Reasonable Limits

How Far Does Religious Freedom Go in Canada?

Special Podcast Episode

Join Father Deacon Andrew Bennett and NextGEN Fellow Kristopher Kinsinger in an extended conversation about the polarizing debate that this report addresses.  

Executive Summary

Freedom of religion is one of the fundamental freedoms enumerated in the Canadian Charter of Rights and Freedoms. Advocacy of religious freedom is often misunderstood as calling for positions that would be difficult for anyone to reasonably accept. A more nuanced discussion of this right can dispel some of this misunderstanding. This paper examines the “reasonable limits” clause in section 1 of the Charter, which states that the Charter “guarantees the rights and freedoms set out in its subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.”

Canadian courts have treated the concept of “reasonable limits” within the Charter. The dominant treatment has been in the form of what is often called a proportionality analysis, and is often linked to a particular Supreme Court of Canada case that results in the Canadian analysis known as the “Oakes test.” This test, a framework for proportionality analysis arising from the case of R. v. Oakes in 1986, has four parts. These are “pressing and substantial objective,” “rational connection,” “minimal impairment,” and “final balancing.” Once an initial infringement of a right or freedom has been established, these four parts of the Oakes test function together to determine whether a legal limit on a right or freedom has met the standards of proportionality so as to be demonstrably justified. A law that fails to meet any of the four parts of the test is considered to have resulted in a violation of Charter rights and thus to give rise to a constitutional remedy, such as the striking down of that law. This paper describes these four parts with reference to religious-freedom cases decided in Canadian courts. It also describes some critiques that have been levelled against the Oakes test or, more broadly, against using proportionality to understand reasonable limits on rights and freedoms.

The paper continues with a discussion of recent policy and law affecting religious freedom in ways that, in our judgment, are not in line with a reasonable-limits analysis. These are the values attestation attached to the Canada Summer Jobs Program in 2018, and Quebec’s Act Respecting the Laicity of the State. In the final section of the paper, we examine some government-imposed restrictions relating to the COVID-19 pandemic and evaluate them in light of reasonable limits on religious freedom.

As a pluralist society of citizens with many different viewpoints, it is important for Canada to find ways to engage with major questions of social policy in rich ways that can overcome polarization. The concept of reasonable limits helps with understanding which limits on religious freedom are acceptable and which are unacceptable. Advocacy for rights and freedoms should be coupled with education to inform the public and facilitate discussion of the complex issues involved. Decision-makers should also redouble their efforts to engage with faith communities in order to better understand the issues and pursue policies that have fewer adverse effects on religious freedom.


For various reasons, policy discussions on certain issues have become increasingly polarized. We have seemingly lost some of our societal ability to see possibilities of complex, challenging compromises and to see nuance in debates. This polarization presents broader societal challenges, but it raises particular issues in the context of discussing religious freedom.

Advocacy of religious freedom is often misunderstood as calling for positions that would be difficult for anyone to reasonably accept. A more nuanced conversation can dispel some of this misunderstanding. In the process, those skeptical of religious freedom, including many without religious beliefs, may see that arguments for religious freedom are for a kind of freedom that they support more than they realize.

At the same time, those advocating for religious freedom who themselves come from religious backgrounds do so within worldviews of their own. According to perspectives in those worldviews, they may assert that their faith traditions must be defended and advocated to the utmost without reservation. A richer conversation can help to make clearer why it is not always right to push for religious freedom beyond all limits. Respect for the inherent dignity of others, a value deeply grounded in many faith traditions, requires accepting some limits on the scope of religious freedom.

The idea of reasonable limits on rights and freedoms is an important one for everyone and relates to rights and freedoms more broadly. By their very nature, constitutional rights and freedoms warrant particular attention and priority. But no right can be unlimited. Issues arise in which one right ends up in a full-fledged contradiction with another right, and it is not clear how to reconcile the two. And issues arise regarding how other interests of a free and democratic society are to be achieved in our shared life together. Understanding rights and freedoms requires also understanding the limits on them, and it is thus important for everyone.

The challenging question in the context at hand is how to understand reasonable limits on religious freedom. That question is one for much philosophical reflection and for theological discussion within faith traditions. But, in terms of how we live together in Canada right now, it is a subject for legal discussion and analysis. That kind of legal discussion may help with further philosophical and theological engagement with the question. But a legal analysis also determines what practically happens with the scope of religious freedom under our constitutional arrangements. It is important to consider where a well-grounded legal analysis of reasonable limits leads.

There may be different ways in which reasonable limits on religious freedom come to be understood in different times and places. Because of combinations of principles and circumstances, we are speaking here of complex policy questions rather than universal truth, even while there might be one such truth on some aspects of the questions. But engaging with legal principles concerning reasonable limits on religious freedom in Canada today helps with thinking about analogous questions elsewhere as well.

