Issue #198: November 2025
The new SEAR. Statutory Interpretation Galore; Impact on the Individual; Remedies.
Welcome to the New SEAR: How to Use the Monthly Newsletter
The monthly edition of the newsletter is arranged in themes, rather than case-by-case. To use the newsletter best for your needs, consider scanning the entire document and highlighting themes in a particular month that might interest you. This month, for example, contains three themes. You can then dive deeper into those themes by reading the case analyses I provide.
At the end of each newsletter, I will also include links to recent issues of administrative law in the news.
I will also be providing a year-end “Administrative Law Wrapped.” While the monthly newsletter will come on the last Sunday of each month, the December edition of the newsletter will come on Dec 14 to account for the holiday season.
Statutory Interpretation Galore
The Supreme Court’s recent statutory interpretation jurisprudence centres around a key concept: the text is the “anchor” under the modern approach (see CISSS A, 2024 SCC 43, at para 24). As Wagner CJC explained in that case, the text is the anchor because it demonstrates the precise way in which Parliament wanted to achieve its goals. This is a concept that has now gained some legs in Supreme Court cases (see Kosicki v Toronto (City), 2025 SCC 28 at para 37), and in the lower courts (see Reference re iGaming Ontario, 2025 ONCA 770, below). But not so fast: the Supreme Court’s recent decision in R v Wilson appears to be an ugly duckling.
My bottom line is simple. The Supreme Court can no longer rely on the repetition of Rizzo and Rizzo Shoes and call it a day. The Court should explicitly clarify the proper approach to interpretation of statutes to avoid the whiplash that we see in the movement from cases like CISSS A to Wilson. And it should be ready to simply let Parliament deal with its own mistakes, such as they are. For their part, and as we will see, lower courts are taking the “text as anchor” jurisprudence seriously.
R v Wilson, 2025 SCC 32
Wilson involves the Good Samaritan Drug Overdose Act, which added s.4.1(2) to the Controlled Drugs and Substances Act. The provision reads as follows:
(2) No person who seeks emergency medical or law enforcement assistance because that person, or another person, is suffering from a medical emergency is to be charged or convicted of an offence under subsection 4(1) if the evidence in support of that offence was obtained or discovered as a result of that person having sought assistance or having remained at the scene.
Parliament’s point in enacting this provision was to confer an immunity on those who call in an overdose. The immunity, according to the plain text of the provision, relates to a charge and conviction. The facts of Wilson, however, raised an important interpretive question. Wilson was arrested at the scene of an overdose and searched incident to that arrest.
Does the immunity in s.4.1(2) apply to arrests?
For a majority, Karakatsanis J concluded that it did. She advanced several arguments for her conclusion, but it is notable that she does not once cite the “text as anchor” jurisprudence—none of the cases the SCC has released on statutory interpretation in the last year make even an appearance. Instead, the lynchpin of her opinion is a through-and-through purposive argument. She begins by noting that though the word “arrest” is not present in the provision, “…the words of a statutory provision can never be interpreted in isolation” [34]. She goes on to say, as a result, that permitting arrest would “greatly undermine Parliament’s intent to save lives by promoting resort to emergency services following a drug overdose” [5, 37, 41]. For her, this law is all about “saving lives.” Even though the word arrest is not present, Parliament’s assumed purpose takes a front seat in Karakatsanis J’s analysis.
To be sure, Karakatsanis J takes an eclectic approach, supporting her purposive argument with a plausible view of the text. For her, it would be nonsensical to argue that Parliament did not intend arrest to be covered by the immunity—it is included “by necessary implication” [71]. This is because, in the broader context of the Criminal Code, the power of arrest is (1) limited, precluding the use of arrest to gather other evidence [57-60]; and (2) importantly, is part of a larger scheme “intimately related to bringing persons before justice to answer criminal charges” [62]. In this way, and for her, to speak of “charge” and “conviction” separately from “arrest” simply does not jibe with the rest of the law.
