of the Canadian Charter, which would “direct courts to consider Canada’s ‘multicultural heritage’ in interpreting rights, raising the spectre of women’s rights being derogated or nullified in the name of protecting cultural practices.”[30] This led to a groundswell of activism from feminists, who lobbied for additional changes. At the time, she argues that their aim “was to ‘ensure that all of the rights and freedoms set out in the Charter will be interpreted so as to apply equally to men and women,’ that sex discrimination is taken ‘equally seriously’ as that of race, and that sex-based distinctions undergo a high degree of judicial scrutiny.”[31] The purpose, then, was to protect gender equality and counteract the addition of s. 27. In the end, the wording of the current version of s. 28 was adopted on April 23, 1981, including its use of the words “Notwithstanding anything in this Charter.”

Froc also highlights that, after the November 1981 “Kitchen Accord” that led to the adoption of the notwithstanding clause in order to bring the provinces (except Quebec) on board with the adoption of supra-legislative human rights norms, there were attempts to subject s. 28 to s. 33. At the time, the opening part of s. 28 would have read, “Notwithstanding anything in this Charter except section 33.” Relatedly, s. 33 would have been modified to add, “section 28 of this Charter in its application to discrimination based on sex referred to in section 15.”[32] In the end, these proposed changes were removed from the Canadian Charter with the consent of the provinces (except Quebec) in response to the substantial advocacy of women in Canada. In other words, it was agreed that s. 33 could not be used to target s. 28 itself.

This leads Froc to argue that a s. 33 declaration cannot prevent a claim based on gender equality generally. As she explains, “[t]here would be no reason for the initial insistence upon its explicit inclusion in the override unless [the framers] accepted that section 28 could operate independently—in this case, to block discriminatory government action after the override was invoked in relation to a section 15 sex equality violation.”[33] Froc notably draws further support for this conclusion from a book authored by members of the November 1981 Saskatchewan constitutional delegation, who note that the removal of the application of s. 33 from s. 28 “in effect […] meant that sexual equality in section 15 could not be overridden.”[34]

Ultimately, Froc concludes that : (i) “[a] section 28 violation cannot be preserved using section 33,”[35] and (ii) where a s. 15 violation concerning gender equality is preserved by s. 33, “section 28 operates to block the effect of that invocation.”[36] These two propositions reflect Froc’s view that s. 28 is an “independent” and/or “absolute” guarantee.[37] At times, she seems to only intend to convey the idea that s. 28 is absolute in its relationship with s. 33 and seems to accept that it is not an independent guarantee, but this is not always clearly stated. It is my aim to explain these distinctions in this article, which I see as important to a proper understanding of s. 28.

While Froc’s view does not reflect the predominant jurisprudential approach, it also has some support in case law. In Syndicat de la fonction publique du Québec inc. v. Québec (Procureur général), which involved pay equity legislation, the Superior Court of Quebec ultimately concluded that part of the scheme at issue was unconstitutional pursuant to s. 15 of the Canadian Charter. In arriving at this conclusion, the court also considered s. 28 of the Canadian Charter at length.

The court reviewed much of the historical background outlined above and turned to the scholarship on the issue in order to conclude that “although the principle of equality under section 15 may be set aside by the legislature under section 33, no law could make a distinction, even expressly, on the basis of sex on pain of invalidity.”[38] It acknowledged, however, that the debate surrounding ss. 1 and 32 of the Canadian Charter and their relationship with s. 28 was more controversial and “does not have to be resolved in the context of this judgment.”[39] In the end, the court appears to have held that s. 28 is an absolute substantive guarantee with respect to s. 33 and its reasons could be read by some to mean that s. 28 is an independent substantive guarantee as well, an issue to which I return in my own analysis.[40]

The view of s. 28 as absolute in its relationship with s. 33 also received some support from the former Chief Justice of Quebec, dissenting in the context of an appeal from the denial of an interlocutory injunction in Hak. While the judgment is, by its nature, only a preliminary view of the merits,[41] the Chief Justice indicated that the introductory phrase “Notwithstanding anything in this Charter” could “lead one to believe that section 28 blocks the effect of a section 33 override when a statute restricts access to certain fundamental rights unequally between the sexes.”[42] However, this latter sentence of the reasons acknowledges that s. 28 is not an independent substantive guarantee.

3 Charting a Middle Ground : The Nature of Section 28

This review of existing positions now leads me to articulate a middle ground by distinguishing between the nature of s. 28 and its application in relation to other constitutional provisions. My analysis rests on the contention that constitutional provisions must be understood purposively and harmoniously with the constitutional scheme.[43] While Froc’s analysis is compelling, my understanding of s. 28 suggests that both scholarly views are partly flawed. This is not meant to be a criticism. Even those who participated in the drafting of the Constitution Act, 1982,[44] namely Mary Dawson, have sometimes acknowledged that the legal effect of this provision is “somewhat unclear.”[45] This is also reflected in the competing accounts of the history surrounding its enactment and the ongoing debate amongst scholars. In highlighting a third way, I build on the work of others in an attempt to contribute to discussions that seek to make sense of this provision. In this part, I begin by considering the nature of s. 28. This entails considering : (i) whether s. 28 is substantive or not and (ii) what kind of substantive provision is contemplated by s. 28.

In my view, the position advanced by St-Hilaire and Honickman cannot be fully accepted because it deprives s. 28 of its substantive role. I start with the language of the provision, which provides that “the rights and freedoms referred to in [the Canadian Charter] are guaranteed equally to male and female persons.” Therefore, unlike ss. 26 and 27, which focus primarily on “construct[ion]” and “interpret[ation],”[46] s. 28 “guarantees” supra-legislative rights and freedoms equally to male and female persons. The use of the French word “garantis” also conveys this idea. Indeed, as Hubert Reid and Simon Reid observe in their Dictionnaire de droit québécois et canadien, a “garantie” is a “[l]egal provision to protect certain fundamental rights.”[47] While s. 28 is in the “general” part of the Canadian Charter, this does not preclude a substantive function. For instance, s. 25 of the Canadian Charter has been interpreted substantively by the courts.[48]

I would add that s. 28’s place in the general part is logical because, as I will explain in greater detail below, it is a guarantee tied to all of the other rights or freedoms in the Canadian Charter. In that sense, it is indeed “general,” as the heading of this part suggests.[49] In my view, the real distinction between the main parts and the general part of the Canadian Charter is that the main parts include independent supra-legislative rights and freedoms, whereas the general part deals with issues that affect all of the supra-legislative rights or freedoms or other aspects of the Canadian Constitution. This entails that the purpose of these provisions can be substantive, interpretive or declaratory depending on the context.[50]

I acknowledge that Gerard Kennedy has argued that all of these provisions are “interpretive.” I note, however, that he explains that these provisions can sometimes function as “interpretative trumps,” pointing to s. 25 as an example.[51] While the distinction between our views of the “general” part may be more superficial than real in some respects, I would explicitly acknowledge that, in certain contexts, these provisions have a substantive function. This will depend on the purpose of each provision and the circumstances of the case. I part company with him, however, if he agrees with St-Hilaire and Honickman that s. 28 is “interpretive” and not “substantive.” While I concede that this would make s. 28 the only rights-granting provision in this part, seeing as s. 25 protects rights and freedoms that are external to the Canadian Charter, I do not see this as an obstacle to its substantive nature where the purpose of the provision leads to that conclusion.

