Mason v. Canada (Citizenship and Immigration)

These appeals require the Court to apply the framework for judicial review developed in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, to two administrative decisions involving a question of statutory interpretation in the immigration context.

The statutory provision at issue, s. 34(1) (e) of the Immigration and Refugee Protection Act , S.C. 2001, c. 27 (“IRPA ”), provides that permanent residents and foreign nationals are inadmissible to Canada on “security grounds” for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada”. The key point of disagreement among the administrative decision makers and courts below is whether the “acts of violence” listed as “security grounds” in s. 34(1)(e) require a link to national security or the security of Canada, or whether s. 34(1)(e) applies to acts of violence more broadly even without such a link.

Both administrative decisions under review interpreted s. 34(1)(e) as not requiring the acts of violence to have a link to national security or the security of Canada. In the first administrative decision, the Immigration Appeal Division (“IAD”) of the Immigration and Refugee Board of Canada (“IRB”) ruled that Mr. Earl Mason, a foreign national, could be found inadmissible under s. 34(1)(e) if his alleged violent conduct were established. Mr. Mason allegedly shot a gun and wounded two people when he was assaulted during a fight at a bar. Charges against him were stayed and he was not convicted of any criminal offence. In the second administrative decision, the Immigration Division (“ID”) of the IRB followed the IAD’s interpretation of s. 34(1)(e) in Mr. Mason’s case and ruled that Mr. Seifeslam Dleiow, a foreign national, was inadmissible under s. 34(1)(e) for acts of violence against two intimate partners. It was not alleged that either Mr. Mason or Mr. Dleiow engaged in acts of violence with a link to national security or the security of Canada.

The Federal Court allowed Mr. Mason and Mr. Dleiow’s applications for judicial review. In Mr. Mason’s case, in reasons released before this Court released Vavilov, the Federal Court ruled that it was unreasonable to interpret s. 34(1)(e) as applying to acts of violence without a nexus to national security. The Federal Court followed that approach in Mr. Dleiow’s case. Thus, neither Mr. Mason nor Mr. Dleiow was inadmissible. In both cases, the Federal Court also certified serious questions of general importance, so that the Federal Court of Appeal could consider whether it was reasonable to interpret s. 34(1)(e) as not requiring proof of conduct having a nexus to national security or the security of Canada.

The Federal Court of Appeal allowed both appeals. In reasons addressing both cases — released after this Court released Vavilov — the Court of Appeal ruled that the IAD and ID had reasonably interpreted s. 34(1) (e) of the IRPA as not requiring a nexus to national security or the security of Canada.

Mr. Mason and Mr. Dleiow now appeal to this Court. Two issues arise. First, what standard of review should the reviewing courts have applied when reviewing the decisions of the IAD in Mr. Mason’s case and the ID in Mr. Dleiow’s case? Second, how should that standard of review have been applied in the circumstances?

In Vavilov, this Court revised the framework for determining the standard of review. The Court established a presumption that the standard of review of the merits of an administrative decision is reasonableness, subject to limited exceptions based on legislative intent or when required by the rule of law (paras. 10 and 17). The revised framework seeks to maintain the rule of law, while respecting a legislature’s intent to entrust certain decisions to administrative decision makers rather than courts (paras. 2 and 14). It also aims to bring simplicity, coherence, and predictability to the law on the standard of review and to eliminate the unwieldy exercise of determining the standard of review based on contextual factors, as had been required by this Court’s jurisprudence following Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190 (Vavilov, at paras. 7 and 10).

Vavilov also explained how a court should conduct reasonableness review. This Court stressed that reasonableness review and correctness review are methodologically distinct (para. 12). Reasonableness review starts from a posture of judicial restraint and focusses on “the decision the administrative decision maker actually made, including the justification offered for it, and not on the conclusion the court itself would have reached in the administrative decision maker’s place” (paras. 15 and 24). When an administrative decision maker is required to provide reasons for its decision, reasonableness review requires a “sensitive and respectful, but robust” evaluation of the reasons provided (para. 12). A reviewing court must take a “reasons first” approach that evaluates the administrative decision maker’s justification for its decision (para. 84). An administrative decision will be reasonable if it “is based on an internally coherent and rational chain of analysis and . . . is justified in relation to the facts and law that constrain the decision maker” (para. 85). This Court also affirmed “the need to develop and strengthen a culture of justification in administrative decision making” (para. 2).