Jamal J dissented. Unlike Karakatsanis J, he spends quite a bit of time emphasizing the “text as anchor” jurisprudence [130, 164, 183]. With that in mind, he starts by noting that the term “arrest” is completely missing from the CDSA provisions, and that the use of specific terms of art—“charged and convicted”—excludes arrest [153]. The context of the provision supports this textual argument for Jamal J. Unlike Karakatsanis J, he viewed arrest as a separate, standalone step in the criminal process; immunity from charge and conviction does not logically necessitate immunity from the investigative step of arrest. Indeed, as he notes, arrest does not require a “prima facie” case for conviction—it only requires reasonable grounds that an offence has been or may be committed in the future.
Finally, Jamal J spends some time reiterating the recent guidance the Supreme Court has offered on the derivation and use of statutory purpose (see para 182). As he notes, legislation does not pursue an objective—such as saving lives—at all costs. As Jamal J notes:
[183] Even an appropriately formulated statement of statutory objectives cannot justify an outright rejection of the text, since the text must remain “the anchor of the interpretive exercise” (CISSS A, at para. 24). The text may “tell an interpreter just how far a legislature wanted to go in achieving some more abstract goal” (para. 24, quoting Mancini, at p. 927).
Jamal J notes that the CDSA amendments were designed not just to save lives, but to protect and balance both public health and public safety. It seems self-evidently true that s.4.1(2) “does not seek to promote public health at all costs”—no law does [188]. Accordingly, preserving the power of arrest would be plausible in pursuit of the public safety purpose. Though evidence seized in the situation described in s.4.1(2) cannot be used to charge or convict, the power of arrest (and search incident to arrest) can be used for purposes of officer/public safety. And he is right to note the ultimate effect of Karakatsanis J’s analysis: it requires the addition of the word “arrest” in the provision which Parliament may, for good reasons, have excluded [196].
Karakatsanis J and Jamal J, analytically, start in different places, end in different places, and take different approaches to get there. While there was a way, in my view, for Karakatsanis J to reach the result she did in this case, the reasoning she used diverges from the Supreme Court’s recent statutory interpretation caselaw. For the sake of sound reasoning, then, Jamal J’s approach is preferable. It does not revolve around the abstract and ultimately slippery purpose of “saving lives”; it gives plausible meaning to the text; and it properly balances Parliament’s competing purposes against each other. And—this is a big deal—it has the virtue of being consistent with the precedents.
Analysis
First, Karakatsanis J’s opinion is lacking because of a failure to at least engage with the Supreme Court’s recent case law. The Supreme Court’s “textual turn” in statutory interpretation is exemplified by cases like CISSS A and Kosicki. Lower courts have taken the ball and run with it (see, for example, this non exhaustive list: Canada v DAC investment Holdings, 2025 FCA 37 at para 17; Toronto Revolver Club v Chief Firearms Officer, 2025 ONSC 4833 at para 12; Animal Justice Canada v Canada (Attorney General), 2025 FC 1703 at para 96; Vancouver (City) v Canada (Attorney General), 2025 FC 1456 at para 105; Canada (National Revenue) v Shopify, 2025 FC 968; R v Allison, 2025 CMAC 3 at para 11). It sends the wrong message for the Court to disappear its own precedents—a frequent problem in administrative law.
Second, consistency with precedent did not have to come at the price of a result that Karakatsanis J, obviously, found inappropriate. A text as anchor approach would not necessarily change the result for Karakatsanis J, though it would bring the reasoning in line with the Supreme Court’s own precedents. As Wagner CJC mentioned in CISSS A (and as noted in the cases below), the level of generality of the text-to-be-interpreted is the quality that courts should ascertain when viewing the text as the anchor (see CISSS A, at paras 30, 33). The level of generality of the text has implications for how purpose can be used in interpretation. Where the text is pitched at a high level of generality (in CISSS A, for example, the terms under interpretation were “correct” and “situation”), purpose and context can be used to specify the meaning of a broad and general term. Where, however, the term is less general (as in Kosicki), purpose and context should only be used as a ”double-check” that the court’s adopted interpretation bears a rational connection to stated statutory purposes.
Karakatsanis J’s opinion could stand while discarding the interpretive approach that she invokes in relation to purpose. Rather than simply asserting that Parliament’s purpose was “to save lives”—and then adopt an interpretation that better achieves that purpose—she could have explained further why arrest is, by necessary implication, included in the immunity. The test of necessary implication is strict, especially where Parliament uses less general and more narrow words as here. It requires, as Jamal J explains, that the omitted aspect—here arrest—is necessarily incidental to the conferred power—here immunity from charge and conviction.