This brings me to the historical context, which bolsters my understanding of the text. While I acknowledge that there are competing accounts of the relevant history, Froc’s meticulous account of this context is compelling. While constitutional analysis cannot be reduced to a search for subjective intentions, as Froc herself also acknowledges, my view is that it is a factor to consider because it forms part of the context that led to the enactment of the Canadian Charter. Therefore, s. 28 is indeed a substantive gender equality guarantee distinct from s. 15 of the Canadian Charter.[52] The historical context discussed above demonstrates why this guarantee was added to the Constitution : to protect gender equality, particularly in light of the adoption of ss. 25 and 27. In my view, this aspect of the nature of s. 28 should not be controversial.[53]

The strongest case for the “interpretive” approach might be made by reference to more recent developments in Quebec law. In 2008, the Quebec National Assembly added s. 50.1 to the Quebec Charter, which provides that “[t]he rights and freedoms set forth in this Charter are guaranteed equally to women and men.”[54] The avowed purpose of the change was to emphasize the importance of gender equality. The parliamentary committee debated extensively about the nature of the proposed provision—a debate similar to the one that is currently taking place between Froc and other scholars regarding s. 28. Alexandre Cloutier, a lawyer and Member of the National Assembly of Quebec at the time, observed that the use of the word “guaranteed” suggested the clause was “not interpretive.”[55] He was assured, however, that this specific wording was chosen because s. 28 of the Canadian Charter, which is similarly worded, “was seen by many precisely as an interpretative provision.”[56] Moreover, he was told that this wording was a safe path because s. 28 had been “relatively little used by the courts”[57] and the aim of the proposed wording of s. 50.1 of the Quebec Charter was to “isolate its real impact.”[58] In response to this potential objection, I would note that, while I agree with Me Cloutier’s assessment of the meaning of the word “guaranteed,” the context surrounding the adoption of s. 50.1 of the Quebec Charter is different from the one surrounding the adoption of s. 28 of the Canadian Charter as a result of this debate. This may justify a distinct understanding of that provision. Moreover, I would note that, in response to Me Cloutier, it was also acknowledged that some have claimed that “perhaps the full potential of s. 28 has not been argued before the courts.”[59] Froc and I would agree with this latter comment.

The real debate with respect to its nature lies elsewhere in my opinion. The question is what kind of substantive guarantee is contemplated by s. 28. I would make clear that s. 28 is an “autonomous,” not an “independent,” substantive guarantee. To date, this specific distinction has not appeared in the debate amongst scholars on s. 28. In some cases, scholars speak of s. 28 as being an “independent” guarantee and their argument seems to track the meaning of this concept. In other cases, scholars appear to use the word “independent” as a synonym for “substantive” or “absolute.” It is also possible that some authors who advocate for the “interpretive” approach truly intend to convey that s. 28 is a non-independent guarantee, but are unaware of the concept or have not provided a complete account of their views. This has the effect of confusing the debate. The ideas must be clearly distinguished.

The concept of “independence” refers to the idea of a stand-alone guarantee, i.e., one that “applies on its own” with “no connection” to another right or freedom.[60] Most of the guarantees in the Canadian Charter are substantive guarantees of this kind. That said, if the guarantee at issue is or must be connected to other rights or freedoms, it is not independent. This would mean that the guarantee must necessarily be invoked in correlation with other provisions. This is not to say that the link must be strict, but that it does or must exist. However, it is also important to acknowledge that this non-independent status does not negate the substantive function of such a guarantee. The concept of an “autonomous” provision, which I explain more fully below, refers to a sphere of action in which a provision plays a distinct role. This explains why such a guarantee can be characterized as a kind of substantive guarantee. It entails that the non-independent guarantee can “be infringed by a measure that in itself complies with the requirements of another article.”[61] These types of guarantees are less common and, in my view, s. 28 is the only non-independent guarantee in the Canadian Charter.

Accordingly, I do not consider s. 28 to be an independent substantive guarantee. The language of s. 28 indicates that “the rights and freedoms referred to in [the Canadian Charter] are guaranteed equally to male and female persons.”[62] The French version is to the same effect. This suggests that the provision is concerned with protecting the equal exercise or enjoyment of the supra-legislative rights and freedoms contained in the Canadian Charter. Section 28 ensures that these other provisions are “guaranteed equally.” It is not a stand-alone gender equality guarantee like s. 15. To my mind, this is also compatible with the historical context that Froc has brought to the fore in her work. It does not deny the substantive nature and function of s. 28, but acknowledges that its primary purpose is to safeguard the equal exercise of other rights and freedoms. For the purpose of applying this gender equality provision, it must be attached to another right or freedom guaranteed in the Canadian Charter.

In my view, a look to Quebec law, specifically s. 10 of the Quebec Charter, can help to understand the nature of s. 28. Section 10 is, much like the equality provision in continental systems, an autonomous guarantee, but one that is not independent.[63] My aim is not to provide an exhaustive account of the case law on s. 10 of the Quebec Charter, but merely to highlight the similarities between the two provisions in service of my argument on the understanding of s. 28.

Section 10 of the Quebec Charter provides as follows :

  • 10. Every person has a right to full and equal recognition and exercise of his human rights and freedoms, without distinction, exclusion or preference based on race, colour, sex, gender identity or expression, pregnancy, sexual orientation, civil status, age except as provided by law, religion, political convictions, language, ethnic or national origin, social condition, a handicap or the use of any means to palliate a handicap.

    Discrimination exists where such a distinction, exclusion or preference has the effect of nullifying or impairing such right.

  • 10. Toute personne a droit à la reconnaissance et à l’exercice, en pleine égalité, des droits et libertés de la personne, sans distinction, exclusion ou préférence fondée sur la race, la couleur, le sexe, l’identité ou l’expression de genre, la grossesse, l’orientation sexuelle, l’état civil, l’âge sauf dans la mesure prévue par la loi, la religion, les convictions politiques, la langue, l’origine ethnique ou nationale, la condition sociale, le handicap ou l’utilisation d’un moyen pour pallier ce handicap.

