HomeCommentaryA Constitutional Pregnancy: A New Approach to Section 1 Following BC FIPA?

A Constitutional Pregnancy: A New Approach to Section 1 Following BC FIPA?

Can you get a little bit pregnant? Of course not. But recently, in BC FIPA, the Supreme Court of Canada held that it matters whether the government infringes constitutional rights by a little or a lot. The difference will determine the strength of evidence the government will need to adduce to justify a rights infringement under section 1 of the Charter. This link has practical consequences for constitutional litigants.

In this short comment, I will offer an opinion on the connection the Court drew between the severity of a constitutional infringement and the evidence required at the s.1 stage. I will argue that this connection presents a basic problem, namely that there is no method for determining the “severity” of an infringement. In turn, there is no way to ascertain what the level of deference should be under s.1—namely, what the government’s evidentiary standard of justification will be in a given case. Instead of entering this morass, the Court should simply rely on its traditional approach to s.1 justification. In s.2(b) cases, if the restricted speech is closely connected to the purposes underlying the freedom of expression, the government will bear a greater burden in justifying that infringement. Moreover, the s.1 case law has also delineated the categories of circumstances where the government’s evidentiary standard might be relaxed: where the government is seeking to protect vulnerable persons, where the harm is difficult sought to be remedied is difficult to measure, or where the government seeks to address a complex social issue. These categories of deference are far more predictable than the Court’s introduction of a “severity of the infringement” approach, and should be reaffirmed to provide guidance for constitutional litigants.



BC FIPA is a non-profit society based in Vancouver. In 2009 and 2013, it sponsored election advertising in British Columbia. Under BC law, anyone who is a sponsor of “election advertising” is required to register under the British Columbia Election Act. There is no minimum amount of spending under the Act that triggers the registration process. Under s.239 of the Act, anyone who is a “sponsor” must register. BC FIPA was caught by this process: it failed to register. It argued that it was an onerous and unconstitutional burden for those who spend less than $500 on advertising to be compelled to register.[1] In other words, according to BC FIPA, the law impermissibly caught “small spenders” [2]–for example, those who put hand-made signs in their windows, or wear hand-made t-shirts. If the law caught such a class of people, there would be an obvious constitutional concern of overbreadth.

The Court ultimately held, as it almost always does in s.2(b) cases, that the right to freedom of expression was infringed but that the BC law was justified under s.1.[3] How the Court related the minimal nature of the s.2(b) infringement to the government’s standard of proof under s.1 of the Charter is of interest.



The Court began its s.2(b) analysis by pointing out that the sort of speech under analysis—political speech—was obviously at the core of the freedom of expression protected under the Charter. This is  reflected in much of the freedom of expression jurisprudence. [4] Speech that advances political ends is jealously guarded by the Charter. In fact, the Court has previously held this sort of speech will receive greater constitutional protection compared to rather “value-less” speech. In Ross, the Court confirmed that “…when the form of expression allegedly impinged lies further from the “core” values of freedom of expression, a lower standard of justification under s.1 has been applied.”[5] This is all consistent with the purposive approach to constitutional interpretation. That approach determines whether an activity is entitled to constitutional protection based on the strength of its connection to the underlying purposes of the right. Because it concerns election advertising, one would expect the speech at the centre of BC FIPA to receive the utmost protection, in turn influencing the level of deference owed under s.1.

But the s.2(b) analysis did not turn on the fact that the “core” of the freedom of expression was engaged. While the Court found a s.2(b) infringement, it found that the right was only a little bit infringed.[6] Thus, the conclusion turned, not on the nature of the speech, but on the “scope” of the infringement – which necessitated an in-depth interpretation of the impugned legislation. This interpretation led the Court to conclude that the requirement of “sponsorship” essentially saved the law from a gross infringement of s.2(b).[7] The Court noted that a “sponsor,” properly interpreted, only included those who are paid to advertise on behalf of a candidate.[8] Only those who are sponsors must register. Thus, the legislative requirement, while imposing a registration on those seeking to engage in political expression, was not onerous because it did not require small spenders—the t-shirt wearers—to register.