This paper examines how the Canadian courts have treated the concept of reasonable limits within the Canadian Charter of Rights and Freedoms generally, though with a particular focus on how they have treated reasonable limits in the context of religious freedom. The dominant treatment has been in the form of what is often called a proportionality analysis—and often linked to a particular Supreme Court of Canada case that results in the Canadian analysis being called the “Oakes test.” 1At the same time, there are complex currents of discussion about difficulties with the assumptions embedded in this sort of analysis, and the paper will also highlight how there have been some calls for a modified approach to understanding reasonable limits on Charter rights and freedoms, even while recognizing that the dominant approach to proportionality analysis is strongly embedded in current Canadian law.

This discussion will include reference to a number of important religious-freedom cases, with these cases offering examples of particular limits on religious freedom that were considered reasonable or unreasonable. The next part of the paper will highlight some examples of recent laws in Canada that have been out of line with legal understandings of reasonable limits on religious freedom, while explaining how bringing the concept of reasonable limits to bear can facilitate a more nuanced, less polarized discussion of such laws.

The final part of the paper will consider some current issues—notably, ways in which religious freedom has been limited during the COVID-19 pandemic—and it will discuss how an understanding of freedoms subject to reasonable limits can facilitate a better discussion on these issues.

Reasonable Limits and the Oakes Test

Section 1 of the Canadian Charter of Rights and Freedoms provides that the 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.” 2From a constitutional-law perspective, in considering any particular Charter right or freedom it is thus a crucial question what constitutes a “reasonable limit.” This section of the paper attempts to analyze this question while mentioning a number of important religious-freedom cases where constitutional law on reasonable limits has been applied.

Before addressing the main analysis, it is worth noting that a number of leading religious-freedom cases raise some complex related questions of how this whole analysis is to be applied in the context of reviewing decisions of administrative boards and tribunals that affect Charter rights. The Supreme Court of Canada has gone through a period of urging a form of deference to these bodies, and thus not applying the full test discussed here, and that phenomenon has affected a number of leading religious-freedom cases through what is called the “Doré analysis.” 3One example would be the concurring opinion in the Ktunaxa case, a case concerning Indigenous religious-freedom claims in the context of a proposed ski-resort development. 4While the main judgment in the case effectively held that the ski-resort development did not raise religious-freedom issues, two concurring justices held that it did, that there were serious negative effects on Indigenous religious-freedom rights at issue, but that nonetheless the way in which the minister had considered these while carrying out a statutory mandate of allocating Crown lands to various uses was acceptable. 5The way the analysis works here does not directly consider the violation of rights to challenge the decision directly but deferentially looks at whether the minister considered the rights issues in the decision-making process. Some would raise questions about whether such an approach is sufficiently protective of Charter rights and freedoms. 6Whether the Supreme Court of Canada has done the right thing on these issues of how to analyze Charter rights in the context of administrative decisions has many complex dimensions and would warrant its own separate analysis.

Section 1 has been read as what is sometimes called an “external-limitations clause.” Many constitutions do not have an explicit clause referring to limits on the rights they protect, but courts then develop those limits in the course of interpreting each right. Thus, in the United States, there ends up being a slightly different understanding of limits on each specific right or freedom, developed in a jurisprudence pertaining to that right or freedom. By contrast, in Canada, the presence of an external-limitations clause in the written text of the constitution seemingly encouraged the courts to attempt to develop a methodology for understanding limits on rights and freedoms at a more general level. The section 1 test then applies to each right or freedom. 7

In doing so, courts also took the view that thinking of limits on rights and freedoms under that external framework would allow them to read the initial scope of rights and freedoms themselves more broadly. This is because there ends up being a different check on any overly broad conceptions of specific rights and freedoms, with that check involving the government justifying its limits but nonetheless allowing for sensible interpretations of the rights and freedoms at issue. Thus, for example, the courts have ended up reading freedom of expression in very broad ways so that they could insist on government justifications of any limits on expression.

In Canadian law, “reasonable limits” for the purposes of limits on rights have come to be interpreted in terms of what is called a proportionality analysis. While there are more details to understand concerning such an analysis, the basic idea is that governments act based on a variety of policy objectives, including protecting the rights of other people. This may lead to the government enacting policies that ultimately limit the scope of certain rights. These limits are subject to tests concerning whether they are appropriately proportionate as between the effects on the right or freedom that is limited versus what is achieved in terms of the policy goal or protection of others’ rights. This analysis is not a simple utilitarian weighing as it might first sound. Rather, it involves a variety of nuanced questions that consider whether these specific limits on rights and freedoms are actually needed and ultimately acceptable.