    Il y a discrimination lorsqu’une telle distinction, exclusion ou préférence a pour effet de détruire ou de compromettre ce droit.[64]

It must be acknowledged that the Quebec Charter not only predates the adoption of the Canadian Charter, but also has a broader reach. Whereas the Canadian Charter focuses on the state’s conduct, the Quebec Charter is not so limited because it applies in the private and public spheres.[65] However, s. 10 of the Quebec Charter, unlike s. 15 of the Canadian Charter, is not an independent equality guarantee. It is a guarantee of equality in the exercise of other human rights and freedoms. There is no dispute in Quebec law that s. 10 is a substantive provision, but unlike s. 15 of the Canadian Charter, it is a different kind of substantive guarantee.

As the Supreme Court of Canada explained in Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Bombardier Inc. (Bombardier Aerospace Training Center), the right to equality in the Quebec Charter “is protected only in the exercise of the other rights and freedoms guaranteed by the Charter.”[66] The English version of the judgment does appear to refer to s. 10 as “independent,” but the official French version of the reasons for judgment makes clear that the guarantee is “autonome”[67] (which literally means “autonomous”). This underscores the need to read court decisions and legal sources in both languages to undertake legal analysis in Canada.[68] More recently, the Court affirmed emphatically in Ward v. Quebec (Commission des droits de la personne et des droits de la jeunesse) that “s. 10 does not establish an independent right to equality.”[69] In the official French version, the reasons are to the same effect. The same has been said of s. 10’s continental cousin, art. 14 of the European Convention on Human Rights, which “may be invoked and applied only together with other substantive guarantees of the Convention and its Protocols.”[70]

This does not negate the important autonomous status of the provision. The Supreme Court indeed explained the substantive status of the provision in Bombardier by observing that the requirement to attach the equality guarantee to other rights and freedoms “does not require a ‘double violation’ (right to equality and, for example, freedom of religion), which would make s. 10 redundant.”[71] It is a provision with its own work to do. Similarly, with respect to art. 14 of the European Convention on Human Rights, it has been said that while it “cannot serve as an independent guarantee, it is nevertheless autonomous, insofar as compliance with [it] must be examined even if the substantive guarantee of the Convention or its Protocols has not been breached.”[72]

David Robitaille explains the autonomous substantive nature of s. 10 particularly well and has done much to conceptually clarify the ambit of this provision. As he observes in defining the provision, “section 10 can be infringed by a measure which in itself complies with the requirements of another section but which creates an inequality of treatment in the recognition or exercise of the right or freedom enshrined in that other provision.”[73]

He cites an example that can assist in understanding the purpose of s. 28. He points to the Belgian Linguistics case, a foundational case decided in 1968 under the European Convention on Human Rights, which involved the right to education and the equal exercise of that right.[74] In that case, the French-speaking applicants namely contested a Belgian law that required them to send their children to study outside their community to receive instruction in their language, in contrast to Dutch children who could receive instruction in their own language in their community. Since the right to instruction does not include the right to receive education in the language of one’s choice, a breach of that right could not be made out. However, the autonomous status of the equality provision meant that, on one of the questions raised, the court ultimately concluded that, while instruction itself may be available and thus consistent with that specific guarantee, it was not equally available to both language communities. This was a breach of the equal exercise of a right or freedom.[75]

The continental approach therefore helps us understand that s. 28 is indeed an autonomous substantive guarantee with a role to play in protecting gender equality in Canadian society. Relegating the provision to a mere interpretive status would defang the guarantee and its utility when s. 15 is not available and a breach of the underlying provision itself cannot be made out. Its utility would be particularly evident in cases of wrongful indirect discrimination,[76] a point which the Belgian Linguistics case and its progeny helpfully illustrate.

At the same time, it must be acknowledged that s. 28 is not an independent substantive guarantee. The distinction explored here is legally significant and has not been fully appreciated. For instance, Froc observes that s. 28 has “independent work to do.”[77] At times, she highlights that the activists who fought to include s. 28 viewed it as having “independent power” that was not “dependent on other rights.”[78] In other cases, she appears to take the position that s. 28 is either independent or autonomous. For example, she has argued that “section 28 precludes the operation of legislation that discriminates on the basis of sex or denies other rights or freedoms (such as freedom of religion) unequally between males and females.”[79] In some cases, while she formally maintains this dual possibility, her argument seems more nuanced and is primarily concerned with explaining that s. 28 is “absolutely protected,” with the word “independent” being used to distinguish her view from those who view s. 28 as “interpretive.”[80] On this view, what she really intends to convey is that s. 28 is autonomously substantive and absolute in its application. The use of the word “independent” merely distracts from her core claim. Indeed, Froc refers to continental jurisprudence on equality rights in her work, which appears to indicate that she concedes that s. 28 is not an independent guarantee.[81] Nonetheless, as explained above, Froc sometimes observes, that s. 28 is or could potentially be an independent guarantee. If this is her view, I do not share this aspect of her reasoning because the autonomous status of the guarantee remains legally significant, and it must be distinguished from an independent guarantee.

In sum, I have tried to demonstrate that the issues in play must be clearly distinguished and have suggested that a look to continental concepts can assist in doing so. It is a mistake to diminish the substantive nature of s. 28. At the same time, I do not understand s. 28 to be an independent substantive guarantee. If Froc disagrees with this view, we are ships passing in the night. However, to the extent that Froc recognizes and affirms that the provision is non-independent, yet nonetheless substantively autonomous, we share the same view. In my view, her core claim that s. 28 is “absolute” in its application can survive an acknowledgement that it is not “independent” in nature. However, as I will explain below, her claim that s. 28 is “absolute” is weakened in these circumstances given the nature of an autonomous, non-independent guarantee.

4 Charting a Middle Ground : The Application of Section 28

Having reviewed the nature of s. 28, I now consider how it applies in relation to other constitutional provisions. This aspect of the analysis focuses more directly on the legal effect of the words “Notwithstanding anything in this Charter.” This entails considering two distinct questions in relation to s. 33, namely (i) whether s. 28 is itself subject to s. 33 and (ii) whether s. 28 is absolute in its interaction with s. 33.

Section 33 of the Canadian Charter provides :

  • 33. (1) Parliament or the legislature of a province may expressly declare in an Act of Parliament or of the legislature, as the case may be, that the Act or a provision thereof shall operate notwithstanding a provision included in section 2 or sections 7 to 15 of this Charter.

    (2) An Act or a provision of an Act in respect of which a declaration made under this section is in effect shall have such operation as it would have but for the provision of this Charter referred to in the declaration.

    (3) A declaration made under subsection (1) shall cease to have effect five years after it comes into force or on such earlier date as may be specified in the declaration.

    (4) Parliament or the legislature of a province may re-enact a declaration made under subsection (1).

    (5) Subsection (3) applies in respect of a re-enactment made under subsection (4).