It is important to note that this measurement of a constitutional infringement based on its scope was not defined by the Court. It is unclear whether the Court found a minimal infringement because the law was not overbroad (it did not include the small spenders), or simply because it interfered with protected rights in a minimal fashion (it was not onerous for the for-profit advertisers to register). As we shall see below, this adds to the confusion surrounding what it means for a right to be a little bit infringed.



All parties agreed that a minimal infringement,[9]such as the one found by the Court, would be justified under s.1.[10] In fact, BC FIPA itself did not contend that a registration requirement which only applied to for-profit advertising entities would be unconstitutional.[11] As such, the only s.1 question was whether the registration requirement as applied to these entities was justified.

The traditional approach to s.1 in s.2(b) cases, as noted above, depends on the impugned speech’s connection to the underlying purposes of s.2(b): speech involved in a democratic system, the promotion of truth, and self-realization and individual autonomy.[12] The closer a form of speech is connected to these purposes, the more difficult it will be for a government to justify an infringement of the speech because the government objective will not outweigh the inherent value of the speech.[13] This is simply logical.

The evidentiary standard under s.1 is also influenced by other “contextual factors” in every Charter case.[14] The s.1 justificatory exercise does not operate in the ether—it is directly connected to the social problem sought to be solved by Parliament, which relates to the “type of proof which a court can demand of the legislator to justify its measures under s.1.”[15]For example, where Parliament is dealing with a vulnerable social group, courts will not insist on strict evidence because Parliament is balancing competing social interests.[16] Because of Parliament’s representative function, it is best suited to strike this balance, and a court should not insist on evidence of the efficacy of the balance.[17] Courts will not insist on social science evidence, necessarily, where there is a complex social issue at the centre of the impugned legislation—particularly because in such cases, the social science evidence of a particular harm may be conflicting, and Parliament is best suited to assess the conflict.[18] All of these “categories” of deference are justified because they erect limits on a court’s power of constitutional review and ensure that legislative efforts are not stymied simply because a more efficacious solution might be dreamed up by a court. In these sorts of cases, “logic and reason, assisted by social science evidence is sufficient proof of the harm Parliament seeks to remedy.”[19]

The wrinkle in this case was that the BC government led no social science whatsoever to support the stated goals of the law. But this was not fatal to the government’s justification under s.1. The Chief Justice reasoned:

By not leading social science evidence at this stage, the Attorney General of British Columbia has seriously diminished its ability to justify the infringement of a Charter right, but it has not eliminated it; though logic and reason, without assistance, can only go so far, they can go far enough. Where the scope of the infringement is minimal, minimal deference to the legislature may suffice and social science evidence may not be necessary. That is this case.[20]

The Court’s conclusion is not the problem. Indeed, the BC government’s law seems to engage a complex social issue involving the freedom of speech and the need for a level playing field in terms of elections. The Court’s analysis respecting the evidentiary threshold, however, is another matter. What the analysis envisions is not a lower threshold of justification based on the purpose of the right infringed—in other words, the government is not held to a lower or higher evidentiary standard based on the sort of speech that is implicated by the government action, as held in Ross, or by the other categories of deference typically explored under s.1. Instead, the Chief Justice’s comments raise the possibility that a court will assess, in Charter cases, just how much a right has been infringed before concluding what the government’s evidentiary burden under s.1. Apparently, one can get just a bit pregnant.

As far as I am aware, this is not a statement which has previously been made respecting either the right to freedom of expression or any other constitutional right. And practically, it presents problems for a court hearing a constitutional challenge. Consider the difficulties. How is a court to determine the “level” of infringement of a Charter right? If an infringement is just slightly more than minimal, will a court demand social science evidence? If so, just how much? And again, what is meant by the “scope of an infringement”? These are all unanswered questions.