Applying the Vavilov framework to these appeals, I conclude that the standard of review of the administrative decisions at issue is reasonableness. No established exception to the presumption of reasonableness review applies, nor should any new exception be created on the basis that the appeals involved a serious question of general importance certified for appeal to the Federal Court of Appeal under s. 74 (d) of the IRPA . The certified question regime is a statutory mechanism for the Federal Court to provide for an appeal from a judicial review decision in certain circumstances.

Both administrative decisions were unreasonable. In particular, the IAD in Mr. Mason’s case, whose interpretation of s. 34(1)(e) was followed in Mr. Dleiow’s case, failed to consider three significant legal constraints bearing on its decision. First, the IAD failed to address critical points of statutory context that Mr. Mason had raised in his submissions to the IAD. Second, the IAD failed to address the potentially broad consequences of its interpretation, which again Mr. Mason had raised in his submissions. These omissions involved significant failures of “responsive justification” that would cause a reviewing court to lose confidence in the IAD’s decision. Third, the IAD failed to interpret and apply s. 34(1)(e) in compliance with international human rights instruments to which Canada is a signatory — specifically, the obligation of non-refoulement under Article 33(1) of the 1951 Convention Relating to the Status of Refugees, Can. T.S. 1969 No. 6 (“Refugee Convention”) — contrary to the express direction in s. 3(3) (f) of the IRPA that it must do so. The IAD’s failure to consider these three legal constraints rendered its decision unreasonable.

In these cases, the relevant legal constraints point overwhelmingly to a single reasonable interpretation of s. 34(1)(e) — a person can be found inadmissible under s. 34(1)(e) only if they engage in acts of violence with a nexus to national security or the security of Canada.

As a result, I would allow both appeals, set aside the judgments of the Federal Court of Appeal, allow the applications for judicial review, and quash the administrative decisions.

Mr. Earl Mason is a citizen of Saint Lucia and a “foreign national” in Canada — that is, he is neither a Canadian citizen nor a permanent resident (IRPA , s. 2(1) , “foreign national”). He is married to a Canadian citizen and has two daughters in Canada. He entered Canada in June 2010 and has remained here as a foreign national. He claimed refugee protection when he arrived in Canada, but he later withdrew this claim when he applied for permanent residence with his wife’s sponsorship.

The following allegations were made against Mr. Mason in the inadmissibility proceedings under review. In May 2012, Mr. Mason had an argument with a man at a concert in a bar in Surrey, British Columbia. The man broke a beer bottle over Mr. Mason’s head, and Mr. Mason responded by drawing a gun from his waistband and firing it eight times, wounding his assailant and another man. In May 2014, Mr. Mason was charged with two counts of attempted murder and two counts of discharging a firearm with intent to wound or disfigure. The shooting was not linked to terrorism or organized crime. In 2015, the charges were stayed because of delay. As a result, Mr. Mason was not found guilty of any criminal offence.

The IRPA sets out several grounds on which a permanent resident or foreign national may be found inadmissible, and thus may be denied entry to or be required to leave Canada (ss. 34 to 42). In April 2016, a Canada Border Services Agency (“CBSA”) officer prepared a report alleging that Mr. Mason was inadmissible on “security grounds” under s. 34(1) (e) of the IRPA , which provides that a permanent resident or foreign national is inadmissible for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada”. Section 34(1) states:

Security

34 (1) A permanent resident or a foreign national is inadmissible on security grounds for

(a) engaging in an act of espionage that is against Canada or that is contrary to Canada’s interests;

(b) engaging in or instigating the subversion by force of any government;

(b.1) engaging in an act of subversion against a democratic government, institution or process as they are understood in Canada;

(c) engaging in terrorism;

(d) being a danger to the security of Canada;

(e) engaging in acts of violence that would or might endanger the lives or safety of persons in Canada; or

(f) being a member of an organization that there are reasonable grounds to believe engages, has engaged or will engage in acts referred to in paragraph (a), (b), (b.1) or (c).

In May 2016, a delegate of the Minister of Public Safety and Emergency Preparedness referred the CBSA’s inadmissibility report to the ID for an admissibility hearing.