  • 33. (1) Le Parlement ou la législature d’une province peut adopter une loi où il est expressément déclaré que celle-ci ou une de ses dispositions a effet indépendamment d’une disposition donnée de l’article 2 ou des articles 7 à 15 de la présente charte.

    (2) La loi ou la disposition qui fait l’objet d’une déclaration conforme au présent article et en vigueur a l’effet qu’elle aurait sauf la disposition en cause de la charte.

    (3) La déclaration visée au paragraphe (1) cesse d’avoir effet à la date qui y est précisée ou, au plus tard, cinq ans après son entrée en vigueur.

    (4) Le Parlement ou une législature peut adopter de nouveau une déclaration visée au paragraphe (1).

    (5) Le paragraphe (3) s’applique à toute déclaration adoptée sous le régime du paragraphe (4).[82]

In my view, it is evident that s. 33 cannot be used to target s. 28 itself. The crucial question, however, is whether s. 28 is absolute. If it is, then s. 28 can be used to guarantee the equal exercise of all supra-legislative rights and freedoms in the Canadian Charter, irrespective of a legislature’s decision to invoke s. 33 with respect to the underlying right or freedom. If it is not, this means that s. 28 continues to apply, but that it will have no utility in relation to laws that continue to operate notwithstanding an underlying right or freedom to which s. 28 must be tied. The result of this alternative approach is that there are contexts in which s. 28 will always be relevant, namely with respect to democratic (ss. 3–5), mobility (s. 6) and language rights (ss. 16–23). Indeed, this article aims namely to emphasize that s. 28 can continue to apply substantively in these contexts.

While Froc’s argument is plausible and could be adopted by the courts, I am not entirely convinced that s. 28 is absolute, as she contends. This is a point which the jurisprudence mentioned in Part 1 does not address because it is concerned with the nature of the provision. St-Hilaire and Honickman, while defending the predominant jurisprudential view, speak to both the provision’s nature and its application in relation to other provisions. While I do not share their view of the first issue, their view of the second issue is on much more solid footing. In coming to this conclusion, I also consider other provisions that bear on the debate, namely ss. 1 and 32. In my view, the words “Notwithstanding anything in this Charter” undoubtedly convey the idea that the substantive guarantee in s. 28 will continue to apply, but this does not necessarily eviscerate the purpose of other constitutional provisions with respect to the underlying supra-legislative rights and freedoms with which s. 28 must be tied.

I agree with Froc that the existing wording of s. 28 predated the adoption of s. 33.[83] This means that the words “Notwithstanding anything in this Charter” contained in s. 28 were added before s. 33 was being contemplated by the drafters. At the time these words were included, the purpose was notably to counteract s. 27 in constitutional adjudication. It conveyed the idea that substantive gender equality in the exercise of other supra-legislative rights and freedoms would continue to apply. Their purpose was not to counteract s. 33. Once s. 33 was added to the Canadian Charter, the debate centred around adding s. 28 as one of the provisions that could itself be targeted by s. 33. As mentioned previously, there were attempts to do that, which were ultimately rejected in the final version of the Canadian Charter.[84] Therefore, in my view, the result is that, when there is constitutional adjudication to be done, s. 28 can notably be used to counteract s. 27. In addition, the choice made to remove s. 28 from s. 33 implies that this guarantee of gender equality cannot itself be targeted by the notwithstanding clause.[85] Overall, this means that s. 28 will remain relevant in some circumstances. This comes out clearly from the text and history surrounding the enactment of the Canadian Charter. Indeed, as Froc highlights, part of the reason for removing s. 28 from s. 33 was the concern expressed by officials of the Department of Justice that it would indirectly allow legislatures to invoke s. 33 in relation to provisions that were not meant to be subject to it, such as ss. 3–6 and 16–23.[86] Accordingly, the result of the change was to ensure s. 28 could not itself be targeted by s. 33 and effectively solidify the place of the provisions not subject to s. 33.

Froc, however, goes one step further and concludes that the historical context bolsters her view that s. 28 is absolute with respect to all of the underlying supra-legislative rights and freedoms with which it can be tied. In my view, this leap does not necessarily follow from the text and history and stands in some tension with the scheme of the Canadian Charter. As mentioned above, Froc notably points to some officials of the Saskatchewan delegation to support her position.”[87] However, she concedes that this understanding was not “uniformly held amongst the politicians,”[88] pointing notably to the fact that some provincial premiers did not share this view. I acknowledge that the feminist activists from 1981 might object to my specific understanding of s. 28, but my view is that, even if this subjective intention did exist, it did not ultimately manifest in the text and scheme of the Constitution.

The argument, then, is that s. 28 does not eviscerate the function of s. 33 in relation to the provisions with which it is concerned, i.e., to provide legislatures with an opportunity to maintain the operation of their laws notwithstanding ss. 2 and 7–15. The result is that, when the underlying supra-legislative constitutional provision has been targeted by a legislature pursuant to s. 33, s. 28 of the Canadian Charter can indirectly play no role to invalidate legislation. There is simply nothing to protect equally because the underlying supra-legislative provision has essentially been withdrawn. This is to say that s. 28 cannot be of any utility to invalidate legislation when s. 33 has been invoked in relation to ss. 2 and 7–15 because it is not legally possible to guarantee the equal exercise of a supra-legislative provision that is reasonably limited or inapplicable. In other words, there was no need to include s. 28 within s. 33 to permit legislation to fully operate notwithstanding ss. 2 and 7–15, since s. 28 is not an independent guarantee of gender equality, as explained previously. However, when the notwithstanding clause is invoked, the s. 28 guarantee of gender equality can continue to play a role for the provisions that are not or cannot be subject to s. 33, notably democratic (ss. 3–5), mobility (s. 6) and language rights (ss. 16–23). This is so even when the relevant legislature invokes the notwithstanding clause in relation to the gender equality guarantee outlined in s. 15. In my view, this is the proper legal significance of the 1981 “Kitchen Accord” and subsequent developments.

This means that I agree with St-Hilaire and Honickman with respect to the legal effect of s. 33 in the contexts in which it can be invoked.[89] Unlike them, I would not deprive s. 28 of its autonomous substantive function in reaching this conclusion. Significantly, this understanding applies whether or not one accepts the view advanced by Webber that the use of s. 33 does not prevent judicial review.[90] In other words, the debate that opposes Webber to other scholars in relation to s. 33 is immaterial to my understanding of s. 28. Of course, if the Supreme Court ultimately accepts Webber’s view, s. 33 would serve to maintain the operation of laws that apply notwithstanding ss. 2 and 7–15. However, if judicial review is foreclosed as others have argued based in part on the precedent in Ford v. Quebec (Attorney General),[91] the view advanced here is that s. 28 would not be engaged in any event because there would be no supra-legislative right or freedom guaranteed by the Canadian Charter with which to tie it.