The Chief Justice’s comments inject uncertainty into the already muddied waters of s.1. The Court should reaffirm its traditional approach under s.1, which is far more predictable and categorical. First, adjusting the level of deference based on the nature of the speech implicated is, as noted above, consistent with the canon of purposive interpretation. For those of us who believe in doctrinal consistency, this is an added benefit. Second, the typical categories undergirding deference under s.1 provide certainty and predictability in the s.1 analysis. The common thread running through the categories of deference explored above is predictability—they do not necessarily require a court to weigh immeasureables. For example, an issue is either complex, warranting deference, or it is not. Given what courts are asked to do under s.1, predictability should be a welcome development. As is well known, s.1 and the Oakes test involve a substantial policy element. They require courts, to a certain degree, to wade into the swamp of politics by evaluating the impugned law’s mechanism of action. The Oakes test goes some way to structuring the discretion of a court faced with a myriad of policy issues under s.1. The categories of deference under s.1 do the same. The categorical analysis is a better way to evaluate whether deference should be owed compared to the nebulous “scope of the infringement” approach introduced in BC FIPA—especially if we value predictability and equality of judicial treatment in the interpretation of constitutional rights.



Chief Justice McLachlin’s one line treatment of the relationship between Charter rights and s.1 seems to be without precedent in the cases. It is possible that the Court did not mean to pronounce a new rule of constitutional law broader than the case before it required; indeed, to do so would be to run contrary to one of the core canons of constitutional interpretation.[21] Nonetheless, the statement is doctrinally inconsistent with the Court’s approach to s.2(b) and s.1, and injects uncertainty into the constitutional analysis. Enterprising constitutional litigants may now readily attempt to characterize a constitutional infringement as “gross” in order to cast a more onerous burden on the government, thereby passing the actual hard, doctrinal work of constitutional analysis. But this is an unprincipled way to conceive of the limitations clause. You cannot get a little bit pregnant. This is a truism which equally applies to constitutional law. The government either infringes rights or it does not. There is no surrogacy period.



[1] BC Freedom of Information and Privacy Association v British Columbia (Attorney General), 2017 SCC 6 at para 12 [BC FIPA].

[2] BC FIPA, supra at para 21.

[3] BC FIPA, supra at para 45.

[4] See the Court’s recount of the freedom of expression jurisprudence at ibid at para 16.

[5] Ross v New Brunswick School District No 15, [1996] 1 SCR 825 at 876-77; The same point was made in R v Keegstra, [1990] 3 SCR 697 at 765.

[6] BC FIPA, supra at para 58.

[7] BC FIPA, supra at para 31.

[8] Ibid.

[9] Notably, we should distinguish between the court’s description of a “minimal infringement” and the “minimal impairment” step of the Oakes test. In the latter case, the Court evaluates whether a right has been minimally impaired; in other words, whether the law interferes with a protected right in the least drastic way possible. At this stage, an infringement of a protected right has already been found by a court. The question is simply whether it was justified. But the minimal infringement analysis introduced by the Court in BC FIPA is different. It evaluates the strength of an infringement at the first step of the constitutional analysis under s.2(b). It then adjusts the level of deference owed to the government under s.1. based on the “strength” of the infringement. Under the traditional Oakes-minimal impairment analysis, deference is not dependent on how minimally the right is impaired.

[10] Ibid at para 45.

[11] Ibid.

[12] Ford v Quebec, [1988] 2 SCR 712 at 765.

[13] Thomson Newspapers Co v Canada (Attorney General), [1998] 1 SCR 877 at para 91 [Thomson Newspapers].

[14] See Thomson Newspapers, supra at paras 87-95.

[15] Thomson Newspapers, supra at para 88.

[16] Irwin Toy v Quebec (Attorney General), [1989] 1 SCR 927 at 994.

[17] Ibid.

[18] Canada (Attorney General) v JTI-Macdonald Corp, 2007 SCC 30 at para 43; RJR-Macdonald v Canada (Attorney General), [1995] 3 SCR 199 at 277-278.

[19] Ibid at para 79. See also R v Butler, [1992] 1 SCR 452 at 503; Keegstra, supra at 768 and 776; RJR, supra at 333.

[20] BC FIPA, supra at para 58 (my emphasis).

[21] Ashwander v Tennessee Valley Authority, 297 US 288 (1936)