The government that has imposed limits needs to justify them, so this sort of proportionality analysis, while accepting the possibility of reasonable limits on rights and freedoms, is ultimately focused on respect for rights and freedoms. Such an analysis demands sufficient justifications for any limits on rights and freedoms. This onus of justification on the government is grounded in the terms of section 1 itself, which specifies that the government must show that the limits in question are “demonstrably justified in a free and democratic society.” Any failure to advance an adequate argument for a justification will lead to a conclusion against the government. 8

Analyses of proportionality as a way of engaging with tensions between different moral and legal considerations have deep foundations in Western legal, cultural, and religious traditions. Parts of the Hebrew Bible concerned with warfare set out principles of proportionality in how warfare was to be conducted so as to limit negative effects on civilian populations (Deuteronomy 20). This was of course located within a different cultural context in which permitted practices still appear harsh to us today, but it represented improvements within prevailing cultural contexts. The just-war tradition within Jewish rabbinical and Christian scholastic thinking continued and built on this concept over the centuries. 9Modern human-rights frameworks have similarly embraced this concept, with significant use of proportionality analyses especially in German law, European human-rights law, Canadian constitutional law, and in other systems influenced by these jurisdictions, including Israel and Commonwealth states such as South Africa. 10

Seemingly influenced in part by European human-rights law approaches, the Supreme Court of Canada developed a framework for proportionality analysis here in 1986 in the case of R. v. Oakes. Although the parts of that test are sometimes put into subgroups, it is easiest to understand the test as having four parts. These are often called “pressing and substantial objective,” “rational connection,” “minimal impairment,” and “final balancing.” Once an initial infringement of a right or freedom has been established, these four parts of the Oakes test function together to determine whether a legal limit on a right or freedom has met the standards of proportionality so as to be “demonstrably justified.” A law that fails to meet any of the four parts of the test is considered to have resulted in a violation of Charter rights and thus to give rise to a constitutional remedy, such as the striking down of that law.

We can say a bit more about each of the four components of the Oakes test. 11First, the “pressing and substantial objective” portion of the test is concerned with whether the government objective at issue is sufficiently important so as to potentially justify the limitation of a right. Obviously, protecting the rights of somebody else will qualify, but so may other sufficiently important government objectives. Since governments usually limit rights for important purposes rather than at random, this part of the test has seldom come into play despite remaining an important part of it. Notably, there could be an unconstitutional purpose if the very purpose of a law were to do something contrary to a Charter right or freedom. The Supreme Court of Canada held as such in relation to a Sunday closing law meant to enforce through the legal system the religious obligation of one faith where there was evidence of the historical purposes of the law to that effect. 12

There could also be a problem on this part of the test if limits were imposed on religious freedom for trivial reasons. The Amselem decision of the Supreme Court of Canada in 2004 was technically a case arising under the religious-freedom provisions of Quebec’s Charter, which applies to private parties, rather than the Canadian Charter, which does not. But the Court’s analysis of religious freedom has come to be influential for the religious-freedom analysis under the Canadian Charter. The facts of the case concerned a condo board’s restrictions on the construction on balconies of temporary succahs (dwelling huts) during the nine-day Jewish festival of Succot. 13The Supreme Court of Canada held that such a limitation on religious freedom offered only trivial benefits for other owners and was thus unreasonable. 14

Second, the “rational connection” component of the test says that the law must be logically connected to the furtherance of the government objective at issue. The requirement of a rational connection is not a demanding standard, and this part of the test has also not been applied often, but it does potentially say that the courts could strike down a law that was obviously counterproductive to its goals, since there would then be a limitation of a right or freedom for no reason. Many of the cases where courts have used this part of the test have been critiqued—it is, after all, peculiar to end up with a conclusion that what the government did has no logical connection to its objectives. 15Examples do exist, however. In one case, a law imposed different security requirements on citizenship processes of children born abroad to Canadian mothers as opposed to those born abroad to Canadian fathers; the distinction between mothers and fathers in this context had no rational connection to the objectives of the law. 16But such situations will tend to arise mainly with a law that has simply been poorly drafted and should not be a common issue under the Oakes test.

The third component of the test, traditionally called “minimal impairment,” has been the site of the most discussion in the most cases.

The third component of the test, traditionally called “minimal impairment,” has been the site of the most discussion in the most cases. Read strictly, it would say that a law must impair rights and freedoms no more than necessary in order to achieve the objective that it sets out to achieve. If there were a way to achieve the same objective without limiting rights and freedoms as much, it would follow that the limitation at issue was unconstitutional because it interfered excessively with rights. Sometimes, then, lawyers put before the courts ideas about hypothetical alternative policies—or even real policies in other jurisdictions—that could achieve the same goals while having fewer negative effects on rights and freedoms. 17On the strictest reading of this part of the test, where they can convince a court that a particular alternative policy will achieve the same objectives, they are entitled to have the court declare the overly intrusive policy unconstitutional.