Consequently, if a legislature invokes s. 33 to prevent the application of s. 15 to invalidate legislation, this will not negate the substantive role played by s. 28 in securing the equal exercise of the other supra-legislative rights and freedoms to male and female persons. Similarly, if s. 33 is used to prevent the application of ss. 2 and 7–15, s. 28 will continue to safeguard the equal exercise of the other provisions that cannot be rendered inapplicable, such as ss. 3–6 and 16–23. Indeed, the historical context teaches us that an important purpose of the changes made to ss. 28 and 33 was to solidify the place of these provisions in the Canadian constitutional order, thereby ensuring that the rights and freedoms that shape the core of Canada’s democratic and federal nature are guaranteed equally to male and female persons through the courts.

This view of the relationship between ss. 28 and 33 is bolstered by considering the relationship between s. 28 and ss. 1 and 32 of the Canadian Charter. There is an important debate amongst scholars on the application of both provisions in this context, as the Superior Court of Quebec acknowledged in Syndicat de la fonction publique du Québec inc. v. Québec (Procureur général).[92] Indeed, while Froc maintains her view that s. 28 is absolute in relation to s. 1, she does not carry this argument over to s. 32. This tends to suggest that there are good arguments on both sides of the debate on the application of s. 28 in relation to other constitutional provisions. My understanding of s. 28 is like the one advanced by scholars who acknowledge its substantive function, yet nonetheless argue that ss. 1 and 32 remain relevant in this context. I contend that this reasoning can be extended to s. 33 as well.

It has been argued that since s. 28 is “absolute,” it applies to block s. 32 as well.[93] Section 32 of the Canadian Charter provides as follows :

  • 32. (1) This Charter applies

    • (a) to the Parliament and government of Canada in respect of all matters within the authority of Parliament including all matters relating to the Yukon Territory and Northwest Territories ; and

      (b) to the legislature and government of each province in respect of all matters within the authority of the legislature of each province.

  • (2) Notwithstanding subsection (1), section 15 shall not have effect until three years after this section comes into force.

  • 32. (1) La présente charte s’applique :

    • a) au Parlement et au gouvernement du Canada, pour tous les domaines relevant du Parlement, y compris ceux qui concernent le territoire du Yukon et les territoires du Nord-Ouest ;

      b) à la législature et au gouvernement de chaque province, pour tous les domaines relevant de cette législature.

  • (2) Par dérogation au paragraphe (1), l’article 15 n’a d’effet que trois ans après l’entrée en vigueur du présent article.[94]

If the view of s. 28 advanced by these scholars were correct, this would entail that s. 28 could safeguard the equal exercise of rights and freedoms to male and female persons in both public and private spheres. In other words, while s. 32 (1) directs that the rights and freedoms in the Canadian Charter are limits on state action, s. 28 is absolute and is therefore not limited to state action. This would also lead to the result that the independent guarantee of gender equality would have immediate effect, rather than having effect “three years after this section comes into force,” as contemplated by s. 32 (2). Froc rejects the contention that s. 28 is absolute in this context and emphasizes the need for contextual interpretation.[95] In her view, since the purpose of s. 28 is to “[block] the effect of certain provisions insofar as their application result in unequal rights,”[96] it is not absolute in relation to s. 32.

I share Froc’s view that s. 28 must be understood in context and is not absolute in relation to s. 32, but we differ somewhat in our understanding of how s. 32 applies in this context. In my view, s. 32 does indeed have the effect of allowing for unequal rights : since it states that the Canadian Charter applies only in the public sphere, it prevents the possibility of securing equal rights in the private sphere. In that sense, it is less ambitious than the Quebec Charter. Section 32 of the Canadian Charter also prevents the s. 15 gender equality guarantee from having immediate effect. These are significant limitations on supra-legislative gender equality rights. If s. 28 is absolute, then it should in theory have the effect of also completely blocking s. 32. However, following the logic outlined above, my view is that s. 32 applies to the underlying provisions to which s. 28 must be tied. Accordingly, while s. 32 cannot be applied directly to s. 28, the latter provision is indirectly concerned with state conduct because it is a guarantee tied to other supra-legislative rights and freedoms that are only held against the state. Similarly, the independent gender equality guarantee could not apply immediately when the Canadian Charter came into force because s. 32 (2) prevented this possibility in relation to the underlying provision. It withdrew it from immediate application. In other words, s. 28 is not absolute. However, I share Nicole Duplé’s view that this would not have precluded a s. 28 claim in relation to other provisions.[97]

This understanding of the relationship between ss. 28 and 32 is also consistent with the case law, namely the Nova Scotia Court of Appeal’s decision in Re Boudreau and Lynch, which confirmed that the guarantee of gender equality in s. 15 could not circumvent s. 32 (2) and have immediate effect, despite the presence of s. 28.[98] The non-independent status of s. 28 is critical to this reasoning. I acknowledge, however, that a scholar like Froc might respond by also indicating that s. 32 was not mentioned as part of the discussions that led to the adoption of s. 28 in its current form, unlike ss. 1 and 33. I concede that this context can form part of the analysis and that the point I am making in relation to s. 32 is less persuasive as a result, albeit not entirely irrelevant to an understanding of the constitutional scheme.

That brings me to s. 1. Some scholars, such as Froc, are of the view that s. 1 does not apply in this context. Others, such as Duplé and Mary Eberts, have expressed the view that s. 1 does apply.[99] Froc’s reasoning is, of course, focused on the text, but it is informed by the historical context. She begins by observing that the initial proposal for s. 28 would have ensured that “the rights and freedoms under the Charter are guaranteed equally to men and women with no limitations.”[100] She explains that there were indeed attempts to amend s. 1, but this was not ultimately secured.[101] Instead, Froc’s view is that the use of the “notwithstanding” language in s. 28 was meant to “preclude use of the rights limitation mechanism in cases of sex discrimination.”[102] She acknowledges that there is “some ambiguity as to whether the import of ‘notwithstanding’ and particularly its effect on section 1 was specifically addressed.”[103] However, she ultimately concludes that correspondence from the Department of Justice indicating that s. 28 was a “significant provision” and later events surrounding s. 33 “constitut[e] an acknowledgement” of her interpretation.[104]

I have a somewhat different view of s. 28 based on the logic outlined above. I share the position of those who suggest that s. 1 remains relevant in this context. Somewhat like s. 33, s. 1 assists in establishing the scope of the independent supra-legislative rights and freedoms outlined in the Canadian Charter.[105] Section 1 provides that the Canadian Charter “guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.”[106] Because s. 28 is a guarantee tied to other supra-legislative rights or freedoms, reasonable limitations are possible. This is consistent with one view of the historical context. For instance, Roger Tassé, Canada’s Deputy Minister of Justice at that time of patriation, reasoned as follows : “[S]ection 28 says notwithstanding anything in this Charter. So what does it mean ? I don’t think it means absolute equality of treatment between men and women. But it does mean that it would not be possible to use section 1 to limit the right guaranteed by section 28.”[107] In other words, I understand him to be saying that s. 1 cannot, of course, apply directly to s. 28, but this does not mean that s. 28 is absolute in relation to other constitutional provisions. On my view, then, s. 1 applies indirectly because it applies to the underlying supra-legislative rights and freedoms with which s. 28 must be tied. If the underlying provisions are reasonably limited, there is nothing for s. 28 to guarantee equally. Once again, the non-independent status of s. 28 helps to understand this legal result.

