Dore – Double Aspect

The Metastasis of Charter Vibes…Again

**A version of this appeared in my newsletter, the Sunday Evening Administrative Review**

For the two years or so that I have been doing this newsletter, the story has been a good one. Vavilov settled the law of judicial review to such a considerable extent that I found myself optimistic about the state of things. However, this NWT case—and to a lesser extent, Mason v Canada (Citizenship and Immigration), 2023 SCC 21—cause me worry about the future of Canadian administrative law (see Issue #108). I wrote about this NWT case as representing the “Metastasis of Charter Vibes”—the idea that a “Charter value” can impose a legal duty even where the Charter right does not. This creates a two-track Constitution, where the “true” Constitution of Values applies where the Charter of Rights—apparently only a pale reflection of these values— does not. This is not our constitutional settlement. This theoretical distortion leads to some practical issues, ones that suggest that we may be heading back into a world where the Supreme Court’s administrative law jurisprudence says one thing but does another, with methods “evolving” alongside the composition of the Court. It is unfortunate that this case provides us little justification for the newest “evolution.”

Problems in Principle

Let me start with the problems in principle with this case. The oddity is the particular context of s.23, which is unique in the Canadian constitutional landscape. Not only does s.23 impose positive obligations on the state to provide certain citizens with minority language education, but it is also a right that is collective in scope [1-3]. Despite these features, s.23 is also precisely defined in the text, as the NWTCA pointed out [NWTCA Decision, at para 57]. Section 23 “was carefully crafted to give a narrow bundle of rights to a defined sub-population of Canada” [NWTCA Decision, at para 57].

Not so, for the Supreme Court. Côté J ultimately concludes that, even though this case involves non rights-holders under s.23, there is still a positive obligation on a decision-maker to consider Charter values—this is so “not only where an administrative decision directly infringes Charter rights but also in cases where it simply engages a value underlying one or more Charter rights, without limiting these rights” [64]. Decision-makers must meaningfully grapple with relevant Charter values, reflected in the governing statutory scheme, the parties’ submissions, or “because of the link between the value and the matter under consideration” [66]. More on this in a minute. For now, one need not plead a Charter right, or demonstrate that a right has been infringed according to the typical tests that are associated with each Charter right—in this case, doing so would be no help, since the case involved non-rights holders. Instead, showing that a Charter value is engaged is enough to impose an obligation on decision-makers to consider the value.

What we have, then, is an actionable Charter value that transcends the Charter’s written (and as I will point out, purposively understood) limitations. True, it is not actionable in the sense that it requires a decision-maker to render a decision consistent with the true meaning of the Charter value. Rather, it is a procedural duty, one that imposes a requirement of consideration on the decision-maker.

Even so, this is a significant move, one that was not evident—except in passing remarks—in its previous cases. Saying, as the Court did in Loyola, that the Doré framework attaches to Charter rights and values tells us nothing about the relationship between the two, and certainly does not imply that a Charter value can impose obligations on the state when the Charter right does not. This is especially so when, in the Supreme Court’s previous cases, Charter values and Charter rights seemed one in the same. In Trinity Western, for example, the majority (applying the Doré framework) simply applies the traditional Charter test associated with s.2(a). But in this NWT case, the Charter value imposes an obligation where the Charter right does not.

Nor is it enough to equate this obligation with a purposive approach to constitutional interpretation. Côté J notes that, because “Charter values are inseparable from Charter rights” (a claim that this case actually undermines considerably) “[t]he choice made by the framers to entrench certain rights in the text of the supreme law of Canada means that the purpose of these rights is important for Canadian society as a whole and must be reflected in the decision-making process of the various branches of government” [75]. One can be mistaken for thinking, on this account, that Charter values are just an analogue to a purposive interpretation, especially where—as here—the Court seems to equate the purposes of s.23 with its values, which the Minister was obligated to consider.

But this is not purposive interpretation as we typically understand it. How does one square this case with the Court’s commitment to purposive interpretation as reflected in Quebec (Attorney General) v 9147-0732 Quebec Inc, 2020 SCC 32 and Toronto (City) v Ontario (Attorney General), 2021 SCC 34? Of course, the NWT case is not a case of Charter interpretation per se, but it does raise the prospect of what interests the Constitution prioritizes. In these cases, the Court wisely held that the text of the Charter contains the legal norm that courts must implement; purpose is an aid to interpretation: “A purposive interpretation of Charter rights must begin with, and be rooted in, the text” (City of Toronto, at para 14). The new, hopped-up version of Charter values endorsed in the NWT case implies instead that the value (or the purpose, or the “protection,” or…) is the legal norm that courts must implement, and the text is just a pale reflection of the true Constitution.

