This post is the fifth of a series considering three major issues under the Canadian Charter of Rights and Freedoms: the impact of how the Supreme Court of Canada (SCC) has defined rights; the relationship between rights; and the relationship between guarantees of rights and freedoms and section 1 of the Charter. It is the Conclusion to the previous discussions.

I focus the discussion of these issues through the lens of section 3, which guarantees the right to vote and to be eligible to sit in the legislature. Following the exploration of the SCC jurisprudence relating to each of the three issues in relation to section 3, I then consider how the Ontario Superior Court of Justice and the Ontario Court of Appeal applied that jurisprudence in Working Families Ontario v. Ontario (“Working Families I”) and Working Families Coalition (Canada) Inc. v. Ontario (“Working Families II”), both decisions of Morgan J. in the Superior Court, and Working Families Coalition (Canada) Inc. v. Ontario (Attorney General) (“Working Families (ONCA)”), respectively.

Part 1 of the series included the (I) Introduction, (II) a brief summary of the impugned spending provisions of the Election Finances Act (“EFA”) and (III) an overview of the Working Families decisions (see here).

Part 2 began consideration of section 3 of the Charter, specifically the scope of the SCC’s interpretation of section 3 with an Introduction and segment reviewing the jurisprudence (see here). Part 3 considered the relationship between rights, primarily the relationship between section 2(b) and section 3 of the Charter (here). In Part 4, I discussed the impact of the meaning ascribed to section 3’s “right to vote” on the relationship between that section and section 1 (here).

In this Part, Conclusion, I place section 3 in the context of the Charter more broadly by comparing it with some of the other guarantees and bring together the import of the previous discussions.

In addition, in the Working Families Morgan J. and the Ontario Court of Appeal decisions comment on the application of section 33 of the Charter and I complete the series with a rundown of those comments in Part 6.


The wording of Charter guarantees of rights and freedoms and the interpretation the courts, especially the Supreme Court of Canada, give to them mean that there is no simple or single way to interpret them. And there is no simple or single understanding of how section 1 relates to the rights and freedoms. Section 3 captures these intricacies in Charter wording and interpretation, as the earlier posts in this series show.

In this Conclusion, I place section 3 in the context of issues running throughout the Charter by briefly comparing it to several other guarantees. I do not delve into the meaning, interpretation or application of these rights in any detail, which are in many cases extensive and by now can cover a great deal of subject territory. Rather, I simply want to show how section 3 appears to conform or not to how other sections of the Charter are drafted or (briefly) have been interpreted. I group this discussion by characterization, not sections: nature of wording; breadth of scope; process of analysis; changes in interpretation; interchangeability with other sections; and relationship to section 1.

(A very helpful detailed exploration of the individual sections can be found here.)

I begin with the wording and scope. I considered the scope of section 3 in Part 2 of this series (here).

On its face, section 3 seems to protect two things – very important things, but two nonetheless: the acts of being able to cast a ballot and of being able to run for and sit in the House of Commons and the provincial and territorial legislatures. The cases that involve denial of that right illustrate how the Charter works in a straightforward way.

Decisions involving denial of the right, whether to prisoners (the first Sauvé [see Ontario Court of Appeal decision for extended reasons], involving all prisoners, and the second Sauvé, involving prisoners sentenced to two or more years), to Canadians living outside Canada on a temporary basis (Frank) or on the basis of age (Fitzgerald [ABCA]), among other groups, follow the standard process. Those claiming they have been denied the right guaranteed them under section 3 are required to make a prima facie case and if they do so, the analysis shifts to section 1 for the government to defend the infringement.

However, the SCC has also interpreted section 3 to encompass laws that have an effect on the value of the vote for individuals or on the information available to voters to help them make an informed decision. Accordingly, the redistribution of constituencies resulting in inequality of voting power (Saskatchewan Electoral Boundaries), the number of candidates a party must run before it can register as a party (Figueroa) and limits on the release of opinion surveys (Thomson Newspapers) are examples of actions can be challenged under section 3 on the basis that they affect voting, although it may be that they do not contravene section 3 (particular redistribution of boundaries did not; required number of candidates did; the majority determined the time limit on the release of surveys did under section 2(b) and therefore did not consider section 3, while the dissent held provision did not contravene section 3).