Karakatsanis J’s analysis of the inextricability of arrest and charging might meet it. This argument treats the text as the anchor, and then asks how it promotes Parliament’s stated purposes. But this is not what Karakatsanis J does. Rather, the connection between arrest and charge/conviction is justified by an appeal to “saving lives,” rather than showing how the preservation of arrest does not sufficiently connect to all of Parliament’s stated purposes. This is what I, and some courts, have described as a “purpose error.” It did not have to be this way.
Third, Jamal J’s view of the relationship between text, context, and purpose allows him to render a more plausible interpretation of the provision because he follows the SCC’s own precedent. The heart of his opinion is the simple idea that because Parliament always compromises multiple objectives against each other—here public safety and public health—it supplies a valid reason to preserve arrest as a live power in the teeth of the CDSA immunity. Karakatsanis J’s opinion revolves around what appears to be a common sense intuition: why would Parliament confer an immunity for prosecution but then allow an arrest to unfold? Jamal J’s opinion provides an answer: because preserving the power of arrest would have an important rational connection to public safety, one of the recognized purposes of search incident to arrest [197]. If a purposive approach to interpretation means anything, it cannot maximize one of Parliament’s purposes at the expense of another. Now, one might still maintain that Karakatsanis J’s view better takes account of Parliament’s main purpose. But this begs the question: at the end of the day, we do not know what Parliament intended to do with the arrest power.
Finally, as noted above, I get the impression that Karakatsanis J was driven by the apparent ill-fit of a conclusion that an immunity, like the one in the CDSA, should be limited to arrest. Jamal J offers plausible reasons why this might be so. But even if we do not accept those reasons, and instead accept that the CDSA immunity suffers from an omission or drafting error, it does not mean the Court should step in to correct that error. If Parliament failed to include arrest in the immunity, why should the court do a clean-up job of Parliament’s mess? Our law should encourage democratic responsibility over issues that clearly fall within Parliament’s exclusive jurisdiction. Parliament should suffer the consequences if the court is forced to render an interpretation that might not be ideal. That’s democracy.
The Supreme Court’s recent statutory interpretation cases—up until Wilson—were remarkable for their clarity and consistency, especially on a Court sometimes not known for those qualities. They all emphasized the importance of the statutory text, and the important reasons why the text carries weight: it explains the goods the legislature wanted to achieve, and how it wanted them balanced. That balance is best reflected through the statutory text.
Reference re iGaming Ontario, 2025 ONCA 770
As noted in the discussion of Wilson, the “level of generality” of a law is a central feature grounding the “text-as-anchor” approach. Disputes over the level of generality arose in Reference re iGaming.
In iGaming, linked above, the interpretation centred around s.207(1)(a) of the Criminal Code:
Notwithstanding any of the provisions of this Part relating to gaming and betting, it is lawful for the government of a province, either alone or in conjunction with the government of another province, to conduct and manage a lottery scheme in that province, or in that and the other province, in accordance with any law enacted by the legislature of that province[.]
The question: is Ontario still operating a lottery scheme for the purposes of this section if it permits players in Ontario to play with others in other countries?
The Court majority begins its analysis by noting that there is debate—as there often is in these cases—between a narrow and broad interpretation of s.207(1)(a). This is a question pertaining to the level of generality. On one hand, the phrase in s.207(1)(a)—“in that province”— “is capable of being read narrowly to mean ‘within that province’s geographical boundaries’” [149]. That is one plausible view of the text. However, context—as it often does—helped to shed light on whether this narrow meaning was correct. The Court implicitly concluded that because the level of generality of the text was pitched highly, the context helped to confirm that this meaning was intended by the legislature. For one, the framing of s.207(1)(a) demonstrates that Parliament contemplated interprovincial lotteries: “…it had no objection to provincial government-run lottery schemes extending into other provinces, as long as the scheme does not interfere with the sovereignty of the other provinces” [151]. From this, the Court jumped to its conclusion: as a result, “[i]t can be inferred that Parliament similarly had no objection to provincially-run lottery schemes being linked with other countries…” [151].