Matters on this part of the test are more complex, however. To take it to refer literally to the absolute minimal impairment, or absolutely least restrictive means of achieving the objective of the legislation, has the potential to impose an extraordinarily demanding standard and invite courts to make extensive decisions related to detailed policy arenas for which they may be ill-equipped. As a result of such considerations, the Supreme Court of Canada moved toward a position where the “minimal impairment” branch is read in a more qualified way, such that the test considers whether a particular law is reasonably minimally impairing—whether the law impairs rights and freedoms “as little as is reasonably possible” to achieve the objectives. 18

Despite the increasingly deferential tone of this part of the test, inappropriate restrictions on religious freedom have still been struck down under it. For example, in the Supreme Court of Canada’s 2006 decision in the Multani case, the Court held that an absolute prohibition on Sikh high-school students possessing kirpans—ceremonial knives carried by those Sikhs who have been specially initiated into the community known as the Khalsa—was unreasonable given that there could be alternative policies that would ensure that kirpans were possessed in a way that prevented any risk of their misuse. 19

A 2009 Supreme Court of Canada ruling in Hutterian Brethren, which upheld Alberta’s mandatory photos on driver’s licenses even for a small group of two hundred Hutterites with religious objections to having their photos taken, has been extensively critiqued. The majority justices took the view that there was no other reasonable means of achieving security of the system, while the dissenting justices thought there were possible accommodations that would have made the law more carefully tailored in light of rights and freedoms affected. 20

The fourth component of the test, the “final balancing” component, has been less-utilized. One leading constitutional-law expert has even long suggested that it might be a redundant component, expressing simply a conclusion resulting from the other components. 21However, the Supreme Court of Canada has made clear that it is indeed a separate part of the test and that a law that passed the first three parts of the test could still fail on the basis that its overall detrimental effects on rights were excessive in relation to the policy objectives achieved. 22This conclusion is right in principle. It says that even an important policy objective pursued in the manner with the least detrimental effects on rights could nonetheless be something that cannot constitutionally be pursued if those effects on rights are excessive. In the 2009 Hutterian Brethren case, the dissenting justices suggested that even someone concluding that the law had passed the minimal-impairment step should still reject the law as unconstitutional because of the severity of the effects on the small group of people at issue next to the marginal gain for the province. 23

The Oakes test has always been subject to critique. Some critiques have focused on practical bases concerned with consistency of application, and others have focused on more theoretical bases, such as a concern that the Oakes test treats rights in an excessively utilitarian manner inconsistent with the basic idea of rights. 24Within many conceptions of rights, the very idea of rights is that the claim of a rights-holder takes priority over any sort of utilitarian balancing; if a right did not have such a priority, there would be no need to speak of rights at all, as one could speak simply of a utilitarian balancing based on policy objectives and outcomes. If the Oakes test ceases to treat rights as rights, then this critique would suggest that it has gone astray. There are reasonable arguments that the Oakes test was not the sole means of reading section 1 of the Charter and, indeed, that it might not even fit well with some aspects of the text, especially when considering both language versions of section 1. 25

In recent years, in some ways carrying forward the concern that the Oakes test treats rights in an excessively utilitarian manner, some scholars have critiqued the entire idea of using proportionality in understanding limits on rights and freedoms. Some scholars have suggested that the method of proportionality analysis invites an approach to rights themselves that effectively denigrates what a right is. 26In an extended argument, Francisco Urbina has recently levelled a number of critiques at proportionality analysis, including at the ways in which it may treat rights as interchangeable with other considerations and the ways in which it may invite unconstrained moral reasoning by judges. 27

Some justices of the Supreme Court of Canada have shown appreciation for these sorts of critiques and suggested the possibility of modifying the Oakes test. Indeed, even as long ago as 2009, a dissenting judgment of Justice LeBel in the Hutterian Brethren decision raised the possibility of thinking of the Oakes test more holistically—that is, trying to think more broadly about whether particular legislation strikes an appropriate balance in terms of reasonable limits. 28More recently, some justices have raised questions about whether the Oakes test has invited too much thinking about rights limits in terms of justified infringements rather than reasonable limits and have suggested that various aspects of Charter analysis might need rethinking. 29There might thus be a conception of reasonable limits on rights that is less focused on the Oakes test and more focused on seeing the idea of reasonable limits as guiding the understanding of each right or freedom as being subject to more detailed specification as to its scope.