None of these laws prevents anyone from voting. This far broader application of section 3 stems from Saskatchewan Electoral Boundaries, in which the SCC equated the right to vote to “effective representation”. Whether actions such as these infringe section 3 depends on whether they interfere with the ability of the voter to be well-informed or to play a “meaningful role in the electoral process”. Importantly, effective representation does not mean equality of voting power or equally populated constituencies, since when other factors (such as geography, community history and interests and minority representation, among others) are taken into account, there may not be voting parity. Diminishing voter parity may be necessary to achieve effective representation.

As I indicated above, however, there are limits to this process of, in effect, “reading in”. The SCC has refused to extend section 3 beyond its literal wording to voting in referenda and municipalities. Section 3 applies only to the House of Commons and to a legislative assembly. Efforts to go beyond those institutions have failed.

Those seeking the right to vote in other contexts, such as band councils (Corbiere), municipalities (Fitzgerald) or referenda (Libman, Haig), among other contexts, must claim the right elsewhere, including section 2(b), freedom of expression (Libman, Haig), or section 15, the right to equality (Corbiere, Fitzgerald).

Despite its narrow wording, the broader approach to section 3 brings its interpretation somewhat closer to some other sections of the Charter of which the interpretation is broad because of the open-ended wording. As I have already discussed in Part 2 of this series, the courts have given the broadly worded (or open-ended) section 2(b), freedom of expression, very broad interpretative scope. Section 2(b) protects almost all forms of expression, that is, any activity or wording that the actor or speaker intends to have meaning. As a result, although the SCC created a three-part test to determine if government legislation or activity has infringed section 2(b), generally the analysis moves quickly from section 2(b) to section 1.

The same is true of section 2(a), freedom of religion, which protects a very broad view of religious beliefs that do not require any kind of “objective” or “expert” proof (Amselem). However, it does not extend to acting on the beliefs (Trinity Western, noting this is an administrative law analysis). Some judges in dissent have also been prepared to carve out certain activities from protection under section 2(a) (Children’s Aid Society of Metropolitan Toronto [Jehovah’s Witness parents refusing child blood transfusion]).

The broad meaning ascribed to section 2(b) advantages the applicant, since it is usually not difficult to establish a prima facie case of an infringement, forcing the inquiry into section 1. The opposite is the case with section 3: here the broad understanding (not just voting, but the amorphous “effective representation”) may disadvantage the applicant who must show that the impugned legislation interferes with effective representation. Depending on the legislation, however, the court may determine that the provisions actually enhance effective representation, not undermine it, in which case the applicant will be unsuccessful (Saskatchewan Electoral Boundaries, Working Families 1 [overturned by the Ontario Court of Appeal in Working Families (ONCA)]).

Unlike some rights or freedoms, such as sections 8, 9 and 11, section 3 does not contain an explicit limitation. Section 8 protects “everyone” from “unreasonable search or seizure” (emphasis added). Section 9 is a guarantee against “arbitrary” detention or imprisonment (emphasis added). Under section 11, everyone charged with an offence is entitled not to be denied “reasonable bail without just cause” (emphasis added). Because of these internal qualifications, it may be difficult, although not impossible, to justify infringements under section 1.

Other guarantees, including sections 7, 15 and 23 also contain requirements that must be addressed within the confines of the right.

Section 7 provides that “[e]veryone has the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice” (emphasis added). The first part of section 7 does not apply if the conduct diminishing the protections there is in accordance with the principles of fundamental justice, whether procedural or substantive. Similarly, section 7 does not protect principles of fundamental justice by itself. The two parts are inextricably linked. The applicant bears the onus under section 7 (that is, the onus does not shift to the government to show that there is a principle of fundamental justice applying to the impugned conduct).