This was not written in the stars. Vavilov could have been integrated differently. We could have started with the basic proposition, accepted by Rowe J (who did not sit in the NWT case), McLachlin CJC, and Brown and Côté JJ in Trinity Western: “[w]hen courts review administrative decisions for compliance with the Charter, Charter rights must be the focus of the inquiry—not Charter values” (TWU, at para 166, per Rowe J). Like purposes, this account allows Charter values to play a “supporting role in the adjudication of Charter claims” (TWU, at para 170) but “[a]n analysis based on Charter values should not eclipse or supplant the analysis of whether Charter rights have been infringed” (TWU, at para 175). As I have previously written, using this orthodox starting point still allows us to integrate Vavilov, particularly the focus on more robust reasonableness review, but it does so on the starting point of well-defined Charter rights. Unfortunately, Côté J does not refer to any of the contrary views in Trinity Western, nor the critics of Charter values in the lower courts, nor virtually any counter-arguments of any kind.

That makes this decision all the more difficult to understand. It might be said that the holding in this case is minimalistic, only imposing a duty of consideration. But this does not change the fact that this duty is being imposed where the Charter right imposes no such duty. The coming-apart of “values” and “rights” is a significant move, one for which we are owed a better account.

Practical Problems

I want to point out two practical problems that this decision creates. The first is an ill-defined “relevancy” requirement for the consideration of Charter values and the parties’ arguments. The second is reweighing.

Readers who have looked at the case will note that—maybe—the problems in principle can be remedied by the Court’s apparent integration of Vavilov into the world of Doré . Paul Daly outlines what this framework looks like here, but for our purposes, what is required is the following (1) A demonstration that a decision-maker considered a relevant Charter value [66]; (2) once engaged, a proportionality analysis nourished by Vavilov, in which the decision must show that the decision-maker adequately considered the Charter values with reference to then impact on the individual [68]. Notably, however, there are two important deviations from Vavilov’s reasonableness standard. First, and remarkably, it may be a requirement for decision-makers to consider Charter values even where not argued [66, for example in cases where there is a “link between the value and the matter under consideration].” Second, unlike under Vavilov, courts are entitled to reweigh the weight put on Charter values by decision-makers, which Côté J suggests is “a necessary consequence of the robust analysis required by Doré ” [72].

I cannot help but comment on the oddity of seeing Côté J reinforce the robustness of Doré when in Trinity Western she, along with Brown J, refuted the majority’s same claim with the pithy “[b]ut saying so does not make it so” (TWU, at para 304). And, of course, one might think it is good to see more robust reasonableness review in this context—this is a possibility that I explored, happily, in early work after Vavilov. But as I pointed out above, the way this review has been specified leaves much to be desired.

Consider first the “relevancy” requirement. Côté J calls on Vavilov to outline the three situations in which a Charter value will be relevant, imposing a requirement on a decision-maker. It is to the benefit of enterprising parties—especially under the “link between the value and the matter” branch—to claim broad Charter values (which may or may not be rooted in the purposes of a provision) to impose a requirement on decision-makers where the Charter right otherwise does not apply, under the typical Charter infringement tests. In this case the “link” appeared clear, but it isn’t necessarily so, and I suspect that parties will make much of this. In other words, I do not believe this is something that will be obvious to decision-makers, courts, and parties.

Relatedly, when listing the situations in which a Charter value will be “relevant,” only one of those situations involve cases where the parties’ raised Charter values in their submissions. By implication, this means that there may be an obligation to consider Charter values when a claimant (a) does not have the benefit of the right; and (b) the claimant does not argue a Charter value. This leads to several problems. First, it means that a decision-maker will need to, of her own volition, identify whether a Charter value is present (linked) to the matter, and assign it the appropriate weight, even if not argued. While decision-makers should be held to the letter of the Constitution, it is another thing to assign them a responsibility that is not easily identifiable. Second, this “robustness” seems in tension with other basic requirements of the law of judicial review. The courts have always insisted that, even in administrative contexts, Charter arguments should generally be raised at first instance to permit the development of an adequate record: see Forest Ethics Advocacy Association v Canada (National Energy Board), 2014 FCA 245 at paras 44-45; Mackay v Manitoba, [1989] 2 SCR 357. This case suggests that this longstanding requirement does not apply to discretionary decisions, which leads to another problem: it appears that parties will be able to raise Charter values on judicial review, even if not argued at first instance—or at least that such a practice is now not viewed with inherent suspicion. Considering the difficulties lower courts have already had with this issue, I suspect things will now become more complex (see e.g. McCarthy v Whitefish Lake First Nation #128, 2023 FC 220, Issue #78).