On either conception, though, it is possible to think of limits on rights and freedoms in nuanced ways and to realize that most rights and freedoms are subject to reasonable limits. Reasonable limits are different from unreasonable limits in being “appropriately tailored”—as a certain metaphor often puts it—in light of other considerations. That is to say, a reasonable limit must be carefully designed in a way that lessens unnecessary or otherwise inappropriate impacts on rights. The metaphor suggests that the “tailoring” or design of a limit could be developed differently so as to achieve other important purposes of governments while respecting rights as best as possible within the sometimes unfortunate tradeoffs in which they must engage. The Oakes test provides one central and still legally entrenched way of understanding whether limits are appropriately tailored. While there have been critiques of the Oakes test, the critics would largely still be considering whether limits were appropriately tailored, with some slight differences in the analyses. Given its ongoing dominance in Canadian law, even while recognizing the possibility of developing understandings over time, this paper will primarily have reference to the Oakes test in understanding when a limit is reasonable or unreasonable. The main reference points will ultimately be whether a limit impairs rights and freedoms as little as reasonably feasible to achieve a pertinent objective, and whether there is an overall proportionality between the reasons for the limits and the extent of the limits.

Under such an approach, case law has thus upheld carefully tailored limits on hate speech as being reasonable limits on freedom of religion (and expression). 30Limits will be considered appropriately tailored where they achieve very important objectives while not limiting religion and expression more than necessary and, generally, not limiting religion and expression in any unacceptably excessive way, while also trying to be clear so as not to “chill” expression that is actually permitted—that is, causing the person to self-limit out of a fear that the expression might not be permitted. Even where somebody attempts to invoke some subjective interpretation of religion in support of his proliferation of speech that makes highly abusive comments intended to devalue certain groups in society, it is a reasonable limit on freedom of religion to adopt carefully tailored restrictions on hate speech. Those advocating for freedom of religion need not support claims to engage in hate speech but may properly suggest that carefully designed limits on hate speech can be reasonable limits.

Laws and Policies Not in Line with Reasonable-Limits Analysis

Canada has a complex history of religious freedom. While in many ways a country in which religious freedom has been a defining value, Canada has at various points in its history nonetheless sufficiently misunderstood some minority religious traditions and beliefs as to impose inappropriate limits on religious freedom. 31Such problems are not merely historical, as two recent examples highlight.

One recent policy that has met with widespread condemnation has been in changes to the Canada Summer Jobs Program, initially launched in 2018. At that time, employers’ applications were effectively made subject to a widely criticized requirement of signing a statement indicating that their “core mandate” did not involve opposition to abortion or to freedom from discrimination based on “sexual orientation or gender identity or expression.” 32The opposition led to changes in the following year’s program, although there would remain a requirement that organizations seeking to participate in the Summer Jobs Program make a declaration that they were not involved in working against any legal rights existing in Canada (which would presumably include abortion), apparently a more acceptable wording for a variety of organizations. 33The way in which this program was administered, however, would end up reflecting government bureaucrats’ making unfounded assumptions about certain organizations, a problem identified in recent court decisions.

While successful lawsuits against the government policy and its application have been based on administrative law, 34they have shown significant hints of considering the policy itself to involve ongoing interference in freedom of religion. For example, in a June 2021 decision involving a denial of Summer Jobs funding to Redeemer University in 2019, the Federal Court suggested that while it could and would decide the case based on administrative-law arguments about procedural fairness, there were preliminary indications that the officials administering the program had discriminated against Redeemer University on religious grounds, 35which would also amount to an interference with freedom of religion. The judge’s decision in the case to grant a costs order on a full-indemnity basis was unusual and spoke to the judge considering the government officials’ conduct highly problematic. As described by Redeemer University lawyer Albertos Polizogopoulos, “I have never seen that in any court, let alone the federal court.” 36

The government objective behind the shifting Summer Jobs policy requirements appears to have been to ensure that those receiving Summer Jobs funding were not using it to advance views on social issues that were at odds with the government’s position on rights established within Canadian law, what the government called “the values underlying the Canadian Charter of Rights and Freedoms as well as other rights.” 37Were that policy to face direct Charter analysis, it is clear that it would infringe on freedom of religion, which includes a right to manifest religious belief. The ability to manifest religious belief includes an ability to argue for positions on controversial issues other than those currently entrenched in Canadian law (something also protected by freedom of expression). That people disagree on issues makes life more complex, and it can be challenging for each of us to encounter people who hold views very different from our own, but they nonetheless have a right to hold and manifest those views.

A right to hold and manifest differing views on controversial issues is protected in various freedoms contained within not only the Charter but also predecessor instruments such as the Universal Declaration of Human Rights (UDHR), adopted by the United Nations in 1948. Among a set of related freedoms, the UDHR recognizes freedom of thought and freedom of opinion. 38Such freedoms always protected the integrity of the human person. The drafting history of the UDHR also shows how conscious the drafters were of the horrors of governmental systems that denied these freedoms. 39Reflecting back on such freedoms, a United Nations report on freedom of thought that went before the United Nations General Assembly in October 2021 saw discussion of the importance of the ability to engage with diverse viewpoints for the “dignity, agency, and existence of the human being.” 40The ability to hold and manifest differing views matters to the ability to seek truth and may thus also serve beneficial ends for society, 41but it also matters to the very integrity of the human person as a human being, both inside and outside of religious contexts. However, where the ability to manifest a religious belief through arguing for a differing position on a controversial issue is limited, there is certainly in the first instance an initial infringement of the right to freedom of religion.