The language provisions are limited in a different way. They include requirements before they apply. For example, section 20 provides that anyone is entitled to services in English or French in a head or central office of an institution of Parliament, but has the same right in any other office of the institution only if “there is a significant demand for communications with and services from that office in such language” or ”due to the nature of the office, it is reasonable that communications with and services from that office be available in both English and French.

Another form of limitation relates to those who may exercise the rights and freedoms. Although nearly all the guarantees refer to “everyone” or “every individual”, section 6, national and interprovincial mobility rights, applies only to “citizens” and, in section 6(2) also to permanent residents. Section 23, guaranteeing minority language educational rights, also applies only to citizens.

Only citizens can benefit from section 3; however, it does not include other types of explicit internal limitation found in the legal rights, for example. Nevertheless, the notion of effective representation may be thought to incorporate a type of internal limitation because the onus lies on the applicant to show the reason supposedly leading to effective representation does not contravene the right to vote.

The Supreme Court of Canada has attempted to mould the analysis under some rights by developing “formal” steps for the analysis. This is the case with sections 2(a), 2(b), 7 and 15, for example. However, this is not the case with section 3.

I considered the relationship between section 3 and other rights in Part 3 of this series. Not surprisingly, since the rights often reflect similar values, there are times when the same set of facts can be considered under more than one section of the Charter. The Superior Court decisions in Working Families illustrates this point particularly well. In some cases, the court will decide not to address a challenge under a particular provision when it has already decided the matter under another section (Thomson Newspapers [impugned provision unconstitutional under section 2(b) and therefore no reason to consider section 3])

I also referred to how section 2(b) and section 15 may in effect substitute for section 3 when the form or locale of the vote does not fall within section 3. Note, however, that other sections, such as sections 2(b) and 15, are subject to the override in section 33 of the Charter. As a result, claimants who are excluded from section 3 by the wording could also be denied the right to vote because of the operation of section 33.

Although one might have expected that the relationship between rights and section 1 would be similar for all rights, this is not the case. I discussed the relationship between section 3 and section 1 in Part 4 of this series.

As I already stated, in some cases, the right is so broad that it is relatively easy to determine that there has been an infringement. In section 2(b) cases, for example, despite the formula the SCC has developed to determine whether the government has infringed freedom of expression, most conduct falls within the protection of the guarantee and therefore the analysis quickly shifts to whether the infringement satisfies section 1. Nevertheless, once the shift is made, it might be harder to protect some forms of expression than others, depending on how close the nexus is with the core of section 2(b). For instance, while commercial expression does attraction section 2(b) protection, an infringement might face a less stringent analysis under section 1.

In all the examples of explicitly limited rights, the internal caveats mean that the applicant must make the case (that the search and seizure is “unreasonable” or the detention is “arbitrary”) before the government must justify their action under section 1. However, once there has been an infringement of section 8, it is unlikely that it can be justified under section 1.

Applying “the purposive approach” to interpretation is consistent with the dominant principle of Charter rights interpretation. It is from this perspective that the SCC effectively changed the wording of section 3. However, the consideration within section 3 of the impact on broader interests, such as whether the lack of spending limits might allow wealth to dominate elections, risks introducing factors more appropriate to the section 1 analysis. And this, in turn, has the potential of shifting the onus of showing a limitation on the right to the applicants rather than the government. Having said that, efforts in dissent to widen the scope of section 3 even further and thus leave little to address under section 1, particularly by LeBel J. in dissent in Figueroa have not been successful.

The overlap of rights and the thread of values permeating not only the rights and freedoms but section 1 means the guarantees do not exist in silos or as singular protections. Thus section 2(b) protects voting, despite section 3’s explicit guarantee of the right to vote. Although there is a specific guarantee of freedom of religion, it is also explicitly protected under section 15 when it occurs as a form of discrimination and it is also a recognized form of expression under section 2(b). Section 7 and section 8 both protect privacy interests.