Oddly, this is the exact same scenario the Supreme Court invited in its recent Mason case. As I pointed out, there, an international law issue was not argued at first instance, but the decision-maker’s failure to consider it made the decision unreasonable. Now, we have the same issue with Charter values. The Supreme Court in Vavilov settled on reasonableness as the presumptive standard of review, but in both Mason and this case it eats away at the operation of that standard. Courts shouldn’t conduct de novo review just because the court believes the issue is important enough. I can only quote Côté J at para 172 in Mason: “My colleague views the IAD’s “fail[ure] to address the legal constraints imposed by international law” as unreasonable…With respect, I would have concerns, given the emphasis in Vavilov on a “reasons first” approach, with finding a decision to be “unreasonable” based on arguments that were not put before the administrative decision maker and that do not apply to the individuals actually before that decision maker.”

Finally, consider the problem of reweighing. Courts will be invited to reweigh the weight put on a Charter value where a decision-maker considers it at first instance. This isn’t much in terms of reasonableness review. This creates a new distinction between reasonableness in constitutional cases and reasonableness in all other cases.

Conclusion

What are we left with? Many of us saw Doré as both principally and practically flawed. This case suggests that these flaws are here to stay. But perhaps we are even worse off. We now have a legally-enforceable Charter values framework that applies where the Charter right does not, and where a party does not argue it. If this is a function of s.23’s unique status in the constitutional framework, that is one thing, but I will be surprised if the problems are kept to this world. There are quite likely knock-on practical effects that courts may have to work through. The metastasis continues unabated.

For What It’s Worth

University of Toronto professor Richard Stacey recently released an article in the University of Toronto Law Journal (paywalled, which is truly unfortunate), arguing that (among other things) the Supreme Court of Canada’s decision in Vavilov “affirm[s]” the Supreme Court’s controversial decision in Doré (340; see also 351). To be specific, Stacey says (340-341):

Read together, and building on a rich body of Canadian case law that came before, Vavilov and Doré definitively mark a culture shift way from an outcomes-oriented conception of judicial review toward one that engenders a culture of justification…

…in a culture of justification built on a robust conception of reasonableness, constitutional law and administrative law come together in a unified system of public law.

Stacey also makes a number of other arguments, including: (1) the culture of justification apparently endorsed in the same way in Vavilov and Doré necessarily and logically excludes correctness review (see pg 349) and (2) so-called “Charter values” act as “justificatory resources” that together bind a unified system of public law, bringing Vavilov and Doré together (357 et seq). That is the core of Stacey’s argument: the so-called “unity of public law” thesis draws together a requirement that administrative decisions be justified.

Stacey’s argument is interesting, but ultimately unpersuasive. For one, it treads well-worn territory of “administrative constitutionalism” and “the unity of public law,” theories that—as I will argue—are actually undermined by Vavilov’s formalism. While Vavilov does put a focus on justification, Stacey’s article does not deal with the parts of Vavilov that clearly work against his thesis: the dispatch of expertise as a factor governing the standard of review; and the retention of correctness review based on Rule of Law considerations, among other things. Secondly, Stacey’s article does not engage with key scholarship on this issue post- Vavilov that could both strengthen and undermine his case. Finally, Stacey attaches too much determinacy to Charter values, ethereal things that even their most ardent supporters must agree are relatively indeterminate: perhaps not of this world.

***

Stacey sets out his argument on the first issue (the marriage between Doré and Vavilov) boldly: “The foundation of both cases is the same, and both judgments ultimately embrace the same conception of reasonableness” [351]. Stacey cites the Alsaloussi case out of the Federal Court, where the Court relied on Vavilov in a Doré -type case to give guidance “on what a reasonable decision looks like” [351]. To Stacey, this case—and the theory—reinforces “how the two judgments help to draw administrative and constitutional law together in a single, unified system of law” [352].