This view is also supported by the jurisprudence, which has considered s. 1 in the context of gender equality claims brought pursuant to s. 15 of the Canadian Charter—a point Froc herself readily acknowledges.[108] In addition, in R. v. Hess, the Supreme Court held that s. 1 is relevant with respect to the other provisions that are guaranteed equally by s. 28.[109] This means that the view that the limitations clause remains relevant in this context is bolstered by constitutional practice, which has not understood s. 28 as being absolute. In other words, the Canadian Charter does not provide for boundless rights. It guarantees rights or freedoms as reasonably limited.[110]

In my view, however, the justification offered must not relate to gender discrimination per se, but rather to the public interest sought to be pursued generally and its relationship to the other rights or freedoms at issue.[111] This follows from the fact that s. 1 cannot apply directly to s. 28. In practice, this means that the use of s. 1 will likely be more difficult to justify in direct discrimination cases. The story is perhaps more complex with respect to indirect discrimination.[112] For instance, courts have generously interpreted the guarantee of freedom of expression and much of the legal work is left to the limitation stage of the analysis.[113] Accordingly, in my view, if an individual attempted to show, for example, that a provision criminalizing hate speech infringed the equal exercise of the freedom of expression through an indirect discrimination claim, s. 1 could be used to justify a provision criminalizing hate speech.[114] In these circumstances, s. 1 could be used to attempt to justify the limit to the guarantee of freedom of expression, which therefore indirectly limits the guarantee provided in s. 28. In my view, this example indeed shows that s. 28 cannot be absolute as a practical matter. As mentioned above, I do not aim to enter this debate at length in this article. I rather seek to highlight the doctrinal tensions that exist and leave a more in-depth examination to further doctrinal and jurisprudential development.

This brings me back to s. 10 of the Quebec Charter and its utility in understanding the relationship between ss. 28 and 33. As I have explained, s. 10 helps to understand that s. 28 is a non-independent guarantee of the equal exercise of other provisions by male and female persons. This non-independent status has a direct bearing on the legal effect of s. 33. In the context of s. 10 of the Quebec Charter, it must be acknowledged that it is directly subject to the notwithstanding clause outlined in s. 52 of the Quebec Charter. Accordingly, contrary to the Canadian Charter, there is no special status accorded to gender equality in Quebec law that would put the norm beyond the reach of legislative action.[115] Of course, as previously mentioned, the Quebec Charter includes s. 50.1 since 2008. While its nature may be different in light of the context surrounding its adoption and its place in the scheme of the Quebec Charter, I would simply point out, for the purposes of this article, that, assuming it has a substantive nature similar to s. 28 of the Canadian Charter, there is no indication that the change was meant to or could affect the substantive law in Quebec. The existing state of Quebec law is that the remedy of invalidity can only be obtained with respect to violations involving ss. 1–38 of the Quebec Charter.[116] This is not to say that the Quebec Charter should be set aside cavalierly, especially given its status in Quebec’s legal order,[117] but simply that the remedy of invalidity is more circumscribed. However, like s. 28 of the Canadian Charter, s. 10 of the Quebec Charter cannot be directly subject to the limitations clause. This flows from the scheme of Quebec’s Charter because the limitations clause in s. 9.1 was placed in a different chapter.[118] Therefore, words like “Notwithstanding anything in this Charter” were not needed to achieve this result with respect to the Quebec Charter’s limitations clause.

Returning to s. 10 generally then, as Robitaille helpfully explains, where it does apply, the right to the equal exercise of other rights and freedoms will always be dependent on the underlying right or freedom to which it is attached. Accordingly, “if the legislator is justified in not granting a right or freedom in a particular case and for that reason the person complaining cannot legitimately claim the right or freedom in a particular case, section 10 is no longer applicable since it is, in its essence, dependent on the other provisions of the Charter.”[119] This includes contexts in which the limitations clause cannot apply to s. 10. As Mélanie Samson observes, the limitations clause in the Quebec Charter “applies indirectly” in this context because it applies to the underlying right or freedom.[120] This means that where the underlying provision is reasonably limited or inapplicable, so too is the availability of the equality guarantee with respect to that provision. Significantly, the Supreme Court has adopted this understanding of the provision.[121] In other words, while courts have rarely engaged with s. 28, they have been asked to determine the scope of s. 10 of the Quebec Charter and its application in relation to the underlying rights and freedoms. In doing so, the Supreme Court has always held without apparent controversy that a justified limit on an underlying provision indirectly leaves no role for s. 10.

This entails that adopting the interpretation of s. 28 advanced here does not require significant developments in the law or the application of untested constitutional standards. Rather, it requires a recognition that the doctrinal and jurisprudential understanding of Quebec’s distinct equality guarantee can be useful to cases involving the Canadian Charter when s. 15 does not apply. This does not make the substantive guarantee meaningless, as the equality jurisprudence in Quebec and Europe demonstrates. It merely entails recognizing that some provisions may be legitimately limited or rendered inapplicable in certain circumstances. Overall, this demonstrates that when s. 33 is properly invoked to indicate that a law operates notwithstanding certain provisions of the Canadian Charter, s. 28 simply has no utility with respect to the underlying supra-legislative rights and freedoms guaranteed in the provisions targeted by the clause. Section 28 can only play a role in invalidating laws with respect to the other supra-legislative rights and freedoms enshrined in the Canadian Charter. Although this may limit the availability of s. 28 in some cases, its substantive reach with respect to ss. 3–6 and 16–23 should not be discounted. This is an important distinguishing feature between Quebec’s Charter and the Canadian Charter.

5 Charting a Middle Ground : Section 28 and the Canadian Constitution

I now turn to the Canadian Constitution more broadly and its relevance in understanding s. 28.[122] In this part, my aim is to explain that the understanding of s. 28 offered in this article also gives due weight to two constitutional principles behind s. 33 : democracy and federalism. I begin by observing that the genesis of s. 33 is constitutionally significant and calls for caution in constitutional analysis, before describing the function of s. 33 and its relationship to these constitutional principles over time. This then leads me to emphasize s. 28’s role in contexts in which s. 33 is not or cannot be invoked, namely its relevance in securing the right of rights in a democratic society : participation.