When that initial infringement has been established, the question is whether such a policy adopted for such a purpose would involve unreasonable limits on religious freedom. The answer is that it would. First, the government objective comes surprisingly close to simply imposing consequences on particular religious beliefs and thus verges on being an unconstitutional purpose. Second, to the extent that the purpose is something legitimate, such as continuing to secure certain legally entrenched rights, then it is simply unclear why requiring that organizations receiving standardized subsidies to employ summer students must make attestations of their views on issues, or be subject to various kinds of limits on their views, would be the least rights-impairing way of continuing to secure those rights. Third, at a level of final balancing, it is simply implausible that directly limiting rights to manifest religious beliefs on issues (and limiting expression) is a proportionate way of securing legal rights—simply quashing dissent on policy issues is a highly concerning, if not a coercive, way of achieving any government objective.

Another example of governments recently limiting religious freedom—and doing so very directly—is found in what is often called Bill 21, after its name during the legislative process, but is now Quebec’s Act Respecting the Laicity of the State. 42Section 6 of this statute is an example of a section of a law that directly limits religious expression:

The persons listed in Schedule II are prohibited from wearing religious symbols in the exercise of their functions.

A religious symbol, within the meaning of this section, is any object, including clothing, a symbol, jewellery, an adornment, an accessory or headwear, that

(1) is worn in connection with a religious conviction or belief; or

(2) is reasonably considered as referring to a religious affiliation. 43

Section 8 extends secularism requirements to those providing and accessing certain government services, by preventing the wearing of clothing that would cover the face, which affects certain forms of religious clothing:

Personnel members of a body must exercise their functions with their face uncovered.

Similarly, persons who present themselves to receive a service from a personnel member of a body must have their face uncovered where doing so is necessary to allow their identity to be verified or for security reasons. Persons who fail to comply with that obligation may not receive the service requested, where applicable. 44

In enacting this law, Quebec’s legislators made use of section 33 of the Charter, commonly known as the “notwithstanding clause,” which permits the federal parliament or a provincial legislature to ensure the operation of a law notwithstanding a number of rights otherwise held under the Charter. This clause was included in the Charter in a manner that in effect allows legislators to decide that certain priorities outside the Charter take priority over rights and freedoms in the Charter, and thus avoid any Charter challenge that would undermine the operation of the law that employs the clause. 45There is a five-year limit on each use of the clause, though it can be renewed, and the main check on its use is through the democratic process, with a corresponding requirement that the clause be used transparently. 46

Quebec’s view is that this law helps to maintain equality between citizens by removing religion from the public arena, corresponding to ideas of secularism or laïcité developed in French political theory. 47As a result, it has chosen to enact the law despite the Charter issues that would otherwise potentially apply. From a legal standpoint, Quebec’s use of the notwithstanding clause will likely protect the law from any effective challenge, although some litigants are nonetheless attempting to find ways of challenging it through the courts. 48

Quebec is legally entitled to make use of the notwithstanding clause in ways that may shield this law from Charter analysis. If it were not, however, this law would be an example of a law imposing limits on religious freedom that are not reasonable. The law impairs religious freedom both directly by prohibiting the wearing of religious symbols by certain officials and indirectly by prohibiting wearing of clothing on the face—somewhat ironically in an era when many jurisdictions are requiring the wearing of face masks—with significant implications for certain well-known forms of religious clothing. The law thus limits believers’ rights to publicly manifest their religion.

In directly specifying limits on religious symbols, the law runs into issues about its very purposes and whether those could meet the first branch of the Oakes test. Beyond that, though, there are general questions about whether the Quebec model’s aim of trying to achieve a system in which government operations are not seen to suggest any religious exclusion of any citizen needs to be accomplished by actually banning the wearing of religious symbols. Nobody thinks that government operations in other Canadian provinces are religiously determined simply because some staff wear religious symbols, typically of the variety of religions present within our pluralistic society. In my view, the law would not meet the reasonable-limits test, meaning that Quebec’s use of the notwithstanding clause is not a peripheral use but a necessary means by which it has maintained the operation of the law.

These examples show some of the ongoing relevance of religious-freedom discussions in Canada. There continue to be examples of governments that limit religious freedom in ways that do not amount to reasonable limits. A better understanding of reasonable limits adds clarity to the legal problems that could be identified in these laws. That clarity can facilitate a more nuanced and less polarized discussion between those who may have differing views on these laws.