The lynchpin of the decision, however, is that because this text is stated at a high level of generality, it permits a dynamic interpretation according to the purpose of the provision [184-185]. The gaming law at issue in the case should, the Court concludes, apply to regulate provincial players who game with players outside Canada. It does so because it “advances public safety by bringing such gaming under protective regulation, thereby reducing such risks as fraud and addiction” [184]. In a close case, consideration of purpose can act as a “tie-breaker” or “double-check,” as here.
There was a dissent, however, filed by van Rensburg JA. One of the points of departure between the dissent and the rest of the Court concerned the precise question that the “text as anchor” jurisprudence invites: a consideration of the level of generality. For her, this was an easy case—the plain meaning of “in that province” means within the territorial boundaries of the province [268]. While s.207(1)(a) does contemplate interprovincial cooperation, it is another thing altogether to contemplate international play [270].
This is a tough case, and there are two comments to consider about it. First, the judges in this case are disagreeing about what really matters if, indeed, the text is the anchor of the interpretive exercise: the level of generality of s.207(1)(a). This is a productive debate because it focuses on the meaning of the text.
Second, and despite this debate, its very existence shows that the “text as anchor” jurisprudence will not resolve all interpretive questions. The promise of a more textually-grounded modern approach is not that it will lead to an easy resolution of all interpretive disputes. It is, instead, justified because it provides structure to the modern approach, imposing a grounded starting point. More discussion and work on the level of generality will help to flesh out the contours of this approach.
R v Allison, 2025 CMAC 3
In this excellent decision, the CMAC rejects an attempt to rely on an old case to corrupt the “text as anchor” methodology:
[10] Mr. Allison invokes this Court’s decision in R. v. Wehmeier, 2014 CMAC 5 [Wehmeier] and submits that as a civilian he should only be tried in a military tribunal in “very exceptional circumstances” and only when “absolutely essential” (appellant’s memorandum, at paras. 1 and 3).
[11] We reject this submission. It amounts to a rewriting of the above provisions of the National Defence Act. They clearly provide that Canadian law can be applied to Mr. Allison. They do not restrict their application to “very exceptional circumstances” or only when “absolutely essential”. The language of statutory provisions, seen in light of their context and purpose, is the anchor in the statutory interpretation process: Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Directrice de la protection de la jeunesse du CISSS A, 2024 SCC 43 , 498 D.L.R. (4th) 316 at para. 24. CISSS A postdates this Court’s decision in Wehmeier and, thus, binds us. There is nothing in the language of the relevant provisions of the National Defence Act or in their text and context that would support limiting their application in this case, as Mr. Allison urges us to do.
Re Vincent Estate; Vincent v Vincent et al., 2025 MBKB 134
In this case, Rempel J articulately explains the relevance of the “level of generality” in the text as anchor approach:
[82] Further guidance as to the modern approach to statutory interpretation is provided by the Supreme Court of Canada in Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Director of Youth Protection of CISSS A, 2024 SCC 43 […]
[83] The object of the Act and the intention of the legislature was clearly to modernize the limitations legislation by removing a plethora of limitation periods that varied depending on the legal principles or doctrines they arose from. Only a select few kinds of claims were carved out of the limitation framework and none of them speak to will challenges. Since the text of the Act is to serve as an interpretive anchor, I am obliged to pay attention to the relative generality of the language of the Act which uses broad and expansive terms. It flies in the face of the modern approach to statutory interpretation to narrowly interpret the purpose of the Act to exclude claims that arose only in probate courts at the time Manitoba joined confederation.
Impact in Reasonableness Review
Kimotho v Canada (Citizenship and Immigration), 2025 FC 1719
Since Mason and Pepa, the Supreme Court and lower courts have emphasized the importance of the “impact to the affected individual” constraint from Vavilov. In Pepa, for example, Martin J faulted the IAD for failing to engage properly with the consequences of its interpretation for Ms. Pepa. In total, the bar that this constraint erects is somewhat high.
This was confirmed in the recent case of Kimotho. At issue: the Refugee Protection Division’s “no credible basis” finding. In refugee law, a “no credible basis” finding is significant: it includes “the elimination of the Applicant’s access to an appeal to the Refugee Appeal Division, the loss of her protection from removal pending that appeal, and the loss of a stay attached to this application for judicial review” [7].