As the Supreme Court has explained, constitutional amendments without provincial consent are not permitted for aspects of the Constitution that engage provincial interests. This follows from the federal nature of the Canadian Constitution.[123] The Court has, at times, used the expression “fundamental term or condition of the union” to describe aspects that cannot be unilaterally amended.[124] In this context, it has also made clear that the “architecture of the Constitution” cannot be amended without provincial consent.[125] Section 33 of the Canadian Charter is certainly one such aspect of the Canadian Constitution,[126] especially if the union’s fundamental terms or conditions are understood in dynamic rather than static terms.[127] In Vriend v. Alberta, the Court also recognized that Canada’s constitutional architecture cannot be understood without accounting for s. 33’s place in that structure. It observed that the notwithstanding clause establishes that, in some contexts, “the final word in our constitutional structure is in fact left to the legislature and not the courts.”[128] More recently, in Toronto (City) v. Ontario (Attorney General), the Court held that s. 33 is a fundamental part of the “constitutional bargain” and dismissed attempts to “circumven[t]” it.[129] Accordingly, some caution is in order when engaging with legal issues that involve s. 33 because it is the provincial condition on which the modern system of judicial review is erected in Canada.

This brings me to discuss the function of s. 33 more directly and its relationship with constitutional principle. While the adoption of the Canadian Charter presented a certain break from Canada’s constitutional tradition, this change was not meant to be complete. The judiciary, which had until that point enforced the structural aspects of the Canadian Constitution, most notably federalism, was given an additional role in 1982 : to protect supra-legislative rights and freedoms.

However, it was acknowledged that the rights and freedoms enshrined in the Canadian Charter did not represent an exhaustive list, as confirmed by s. 26.[130] For example, many economic and social rights were not explicitly included in the Canadian Charter, although some debate still surrounds these issues.[131] In addition, as others have explained more fully, many of the rights and freedoms entrenched existed before the Canadian Charter was adopted and were given a new status in 1982.[132] It was also acknowledged that the judiciary might err in its interpretation of these provisions, or fail to give sufficient weight to competing rights and interests in understanding their scope through s. 1. The provinces therefore added a mechanism to ensure the political branches could disagree with the courts or continue to express their understanding of rights and freedoms as they deemed necessary. In addition, it was agreed that the declaration would have to be renewed through the legislative process every five years to ensure accountability.

The adoption of the notwithstanding clause, then, was an attempt to maintain and affirm, at least in part, Canada’s tradition of democratic self-government, which explains why some scholars refer to it as the “parliamentary sovereignty” clause.[133] It does not represent a complete break from the past, but rather an attempt to preserve and affirm elements of the former constitution in the modified constitutional order created in 1982.[134] As Dwight Newman has shown, the context surrounding the adoption of s. 33 is far from unprincipled and demonstrates that provincial politicians from across the political spectrum expressed a concern to “fit together the traditions of Canadian parliamentary democracy and the entrenched rights model.”[135] Constitutional theorists and comparative law scholars have also observed that the Canadian model of weak-form judicial review offers a nuanced response to the problems that can be associated with strong-form judicial review.[136] Indeed, in the context of this article, it should be noted that Canadian history shows that invoking the notwithstanding clause may in fact be required to advance equality interests in certain circumstances.[137] This is consistent with what the Supreme Court affirmed in another context, i.e., that the fact that constitutional provisions are “the product of negotiation and political compromise” does not mean they are “unprincipled.”[138]

This is also significant in a federal system, where provincial autonomy is embedded in the constitutional structure.[139] Indeed, beyond issues involving the division of powers that directly implicate the federal structure, the principle of federalism remains relevant to generally understanding Canada’s Constitution.[140] Diverse approaches to a range of issues falling within provincial jurisdiction are therefore not only desirable, but also foundational. The Supreme Court itself has, on occasion, recognized this in its case law on s. 1 of the Canadian Charter,[141] echoing approaches adopted around the world to accommodate pluralism in human rights adjudication.[142] Accordingly, the reach of the notwithstanding clause extends to preserving federal democracy in circumstances in which a legislature decides that courts have not given or may not give sufficient weight to the constitutional significance of diversity.[143]

This is particularly compelling with respect to Quebec, which has not formally given its assent to the Constitution Act, 1982. While it might sometimes be important to distinguish between questions of legality and legitimacy, these issues are not watertight as a matter of law.[144] Although this article does not permit a full account of this issue, suffice it to note that scholars have discussed the uneasy constitutional situation stemming from the decision to proceed without Quebec’s consent in 1982.[145] This is especially so following the Secession Reference, which, as I have observed elsewhere, led the Supreme Court to recognize that “a failure to permit a federated entity like Quebec, for example, to freely pursue internal self-determination, which includes ‘economic, social and cultural development,’ may ground a right of secession at international law.”[146] Relatedly, the Court also explained that, as a matter of domestic constitutional law, faced with a decision of a clear majority on a clear question, a province could withdraw from the union, and the federation’s other partners would then have a good faith obligation to achieve that end. While the Court has indicated that Quebec is legally bound by the Canadian Constitution as a whole as long as it remains within Canada,[147] my understanding of s. 33 preserves limited autonomy that Quebec has not agreed to give away in the first instance and is consonant with the Secession Reference. Accordingly, when Quebec legislators are of the view that a court’s application of s. 1 does not or may not adequately give voice to Canada’s economic, social and cultural differences, s. 33 can be invoked to do so and assert an alternative understanding of rights and freedoms. Each provision serves the same function in related yet distinct ways.

In the end, beyond our differing understanding of s. 28, the argument is that Froc’s approach may not give sufficient weight to these principles because it has the effect of encouraging the circumvention of uses of s. 33 with respect to the provisions it mentions, thereby depriving the notwithstanding clause of its purpose. The approach advanced here seeks to demonstrate that every constitutional provision has a function within Canada’s constitutional system and must be allowed to play its role. The purpose of s. 33 is to preserve the ultimate say for Parliament and the provincial legislatures with respect to questions involving ss. 2 and 7–15. In doing so, the provision has a role to play in what Sébastien Grammond has called “compact-mending.”[148]

However, my view also assigns weight to s. 28 because it recognizes that, where ss. 2 and 7–15 are not engaged, or when s. 33 is surgically used with respect to specific provisions, gender equality has an autonomous substantive function. Indeed, I would emphasize that s. 28 can play a substantive role in protecting gender equality through the courts, namely with respect to the voice, exit and language rights contemplated by ss. 3–6 and 16–23. In this regard, Froc’s research serves as a reminder that s. 28 was part of the constitutional settlement reached in 1982.