Reasonable-Limits Analysis of Pandemic-Related Restrictions

During the COVID-19 pandemic, places of worship were often explicitly named among the facilities subject to restrictions, whether during complete lockdowns or in the imposition of restrictions on the number of people at religious services. There were also instances of restrictions being imposed immediately before major religious holidays, such as Easter. These sorts of restrictions became the subject of controversy, and some litigation resulted. While most religious leaders sought to cooperate with restrictions, or went even further in light of their own religious values of care for one’s neighbour, some places of worship openly defied some of the restrictions. This sort of response led to polarized reactions, having started out in polarized assumptions by some of a broad scope to religious freedom even amid the pandemic. In this part of the paper, I argue that an application of the reasonable-limits analysis makes it possible to draw distinctions between different religious-freedom claims, to understand better some of the claims that do raise issues, and to understand better why some claims and actions on behalf of religious freedom were less appropriate.

One major contextual factor that affects the analysis in these contexts is how places of worship were regulated relative to other facilities with parallels to them, a sort of comparison that also became important in rulings of the United States Supreme Court on similar challenges. 49In a November 2020 decision in Roman Catholic Diocese of Brooklyn v. Cuomo, the United States Supreme Court issued an injunction overturning certain strict limits imposed on worship services in New York that were not imposed on secular businesses. 50In April 2021, the United States Supreme Court applied that decision while granting an injunction to overturn restrictions on religious worship in California and reiterated a concern about government regulations “whenever they treat any comparable secular activity more favorably than religious exercise.” 51 That contextual factor can be part of understanding why a particular pandemic-related limit as imposed on places of worship is or is not reasonable.

The protection of human life and health through appropriate health measures is obviously a sufficiently important objective that could justify some limits on religious services during a pandemic. If places of worship are limited very differently than other public facilities, however, that point would raise questions about whether the limits on places of worship and the resulting limits on religious freedom are actually necessary means of supporting the objective, or whether some different set of restrictions on other public facilities could support the objective in their place.

Decisions to restrict religious activities were no doubt often developed without sufficient knowledge about religious gathering, the constitutional protections it has, and appropriate ways of engaging with faith communities. For example, when the RCMP in May 2020 interrupted Indigenous religious ceremonies at the Beardy’s and Okemasis’ Cree Nation, it was clear to organizers that they did so with no understanding of the ceremonies or with appropriate ways of engaging with them, including their carrying of weapons in places where it was particularly offensive within the traditions at issue. 52However, one can also wonder if there was an implicit devaluation of religion among policy-makers who were quick to label liquor- and cannabis-sales operations essential services but not places of worship.

Some limits on religious gathering may be necessary in a pandemic context, but it is difficult to say with a straight face that those limits are necessary if they apply only to places of worship. Understanding this point in terms of reasonable-limits analysis enables a more nuanced approach to determining which limits are and are not reasonable.

Those limits on public religious gatherings, or their outright banning, in contexts where there were few restrictions on other public facilities at the same time would not have been reasonable limits. Some Canadian court decisions on such bans, such as an outright ban on indoor worship services in British Columbia in early 2021, seemed to overlook attention to the nature of different treatment of religious and secular places and, indeed, to engage in applications of the Oakes test that simply do not match its usual applications. 53Easter 2021 restrictions on church services in Ontario that permitted services but under stricter requirements than at secular facilities have led to litigation still to be adjudicated.

At the same time, it must be acknowledged that places of worship that persisted in holding services in defiance of numbers restrictions, at times when other public facilities were under equivalently severe restrictions, may have called for protections beyond what was appropriate. The reasonable-limits analysis can enable a less polarized discussion.

In litigation in Manitoba, Chief Justice Joyal considered a challenge to restrictions on worship services in place from November 2020 to May 2021. Significant to his decision that these restrictions were reasonable limits was that Manitoba be able to show that restrictions were carefully based on risk calculations combined with vulnerability considerations, and that places of worship were treated at least as well as secular establishments with similar risk profiles for COVID-19 transmission. 54

Places of worship are entitled to a certain degree of priority in the pertinent considerations. Insofar as these places matter to religious freedom, access to them implicates constitutional rights and freedoms that are not at issue in access to every public facility, thus making it constitutionally appropriate that ongoing access to places of worship receive priority over those places not implicating rights and freedoms in any equivalent way. 55That point could matter to a numbers-related limitation in gathering at places of worship. For example, within many Jewish traditions it is necessary to have a certain number of adherents present for a service, the minyan, and any numbers-related limitation that was slightly under this number would have a disproportionately negative impact on religious freedom for a very small change in health protection. Ontario appears to have operated in tension with this principle for a period commencing in late March 2020, when it banned all gatherings of over five persons rather than offering an exemption for the Jewish minyan. That said, much was unknown about COVID-19 at that point in time, and Ontario was responding on an urgent emergency basis. It appropriately ceased the strict restriction as more became known.