In this case, Battista J concludes that the RPD’s finding did not properly engage with the individual impact constraint. In fact, in this case, he concludes that “…there is no mention at all in the RPD decision of the harsh impact of its no credible basis finding on the Applicant” [42]. This was particularly fatal in this case, because the RPD actually could have decided the case “solely based on the adverse credibility findings it made…” rather than jumping to a “no credible basis” finding [43]. That jump had to be explained in relation to the impact on the individual.
In making this point, Battista J summarizes what the “individual impact” constraint in Vavilov requires [35]:
· The decision maker’s duty of explanation is not discretionary: an explanation “must” be given;
· The decision must reasonably identify the legislative intention leading to the consequences;
· The decision must reasonably identify the consequences of the decision on the affected individual;
· The decision maker is obligated not simply to explain why the decision is consistent with legislative intention, but why the decision best reflects legislative intention. This means that if there is an option that is respectful of other constraints but carries less harsh consequences, a decision maker should explain why that option was not pursued. Applied to determinations pursuant to subsection 107(2), decision makers are required to explain why, in the particular circumstances of the claim and in view of the consequences, it best reflects legislative intention to make a finding of no credible basis rather than simply refuse the claim on the basis of adverse credibility findings.
In my view, Battista J’s remarks clarify the use of this constraint, but raise further questions.
First, it might be that the RPD decision in this case fails on the individual impact constraint because it fails to explain why the “no credibility” finding is a better conclusion than one based on adverse credibility findings. That is a plausible use of the constraint: forcing a decision-maker to explain why the more individually-impactful decision was taken, in comparison to other potential options.
As Battista J says, when the individual impact constraint is triggered, an explanation is not discretionary. But Mason and Pepa might leave the wrong impression: where legislation clearly authorizes severe individual impacts—an expropriation law, for example—it would be duplicative and corrosive to the hierarchy of laws to expect an explanation for clear legislative authorizations. But where the legislature confers a discretion—as in Kimotho—it does make sense to expect heightened justification about potentially extensive impacts.
This is, actually, just simple statutory interpretation—the duty to reason about consequences arises in the same circumstances where courts look at other unwritten principles of law. For example, where there are two rival interpretations—and the statutory term is broad enough to plausibly justify both—the interpretation better connects to statutory purpose should be preferred (see Williams v Canada, 2017 FCA 52). The same is true for consequences in administrative law.
I hesitate to complicate this constraint. Vavilov’s great success is that it does not rely on law office metaphysics and abstract debate about the standard of review. It is qualitative, not quantitative, and it is impossible to reduce judicial review entirely to precise rules. Nonetheless, the clarification here is fundamental, based on the hierarchy of laws.
Remedies and Back-Door Correctness
The signal sent by Mason and Pepa on remedies is intriguing. In both cases, the Supreme Court declared that there was only one reasonable result available in the cases. In Pepa, the majority remitted to the decision-maker—stipulating this one result. In Mason, the Court did not remit and simply decided the matter itself.
For the most part, lower courts have not picked up the invitation from the Supreme Court. And for good reason: both Mason and Pepa, in substance and on remedy, blur the lines between reasonableness and correctness review quite significantly (see Cote J’s opinion in Mason). At first blush, this appears to be yet another example of the Supreme Court’s pre-Vavilov organizing methodology in administrative law: “do as we say, not as we do.” But the consequences are potentially significant: not remitting means, along the way, the court likely conducted something like correctness review rather than reasonableness review. That brings us back to the bad old days of pre-Vavilov, when, as Justice John Evans once said, the organizing theory of the law was disguised correctness review.
But the lower courts have, apparently, started to take notice. Take two lower court decisions rendered this past month. Both of these cases provide good examples of the situations where the judicial review court might properly not remit a decision. Compare these to Mason and Pepa.
Harding v Portugal Cove-St. Philip’s (Town), 2025 NLSC 163
In Harding, the Court concluded that a penalty levied against a municipal councillor did not satisfy the palpable and overriding error standard in the context of a statutory right of appeal. This is a rare finding in itself. But it suggests that the circumstances in which the extraordinary remedial power noted in Mason and Pepa should be limited.