The SCC has also described section 7 as protecting individual autonomy, the right to protect an individual’s bodily integrity and to some extent privacy interests. Section 15, too, is designed to protect an individual’s right to be treated with respect and dignity, recognizing the historical disadvantage of stereotyping and perpetuation of disadvantage. However, discrimination of a marginalized group can be accepted if a program excluding them is an affirmative action program protecting a larger group or related group (Lovelace [non-band Aboriginal group excluded from fund to benefit First Nations bands constitutional], Kapp [subsections 15(1) and (2) to be read together with result affirmative action under subsection 15(2) does not infringe subsection 15(1)]). And section 23 is not just about language, but about cultural considerations that are part of a person’s identity.

Saskatchewan Electoral Boundaries included an element of the remedying of historical disadvantage under section 3’s concept of effective representation, which may justify (within section 3) a lack of parity in voting power. In Figueroa, Iacobucci J. declared that the “best interpretation of s. 3 is one that advances the values and principles that embody a free and democratic state, including respect for a diversity of beliefs and opinions” (Figueroa, para. 27, cited by Ontario Court of Appeal in Working Families (ONCA), para. 62). And as the Ontario Court of Appeal quotes Frank in Working Families (ONCA), “’[A] broad interpretation of s. 3 enhances the quality of our democracy and strengthens the values on which our free and democratic society is premised’” (Working Families (ONCA), majority, para. 62, citing Frank, para. 27).

Over time, the Supreme Court has included within its protections subject matter effectively fenced off from Charter application through interpretation. A major example is the development of freedom of association under section 2(d) or freedom of expression under section 2(b). In both cases, the SCC has included labour activities among protected activities that were not previously acknowledged (Pepsi-Cola, disapproving Dolphin Delivery re secondary picketing; BC Health Services, overruling the “Labour Trilogy” re collective bargaining). Accordingly, just because an activity has not been recognized as receiving protection does not mean it might not in the future. There is nothing to suggest that the SCC is not open to considering whether legislation or government action other than that already addressed affects the right to vote in the sense that it impairs the ability of voters to become well-informed, for example.

The Supreme Court of Canada has been firm since it first considered section 3 of the Charter that the right to vote means more than simply casting a ballot. Even in Frank, which deals with the right to vote simpliciter, the majority refers to the purpose of section 3 as “effective representation”.

As far as federal, provincial and territorial legislatures are concerned, understanding the right to vote as a right to effective representation means that a variety of statutes or government action in some way related to voting fall within the parameters of section 3. If the statute or action affects the pre-conditions to a well-informed vote, effectively limiting the ability of voters to obtain information, for instance, it can be linked to the objective of effective representation and thus to the right to vote. Limitations of some voters’ rights may actually be “right-enhancing” when viewed through the lens of effective representation and thus not a contravention of section 3.

The SCC has attempted to distinguish when it is appropriate to consider factors related to effective representation under section 3 and when they should be assessed under section 1. In Sauvé, McLachlin C.J. stated that section 3 should not be “limited by countervailing collective concerns”. However, in Saskatchewan Electoral Boundaries she had spoken of a number of factors that were to be considered under section 3 as relating to effective representation (including geography and community interests, cultural and group identity and “enhance[ing] the participation of individuals in the electoral process and society”.

Confusion about where factors or interests should be considered, not always delineated as individual or community on the one hand, and societal on the other, is not an insignificant issue arising from introducing the concept of effective representation into section 3. It can shift the onus from the government to the applicant in a way that seems counter to the framework and purpose of the Charter.

Viewing effective representation as an underlying purpose or objective of section 3 would ease this tension. It would permit applicants to raise concerns about impact on the right to vote that if successful in showing an infringement would have to be justified under section 1 by government. This would be consistent with the primary approach to the Charter that requires the applicant to make a prima facie case with a subsequent shift to section 1.

I return now to the Working Families decisions.