This is particularly compelling with respect to the rights in s. 3 that lie at the heart of democracy within both orders of government. It reflects the conception of democracy—and the role of men and women in that system—that has emerged since 1867 and that was a settled feature by 1982, including in Quebec.[149] Although there can be theoretical disagreement amongst scholars over these issues,[150] it might also be said that having courts apply s. 28 in these circumstances can ensure that the decision to invoke the notwithstanding clause is based on a democratic mandate from a wide cross-section of society and that this same cross-section can hold the government accountable for its continued use of the clause should that be necessary. This is not to say that s. 1 has no role to play in the analysis in appropriate cases. Indeed, if one considers the Hak case, for example, scholars have observed that similar measures have been upheld by the European Court of Human Rights in certain contexts, including in cases in which the equality guarantee in art. 14 of the European Convention was at issue.[151] Rather, my argument is that, with respect to these specific provisions, the judiciary is on firmer constitutional and theoretical ground when it does seek to intervene.[152]

6 Charting a Middle Ground : Recent Jurisprudential Developments on Section 28

I conclude this article with the observation that the middle ground position articulated in it is now also reflected in the case law because the trial judge in Hak appears to have adopted a third way on the merits, although the analysis is admittedly confused at times. Relatedly, writing in the context of the appeal from the denial of an interlocutory injunction in Hak, Bélanger J.A., while not stating a strong view on the matter, observed that “it is not clear that section 28 of the Charter precludes the Quebec legislature from invoking the notwithstanding clause.”[153]

In Hak, the trial judge appears to have concluded, contrary to the predominant jurisprudential view, that s. 28 is a substantive provision, notably because its text read in context suggests that it is a “guarantee.”[154] In doing so, he rejected the view adopted by the British Columbia Court of Appeal in McIvor. However, consistent with the approach advanced in this article, the court was not prepared to conclude that s. 28 was an independent provision because it is specifically tied to the “rights and freedoms referred to in [the Canadian Charter].”[155] In its words, it has an “autonomous scope.”[156] This aspect of the court’s reasoning is consistent with my view that s. 28 is an autonomous—yet not independent—substantive provision.

Based on this understanding of the nature of s. 28, the court then held that “to the extent that the legislator withdraws certain rights or freedoms from constitutional protection by using the notwithstanding clause in section 33, there is no longer a substrate of rights or freedoms on which section 28 can then be applied to guarantee this equality between men and women.”[157] This means that, in substance, the court concluded that there was no utility for s. 28 with respect to the provisions targeted by s. 33. It came to the same conclusion in relation to s. 50.1 of the Quebec Charter.[158] This is consistent with my view that s. 28 is not absolute.

I would add that the court observed that, where s. 28 does apply, it cannot be used to “obliterate other rights or freedoms that exist in Canada.”[159] For the court, these interests can be accounted for under s. 1 in appropriate cases. This is also the view that was adopted by Mainville J.A., writing on the appeal of the denial of an interlocutory injunction in Hak. However, he only considered s. 1 to be relevant in the alternative, if s. 28 is a “substantive provision.”[160] This is compatible with my view of the relationship between ss. 1 and 28 of the Canadian Charter and is consonant with the Supreme Court’s comments in Hess. The trial judge in Hak similarly concluded that s. 28 could not be absolute given the effect of this conclusion on s. 32 of the Canadian Charter.[161] This analysis bolsters my understanding of the relationship between ss. 28 and 33. My approach would also make clear that none of these provisions can be applied directly to s. 28 if one is to give it proper meaning.

However, in concluding that s. 28 is not absolute, the court discounted any analogy with s. 10 of the Quebec Charter.[162] Moreover, in endorsing Syndicat de la fonction publique du Québec inc. v. Québec (Procureur général), the court appeared to view the case as holding that s. 28 “marks the interpretation of section 15.”[163] This has led some scholars to conclude that the trial judge held that s. 28 is “interpretive.”[164] To the extent that the court meant to deny the autonomous substantive nature of s. 28, I believe this should be rejected for the reasons laid out in this article. For the sake of completeness, it should also be noted that s. 28 was only raised by the parties in relation to ss. 2 and 15 of the Canadian Charter in that case. Accordingly, while ss. 3, 6 and 23 of the Canadian Charter were also before the trial judge, he was not asked to consider s. 28’s relevance in this context.

I find support for my position that the court adopted a middle ground because it concluded that, when s. 28 is properly invoked, s. 1 would apply. This assumes that s. 28 is a substantive guarantee. Froc acknowledges that this is the result of the trial judge’s decision, which “avoids the doctrinal pitfalls” of the “interpretive” approach.[165] She nonetheless criticizes his finding that s. 28 is not absolute and her comments therefore bring us back to the debate about the nature and application of s. 28 discussed above. Froc is notably critical of his finding that “section 28’s autonomous scope cannot extend to independently invalidating legislative provisions.”[166] If Froc means to suggest that s. 28 is an “independent” guarantee, I disagree. If by “independent” she only means to disagree with his finding that s. 28 is not “absolute” in its relationship with s. 33, our difference of opinion is lessened. In this latter case, I merely disagree with the proposition that understanding s. 28 as “absolute” is the inevitable result of a proper understanding of the provision and its relationship with the constitutional scheme.

Conclusion

In sum, my view is that s. 28 of the Canadian Charter is a substantive guarantee, but its nature and application are more limited than some scholars assume. This is to say that Froc is correct in part and the scholars who have critiqued Froc’s position are also correct in part. It would diminish the work of women leading up to patriation to relegate s. 28 to a mere interpretive provision. Interpreted in its historical context, the language suggests the creation of a substantive guarantee. At the same time, its substantive reach is not unlimited, nor can it be read in isolation. It must be understood purposively and harmoniously with the scheme of the Canadian Charter and the Canadian Constitution generally.

Section 10 of the Quebec Charter helps to understand, then, that s. 28 is an autonomous—yet not independent—guarantee. It safeguards the equal exercise of the supra-legislative rights and freedoms guaranteed under the Canadian Charter to both male and female persons. However, where an underlying right or freedom is appropriately limited or rendered inapplicable, either through s. 1 or s. 33, then s. 28, while still operative, has nothing to protect equally for male and female persons. In the first case, the underlying provision is reasonably limited by s. 1, whereas, in the second case, the legislature renders the underlying provision inapplicable or articulates its own conception of the right or freedom by allowing laws to operate notwithstanding the underlying entrenched provision.

Accordingly, s. 28 can continue to play a substantive role with respect to the provisions that are not or cannot be subject to s. 33. However, when s. 33 is properly invoked, s. 28 cannot overcome the declaration with respect to ss. 2 and 7–15 and serve to invalidate legislation. It continues to apply, but has no utility in that specific context. Although this middle ground may not be satisfying to either camp, in my view, it appropriately draws from both when it comes to interpreting s. 28. Significantly, it respects the purposes of both ss. 28 and 33 in the Canadian constitutional order and preserves a meaningful role for them within their proper ambit.