Similarly, this point could also highlight the importance of government officials approaching sensitively the matter of restricting gatherings immediately before a major religious holiday. Where genuinely necessary, such a restriction could be a reasonable limit, but if it could be temporarily avoided in the context of places of worship so as to facilitate religious observance while still maintaining strong health protection, then it might be unreasonable to fail to take account of the religious holiday.

At some points during the pandemic, there were restrictions applied against even drive-in church services. 56Understanding the reasonable-limits analysis makes clear why these restrictions were problematic. It is difficult to see how an appropriately conducted drive-in service could lead to any increased risk of COVID transmission. Accordingly, a restriction applied against such a service would limit religious freedom in a way that did not meet minimal-impairment expectations or perhaps even the requirement of a rational connection between objective and rights limit. Such restrictions were not reasonable limits.

Even government policies that purport to accommodate religious freedom warrant careful scrutiny. For example, in New Brunswick’s public-health orders late 2021 and early 2022, a vaccine mandate was imposed on many secular facilities (requiring proof of vaccination or medical exemption for patrons), and then churches and other places of worship were offered the choice of obtaining full proof of vaccination by all in the facility or alternatively operating with masks, social distancing, and a ban on congregational singing, among several other requirements. 57While the order purports to offer a choice to faith communities, it nonetheless imposes certain requirements more strictly on them than on secular facilities in not permitting normal operations if an unvaccinated person with a medical exemption is present, as would be permitted in a secular facility such as a restaurant or nightclub. This again raises questions based on various parts of the Oakes test concerning unequal treatment of religious facilities relative to secular facilities. Moreover, there is also an unnoticed but extreme burden of imposing serious barriers to some individuals’ access to religious sacraments that are sincerely believed to offer vital spiritual nourishment within churches of certain sacramental traditions (Catholics, Orthodox, and some Anglicans and Lutherans). Intervention by outsiders in internal religious matters always raises profound concerns. 58These observations on the implications of the concept of reasonable limits in discussing pandemic-related restrictions on religious services help us to see the value in the concept. It can help with a better discussion of why some restrictions have been unconstitutional and have failed to show adequate respect for religious freedom. At the same time, it can show why some other restrictions have been acceptable and called for more cooperation from some who advocated for religious freedom in excessive ways. The challenges of nuanced discussion in our society are not simple, but concepts like reasonable limits can help.


This paper has reviewed the concept of reasonable limits on rights and freedoms. It began with a review of the Oakes test as the approach of Canadian constitutional law to this concept, engaging with some critiques of the Oakes test but also recognizing its ongoing role as a central means for discussing the concept of reasonable limits. It continued to a discussion of examples of two recent government laws or policies that have infringed on religious freedom in ways that defy the concept of reasonable limits, thus illustrating the potential of the concept in critiquing government laws and policies in ways that can draw together those who regularly advocate for religious freedom and those who can benefit from a nuanced way of understanding religious freedom. Finally, it turned to how the concept of reasonable limits on rights and freedoms can help in understanding what went wrong in the context of some policies during COVID—those that went further than reasonable limits on religious freedom—as well as what was right in some policies that may have received excessive critique.

As a pluralistic society of citizens with many different viewpoints, it is important for Canada to find ways to engage with major questions of social policy in nuanced ways that can overcome polarization. The concept of reasonable limits helps with understanding what limits on religious freedom are acceptable and what limits are unacceptable. It thus has the potential to contribute to that nuanced discussion on issues that have sometimes been over-polarized. This paper ultimately calls for ongoing work on how reasonable limits intersect with religious freedom, something that warrants further attention in additional contexts that could benefit from these sorts of nuanced attempts to overcome excessively polarized discussion.

Two specific recommendations follow. First, while there is a role for advocacy organizations that are pressing for the extension of rights and freedoms on behalf of certain causes, there is also an important role for organizations that seek to inform the public and facilitate nuanced discussions on some of the complex intersections of religious freedom. There should be ongoing work on reasonable-limits issues so as to facilitate better understanding of the concept and dialogue around it that can support robust protections of freedoms while recognizing all interests within society.

Second, in the context of an increasingly secularized society, there needs to be creative outreach on how to avoid imposing unreasonable limits on religious freedom. Given that secularized decision-makers often do not understand the religious-freedom implications of what they are doing, there should be better education, better dialogue, and more involvement in the public square among representatives of faith communities, so that these issues may be better understood. At the same time, around particular policies, there is also a role for direct engagement and consultation with faith communities. In a number of COVID contexts, public officials who talked with faith communities in advance of implementing new policies—after the first days of urgent emergency in March 2020, when such conversation may not have been as feasible—have been able to develop policies with fewer adverse effects on religious freedom, by being better informed about religious freedom and about creative alternative approaches. Pre-infringement engagement with faith communities can play an important role and is an approach to be significantly developed in achieving better compliance with the principles of reasonable limits. 59