A reminder that Working Families addressed provincial legislation about election spending by third parties, an issue that the SCC had already addressed in the form of federal legislation in Harper, which the applicants challenged under both section 2(b) and section 3. The majority held in Harper that all but one provision did contravene section 2(b) but they were justified under section 1; however, they did not contravene section 3 because they enhanced effective representation.

In Working Families, the applicants initially challenged the legislation under section 2(b) as a contravention of their freedom of expression; however, the government re-enacted the legislation with the inclusion of section 33, prompting the applicants to bring another challenge, this time under section 3.

At the trial level in Working Families, although dealing with the same set of facts, Morgan J. held in Working Families I that the impugned provisions of the Election Finances Act contravened section 2(b) and were not justified under section 1; in Working Families II, however, he held that they did not contravene section 3. The Ontario Court of Appeal was not required to consider section 2(b), although it did make some comments about it; it overturned Morgan J.’s section 3 decision, finding that the provisions did contravene section 3.

In Working Families I, the Attorney General conceded an infringement of section 2(b), although it is clear that Morgan J. would have found a contravention. Therefore, his analysis moved almost immediately to section 1. The government’s objective was to maintain equality in the political discourse or, in other words, to prevent some voices dominating it, and restrictions on advertising or spending by third parties is rationally connected to that objective. However, the 12 months limitation applying before the writ drops does not satisfy the minimal impairment test when testimony had indicated that the 6 months limitation in an earlier version of the legislation would have met the objective. Although other provisions minimize the deleterious effects, combined with the 12 months limitation, the provisions do not pass the last stage of the section 1 test.

When Morgan J. came to the section 3 case, he emphasized that the applicants were particularly concerned about not being able to advertise on television, the most expensive form of advertising. He noted that third parties are able to use all manner of communication that are not expensive to deliver their message and that television advertising is not particularly effective in doing so, especially in promoting policy debate, which was, he explained, the applicants’ goal.

Importantly, he observed that “the threshold for finding a prima facie infringement of section 3 is not as low as it is for section 2(b)” (Working Families I, para. 104). He stated, “The goal of the voting rights analysis in the first instance is not to restrain government or to make its legislative interventions as minimal as possible. Rather, it is to allow government to do what it takes to foster the kind of ‘equality in the political discourse [that] is necessary for meaningful participation’” (Working Families II, para. 106).

Justice Morgan acknowledged that he had held in Working Families I that the 6 months limitation was more minimally intrusive than the 12 months – this was the major reason for finding the provisions unconstitutional. Yet in the section 3 case, he held that since the 12 months had also been said to be effective, he could not say it had not been “carefully tailored” to the goal. Accordingly, he held there was no infringement of section 3. As Morgan J. said,

Unlike under section 2(b), the analysis does not go right to section 1 where the government must meet a test of minimal intrusion. Under section 3, if the government intervenes in the political advertising market it must do so in a way that is attuned to right of voters to meaningful participation via an informed vote. Only if it were found not to be attuned to that objective would the section 1 analysis become relevant and the question of minimal impairment be raised. (Working Families II, para.110)

The Court of Appeal overturned Working Families II.

The majority relied on Frank to explain the broad meaning and purpose of section 3. This is noteworthy since Frank was a clear-cut denial to vote case: “The Supreme Court has confirmed how s. 3 should be read – broadly and in view of the principles that underlie it. ‘[A] broad interpretation of s. 3 enhances the quality of our democracy and strengthens the values on which our free and democratic society is premised’” (Working Families (ONCA), majority, para. 62, citing Frank, para. 27).

After reviewing the SCC jurisprudence, the majority concluded, “It follows that, in a s. 3 analysis, one cannot start from the proposition that any limiting of third party spending implies a breach of the right to vote in s. 3. Some level of spending limits is, on the contrary, necessary to enhance the right to vote.” Even so, one must remember that “there can come a point when a spending limit goes from being voting right-enhancing to being voting right-infringing.” (Working Families (ONCA), majority, paras. 81 and 82).