The case involved a rather complex factual history, where the Council first found that a councillor did not have a conflict of interest, and then later determined that he did, in fact, have a conflict. That situation is governed by the Municipal Conduct Act, s.6(8), which says explicitly that in this circumstance (where a conflict was initially cleared but later found), council can invalidate the underlying decision, but “shall not impose any other penalties under this Act against the councillor.” The Council, here, chose to impose a penalty.
This is as close to a black and white error as one can get in the law. It is a palpable and overriding error. And so it does make sense, to use Martin J’s language in Pepa, that:
[125] […] it is natural that this process may incidentally eliminate other options that suffer from the same defects as the one under review, and may even narrow the field to only one possible interpretation. It should not be a surprise that there are cases in which one reasonable interpretation may arise, given that when legislatures speak, they intend to speak with clarity and purpose. This outcome will be more plausible when the question of interpretation is narrow, the statutory language is highly precise, and there are functionally very few options to choose from.
But we should be careful not to take this too far. Harding—involving a clear contradiction of a mandatory statutory rule—is a case where there truly is only one clear result. Martin J said Pepa was such a case. But I am not so sure—especially when, as I explored in Issue #185—there is legitimate debate about the meaning of the provision in that case. Here, there was no such debate, and so it was properly a case where the Court could simply reinstate the councillor.
Le v Card, 2025 BCSC 2204
A similar example arose in this case. There, the petitioner was a landlord who entered into a 1-year residential tenancy agreement with the respondent regarding a suite in a house in Maple Ridge. The landlord’s agent served the tenant a two-month notice to end tenancy for landlord’s use of property. The notice indicated that the reason to terminate the tenancy was that the rental unit will be occupied by the landlord or the landlord’s close family member. The tenant filed a notice of dispute against the landlord arguing that the landlord failed to occupy the property in accordance with s. 49(5) of the Residential Tenancy Act and sought compensation representing 12 times the monthly rent she paid for the property. The arbitrator found in favour of the tenant and ordered the landlord to pay $22,800.
The Court (per Baker J) concludes that this decision is patently unreasonable. At issue was the decision-maker’s totally flawed treatment of evidence. The arbitrator’s decision was premised on the rejection of evidence from the landlord’s agent, but the Court concluded that “the basis for doing so is patently unreasonable. At no time was Mr. Davies’ [the agent] evidence challenged as to its veracity, the source of his knowledge, or any of the other faults expressed by the arbitrator in his reasons” [27].
This is a classic arbitrary evidence error. It rises to the level of an error of law—a clear one. So, one can see the analogy to Harding. Where errors in law or evidence are clear, one can remit. But where you have to dig a bit, or where there is legitimate debate—as there clearly was in both Mason and Pepa—choosing not to remit or declaring one reasonable result is a back-door entry to correctness review.
The Supreme Court frequently calls this “robust reasonableness review.” I do not know what that term means. But whatever it means, reasonableness is not correctness review.
In the News
Universal Ostrich Farms Inc v Canada (Food Inspection Agency) (SCC): The ostrich cull saga has reached its legal conclusion. On November 6, 2025, the Supreme Court of Canada dismissed the application for leave to appeal from the FCA’s decision upholding the CFIA’s culling order. The case serves as a stark reminder that while administrative policies must be reasonable, courts will rarely second-guess technical/scientific determinations made by expert regulators under broad statutory grants of discretion—even when the outcome (depopulation of healthy birds) feels intuitively harsh.
Alberta’s Regulated Professions Neutrality Act: On November 20, the Alberta government announced new legislation that will significantly alter the landscape of professional self-regulation in the province. The Act aims to limit the ability of professional regulators to discipline members for “off-duty” expressive conduct (with exceptions for violence or sexual misconduct) and restricts mandatory training to matters of professional competence. Notably for administrative lawyers, the Act purports to impose a standard of correctness review on decisions involving these new protections—a direct legislative intervention into the standard of review framework in the post-Vavilov world. This, of course, is contemplated by Vavilov.
Disclaimer: Nothing in the SEAR is legal advice. The SEAR is not designed to be comprehensive. It is designed to collect some cases that I believe are interesting to practitioners and students, and to provide my quick thoughts. Not all of the cases presented are fully analyzed ie) I’ve selected the most relevant issues in some of the cases. Please complement the SEAR with your own research and study. Any mistakes are my own.