As I have outlined previously, I do not see how this is the case, even on the terms of the reasonableness standard. Doré was positively unclear about what administrators should do when faced with a Charter claim, beyond saying that an administrative decision-maker “balances the Charter values with the statutory objectives” (Doré , at para 55 et seq). The only question for a judicial review court is whether “in assessing the impact of the relevant Charter protection and given the nature of the decision and the statutory and factual contexts, the decision reflects a proportionate balancing of the Charter protections at play” (Doré , at para 57). But, unlike the Oakes test, and unlike Vavilov’s list of constraints, the Doré -line of cases do not provide any guidance on how courts should conduct the proportionality analysis. Vavilov provides a far more robust and detailed schema of reasonableness than Doré does, and so to equate these cases on this front is ultimately unpersuasive. The similarity on the reasonableness front—if it exists—is cosmetic at best.

Stacey also does not address why Vavilov’s comments on constitutional issues do not demand a correctness standard in the Doré context. As a reminder, the Court in Vavilov—while expressly excepting Doré from the scope of the comments for now (see para 55)—said that “[t]he constitutional authority to act must have determinate, defined and consistent limits, which necessitates the application of the correctness standard” ( Vavilov, at para 56). As I have said before, this should logically include Charter issues. But Stacey does not address this point, nor does he address important literature attacking administrative constitutionalism as a general theory (see Leonid Sirota’s paper here).

Moreover, Stacey does not address other post- Vavilov commentary that could actually strengthen his point. For example, Paul Daly argues that all issues going to the merits in Vavilov are, on its own terms, subject to the reasonableness standard. I have my issues with this argument, but I think it is far more persuasive in support of Stacey’s argument than the evidence Stacey actually offers–in part because it takes Vavilov on it’s own terms.

Relatedly, Stacey argues that a joint-reading of Vavilov and Doré renders the correctness standard irrelevant (349). Yet this is not convincing to me. As I have argued, and as Professor Daly argues to a similar extent, Vavilov is not just one thing, easily explained with reference to a catchphrase like “culture of justification.” There are various currents of administrative law thought coursing through the decision. Vavilov’s comments on the Rule of Law, for example, are relatively formalistic, focusing on the role of the courts as the guardian of the Constitution. On the other hand, other parts of Vavilov clearly draw from the justificatory school of administrative law thought, championed by scholars like David Dyzenhaus. As I have argued in previous work, these schools of thought can be complementary, but Vavilov is clearly a product of pragmatic agreement, even if guided by principle to some extent. Professor Stacey does not address this reality when he excludes correctness from the standard of review equation, without as much as addressing the counter-arguments clearly presented in Vavilov. Indeed, if one follows Vavilov’s formalistic side, correctness review still has a valuable—and formally required—role in Canadian administrative law.

Finally, I should draw attention to Stacey’s argument on Charter values. Much has been written on Charter values, and I need not reprise that literature to make my objection: Stacey’s focus on Charter values as justificatory resources is only useful if the set of justificatory resources is relatively bounded and determinate. While we cannot expect perfect or near-perfect determinacy in law, and moral reasoning with regards to rights-claims is inevitable, this does not logically entail an embrace of Charter values. The problem is that Charter values are endlessly indeterminate—they are not necessarily bounded by the text of the guarantees they are supposed to represent, and some Charter values could conceivably not be found in the text. Enterprising courts and litigants could pitch a value at high level of generality, leading to needlessly subjective moralizing about rights in a way untethered to the doctrine of various constitutional guarantees. Since there is no clear agreement on (1) how to determine what Charter values are relevant; (2) how Charter values are different than Charter rights; and (3) on how administrators are supposed to understand Charter values as distinct from Charter rights, this set of justificatory resources is not at all helpful to courts or litigants.

Tying together Stacey’s article is a common claim: “…I see no distinction between administrative and constitutional law in the first place” (357). Of course, this is a common (one might say orthodox) position. And yet it ignores an important function of constitutional law in relation to the administrative state—the Constitution (written and unwritten) is a limitation or constraint on government action (see Vavilov, at 56). The hierarchy of laws exists for a reason, and under that hierarchy administrative discretion is parasitic on a statutory grant, which itself is subject to Charter scrutiny. There is no real, formal equality between administrative law and constitutional law: the Constitution is supreme, and it shapes and constrains government power. It does not liberate administrative discretion.

All told, Professor Stacey’s article contributes to the growing post- Vavilov literature. Unfortunately, I do not find it convincing. Nonetheless, the Supreme Court will eventually deal with Doré post- Vavilov. And until then, my opinion is worth as much as the page it’s written on, for what it’s worth.

An Oddity in Strom

In October, the Saskatchewan Court of Appeal released its much-anticipated decision in Strom. Strom raised a number of important issues: “ “at the intersection between professional regulation, Ms. Strom’s private life, and the s.2(b) Charter guarantee of freedom of expression in the age of social media.”