The majority considered two “proxies” determining “whether a voter’s meaningful participation in the electoral system has been infringed”: whether the restriction is “carefully tailored” (somewhere between allowing third parties to convey information and being overly restrictive) and whether the provisions permit “a modest informational campaign” (they do not have to allow third parties to mount a persuasive campaign capable of determining the outcome).

The majority concluded Morgan J. had erred in not focusing on the 6 months extension but instead on comparing the impact of the 6 and 12 months limitations: “Do they [the increased restrictions] undermine the informational component of the right to vote – because they restrict information in a more severe way than had previously been the case – such that they undermine the right of citizens to meaningfully participate in the political process and be effectively represented?” (Working Families (ONCA), majority, para. 99). It is not clear, however, why Morgan J.’s comparison of the two periods of limitation do not amount in the end to focusing on the 6 months extension, since the evidence is relevant to both approaches.

According to the majority, Morgan J. also erred by not considering the carefully tailored requirement appropriately, failing to consider the six month extension without an increase in the spending limit together. They also stated that Morgan J. confused the analysis between section 3 and section 1 when he stated the 12 months limitation was one of the reasonable alternatives:

These comments reflect a conflation between s. 1 and s. 3. The Supreme Court’s distinction between carefully tailored and perfectly designed, and its reference to reasonable alternatives, are s. 1 concepts, not s. 3 concepts. As we have explained above, although selecting from one of a number of reasonable alternatives may be relevant to whether a Charter breach is minimally impairing, it is not the test for whether an infringement has occurred. (Working Families (ONCA) para. 119)

The majority concluded the provisions were not carefully tailored. In addition, they said that Morgan J. failed to fully consider whether the applicants could mount a modest informational campaign because he did not assess the actual costs. This contrasts with the trial judge in Harper who considered costs of various communication media in detail.

The majority in the Court of Appeal held that the provisions were unconstitutional under section 3 and were not justified under section 1 (the Attorney General did not seek to make the case they were).

As I considered previously in this series, the dissenting judge in the Court of Appeal disagreed with the analysis by the majority, maintaining they applied the wrong test from Harper (not whether the restrictions were carefully tailored but whether the provisions prevented communication of information for well-informed voters), imported justification into section 3 and incorrectly assessed Morgan J.’s decision:

the s. 3 analysis does not require an inquiry into why the government enacted the spending restrictions. Rather, the question is whether the spending limits – as set out in the impugned legislation − restrict information such that they undermine the right to meaningfully participate in the electoral process. In this regard the court is to consider the legislation as it stands, not previous iterations. If the court, on the basis of the evidence, determines that there is no infringement of this right, justification for the legislation is not engaged. (Working Families (ONCA), dissent, para. 169)

Section 3 illustrates some of the challenges in applying the Canadian Charter of Rights and Freedoms. In some ways, it is far from alone in doing so. But much of the difficulty with section 3 comes from the apparent non-congruence between what is a clearly worded and narrow guarantee and the expanded one might say constitutionally recognized right arising from interpretation. This is not the same as activities that might be the subject of challenge under broad wording (deriving from the fundamental freedoms, for example).

On the one hand, this leads to confusion, particularly in the relationship between section 3 and section 1. On the other hand, the decision to rewrite the right to vote as the right to effective representation has opened the door to challenges of legislation or government action that affect the right to vote, not only through an indirect impact but directly. The interests that might in some way affect whether voters are well-informed, a pre-condition for effective representation, appear to be unlimited. This also means that there is a good possibility that factors and restrictions that might otherwise be captured by other sections, particularly section 2(b), that are subject to section 33 might be addressed instead through section 3.

The decisions in Working Families illustrate the complexity in interpreting section 3 and the importance of selecting the appropriate guarantee under which to challenge legislation when more than one is available. Furthermore, the consideration of the same set of facts under section 2(b) and section 3, even though the two sections have different starting points and different tests to determine whether the legislation is unconstitutional, can easily conflate the application of the two provisions to the inquiry, despite admonitions from the Supreme Court of Canada not to do so.


Related Posts