Strom was a registered nurse. Her grandfather tragically passed at a long-term care facility. Strom took to Facebook to criticize the care her grandfather received at the facility. The facility’s employees reported the comments to the Saskatchewan Registered Nurses’ Association (SRNA). The SRNA charged Strom with professional misconduct, and the SRNA Discipline Committee found her guilty.

The Court ultimately overturned the Discipline Committee’s decision. For many reasons that I cannot explore here, I think this is the right decision, in law and in principle. But one aspect of the decision is of particular interest to me: the Court’s standard of review discussion as it related to the freedom of expression arguments raised by Strom.

Of course, in such a case, the framework that governs the standard of review analysis is Doré. Doré holds that the standard of review when a court reviews an administrative decision is reasonableness (Doré, at para 7). Doré also introduces a proportionality framework for assessing whether a decision-maker has struck a reasonable balance between the Charter right at hand and the statutory objective. Post-Vavilov, it is at least an open question about whether Doré is still good law. This is because Vavilov reaffirms that, when litigants challenge a law under the Constitution, the standard of review is correctness (Vavilov, at para 56). I, for one, have questioned why it is that different standards of review should apply, especially since the Court in Vavilov recognized that legislatures cannot “…alter the constitutional limits of executive power by delegating authority to an administrative body” (Vavilov, at para 56).

Enter Strom. In that case, both parties agreed that the standard of review is correctness on the Charter issue (Strom, at para 133). The Court also agreed, but only because the case came to court via a statutory right of appeal, and under Vavilov, statutory rights of appeal invite the appellate standards of review (correctness on questions of law, palpable and overriding error on questions of fact/mixed fact and law) (see Strom, at para 133). The Court noted, though, the following, at para 133:

It is not necessary to consider the question left unanswered by Vavilov, at paragraph 57; that is, what is the standard of review when the issue of whether an administrative decision has unjustifiably limited Charter rights is raised on judicial review, rather than on appeal?

This, in itself, is not really problematic. Given the fact that, for now, Doré lives another day, it would not be appropriate for an appellate court to apply the correctness standard to Doré-type situations. It is defensible—and proper—to simply classify a constitutional issue as a question of law that falls under the scope of an appeal right. And at the end of the day, it does not matter much for a results perspective, because the standard will be correctness either way.

However, in choosing the correctness standard, the Court then did something that is difficult to understand. It said the following, at para 140:

What, then, is an appellate court’s task when reviewing whether the decision of an administrative body unjustifiably infringed a Charter right? In substance, that task is summarily described in Doré at paragraph 6, despite the fact that the standard of review is correctness. The Court’s task is to determine whether the decision-maker disproportionately limited the Charter right or struck an appropriate balance between the Charter right and statutory objectives.

The Court went on to describe the question raised by the case, at para 166:

The question as to whether it has imposed excessive limits is the proportionality question. Here, it is whether the Discipline Committee advanced its statutory objective in a manner that is proportionate to the impact on Ms. Strom’s right to freedom of expression. One aspect of that question is whether the impact on her freedom of speech in her private life was minimal or serious.

The Court then went on to apply the Doré framework, though made reference to some parts of the Oakes test (see para 153).

This strikes me as an oddity. Let’s take what the Court describes its task to be. When courts apply the correctness standard, courts focus on “the conclusion the court itself would have reached in the administrative decision maker’s place” (Vavilov, at para 15). The proportionality analysis advanced in Doré, however, seems to have deference built-in to it. While Doré notes that the Oakes test and the Doré framework “exercise the same justificatory muscles,” (Doré, at para 5), Doré deference asks courts to give some weight to the statutory objective being advanced by the decision-maker. This was, indeed, a bone of contention for the dissent of Brown and Côté JJ in Trinity Western. But when a court conducts correctness review, at least in theory, the court should not give any weight to what the decision-maker’s reasons are for making

That said, there are no perfect Platonic forms in law. It is true that Oakes itself has developed to contemplate deference in its application. And it is also true that courts, post-Vavilov, have applied what I call “light correctness review” (see Planet Energy, at para 31), where a certain amount of weight is given to the decision-maker’s “…interpretation respecting the words of the Act, the general scheme of the Act and the policy objectives behind the provision.” This could be seen as a sort of S kidmore-like deference, under which courts give non-binding weight to an administrative interpretation.

That said, the inherently deferential idea of Doré review seems inconsistent with a stringent application of the correctness standard. This, to me, is a theoretical oddity, even if its effect is blunted on the edges.