Mr. Seifeslam Dleiow is a citizen of Libya and a foreign national in Canada. He entered Canada in June 2012 on a study permit, which expired in 2014. In 2015, he made a claim for refugee protection. In October 2017, the Refugee Protection Division of the IRB refused his claim, and in October 2018, the Refugee Appeal Division dismissed his appeal.

In September 2018, a CBSA officer prepared a report alleging that Mr. Dleiow was inadmissible on security grounds under s. 34(1)(e). The report alleged that since arriving in Canada, Mr. Dleiow had engaged in acts of violence against intimate partners and other persons. Criminal charges flowing from these incidents were stayed, except for 3 charges: being unlawfully in a dwelling house with intent to commit an indictable offence, mischief under $5,000, and uttering threats to cause death or bodily harm. Mr. Dleiow pleaded guilty to these charges and received a conditional discharge. A delegate of the Minister of Public Safety and Emergency Preparedness then referred the CBSA’s inadmissibility report to the ID for an admissibility hearing.

The ID addressed a preliminary question of law as to whether Mr. Mason’s alleged conduct, if proven, could be a ground of inadmissibility under s. 34(1)(e). The ID ruled that a “security groun[d]” under s. 34(1) means a threat to the security of Canada or another country, and that the act of violence in question must have some connection to a threat to the security of Canada. In the ID’s view, Mr. Mason’s alleged conduct involved “mere criminal offences”, which “although very serious”, lacked “any element that would elevate them to security grounds”, and thus s. 34(1)(e) could not apply (para. 24).

The Minister of Public Safety and Emergency Preparedness appealed the ID’s decision in Mr. Mason’s case to the IAD, which allowed the Minister’s appeal, set aside the ID’s decision, and referred the matter back for a full hearing on the merits. The IAD concluded that inadmissibility under s. 34(1)(e) does not require a link to national security or the security of Canada. In the IAD’s view, “security” under s. 34(1)(e) relates to “security in a broader sense”, namely, to ensure “that individual Canadians are secure from acts of violence that would or might endanger their lives or safety” (para. 37).

In Mr. Dleiow’s case, the ID saw no basis to depart from the IAD’s interpretation of s. 34(1)(e) in Mr. Mason’s case, and therefore affirmed that s. 34(1)(e) does not require a link to national security or the security of Canada. The ID also heard evidence and concluded that Mr. Dleiow was inadmissible because he had engaged in violent acts against two intimate partners, which there were reasonable grounds to believe had endangered their safety. The ID based this conclusion on a guilty plea for acts relating to one intimate partner, and on testimony and police occurrence reports relating to the other intimate partner. As a result, the ID ruled that Mr. Dleiow was inadmissible and issued a deportation order.

The Federal Court granted Mr. Mason’s application for judicial review of the IAD’s decision. The court held that the IAD’s interpretation of s. 34(1) (e) of the IRPA was unreasonable because it disregarded the structure of the Act and rendered meaningless statutory provisions for inadmissibility based on criminality. In the Federal Court’s view, s. 34(1)(e) requires a link to national security.

The Federal Court — whose reasons were released before this Court released Vavilov — outlined how a reviewing court should evaluate the reasonableness of an administrative decision maker’s interpretation of a statute. In the Federal Court’s view, a reviewing court must ensure that an administrative decision maker did not overlook a very strong argument — a “knock-out punch”, that is, an interpretation that is internally consistent, withstands scrutiny, and is not met by a countervailing interpretation of similar force — or choose an interpretation when the interpretive “clues” point overwhelmingly in the other direction.

The Federal Court ruled that the IAD’s interpretation of s. 34(1)(e) was unreasonable because it conflicted with the broader structure of the IRPA , thus undermining Parliament’s intent. In the court’s view, this structural argument was a “knock-out punch”. The IAD’s decision upset the carefully crafted structure of the IRPA by including under s. 34(1)(e) a vast range of conduct that “would or might endanger the lives or safety of persons in Canada”. This would thwart Parliament’s intent by bringing under the most serious category of inadmissibility conduct falling below the thresholds for less serious categories of inadmissibility, and it would discard Parliament’s choice under s. 36 of the IRPA to require a conviction when criminal conduct was committed in Canada. Section 36(1) and (2) of the IRPA provide for inadmissibility based on “serious criminality” and “criminality” as follows:

Serious criminality

36 (1) A permanent resident or a foreign national is inadmissible on grounds of serious criminality for

(a) having been convicted in Canada of an offence under an Act of Parliament punishable by a maximum term of imprisonment of at least 10 years, or of an offence under an Act of Parliament for which a term of imprisonment of more than six months has been imposed;

(b) having been convicted of an offence outside Canada that, if committed in Canada, would constitute an offence under an Act of Parliament punishable by a maximum term of imprisonment of at least 10 years; or

(c) committing an act outside Canada that is an offence in the place where it was committed and that, if committed in Canada, would constitute an offence under an Act of Parliament punishable by a maximum term of imprisonment of at least 10 years.

Criminality

(2) A foreign national is inadmissible on grounds of criminality for

(a) having been convicted in Canada of an offence under an Act of Parliament punishable by way of indictment, or of two offences under any Act of Parliament not arising out of a single occurrence;

(b) having been convicted outside Canada of an offence that, if committed in Canada, would constitute an indictable offence under an Act of Parliament, or of two offences not arising out of a single occurrence that, if committed in Canada, would constitute offences under an Act of Parliament;

(c) committing an act outside Canada that is an offence in the place where it was committed and that, if committed in Canada, would constitute an indictable offence under an Act of Parliament; or

(d) committing, on entering Canada, an offence under an Act of Parliament prescribed by regulations.

The Federal Court found all the countervailing points of interpretive context unpersuasive and ruled that only one reasonable interpretation was available. It therefore quashed the IAD’s decision and restored the ID’s decision.

The Federal Court certified the following serious question of general importance for appeal to the Federal Court of Appeal: Is it reasonable to interpret s. 34(1) (e) of the IRPA in a manner that does not require proof of conduct that has a nexus with “national security” or “the security of Canada”?

The Federal Court allowed Mr. Dleiow’s application for judicial review of the ID’s decision. The court applied the Federal Court’s reasoning in Mr. Mason’s case for reasons of comity, set aside the ID’s decision and ordered the matter be reconsidered on the merits by a different decision maker. The court also certified the same serious question of general importance.

The Federal Court of Appeal decided the appeals in Mr. Mason’s and Mr. Dleiow’s cases together, after this Court had released Vavilov. The court held that the administrative decisions reasonably interpreted s. 34(1)(e) as not requiring a nexus with national security or the security of Canada.

The Court of Appeal began by discussing how a court should conduct reasonableness review. The court said that “Vavilov tells us much but it leaves some things unclear” (para. 9). The court cautioned that a reviewing court should not fashion its own yardstick and use it to measure what the administrator did, but should instead conduct “a preliminary analysis of the text, context and purpose of the legislation just to understand the lay of the land before they examine the administrators’ reasons” (para. 17). The Court of Appeal also criticized the Federal Court’s “knock-out punch” approach in Mr. Mason’s case as involving disguised correctness review.

The Court of Appeal concluded that the IAD was alive to the essential elements of s. 34(1)(e)’s text, context, and purpose, and saw no omitted aspects that would cause a loss of confidence in the outcome. It rejected Mr. Mason’s argument that reading s. 34(1)(e) without a nexus to national security is inconsistent with the broader statutory context. The court ruled that the IAD reasonably concluded that the conduct captured by s. 34(1)(e), which speaks of the danger posed to the “lives or safety” of persons in Canada, is only a small subset of what would be considered serious criminality under s. 36 of the IRPA . Sections 34 and 36 address two different matters — conduct and convictions, respectively (para. 55). Section 36 is much broader, and applies to much non-violent criminal behaviour; s. 34(1)(e) is narrower, and applies only to acts of violence. The court noted that when the IAD said that s. 34(1)(e) is not absurdly broad because the conduct captured by the provision is “narrowly defined”, this could only mean that it interpreted “safety” in s. 34(1)(e) as “something approaching the level of a threat to life, not just minor harm” (para. 57). The court stated that although the IAD did not address certain contextual arguments, this failure was not a fundamental gap and did not cause the court to lose confidence in its outcome so as to make its decision unreasonable.

In contending that s. 34(1)(e) requires a nexus with national security or the security of Canada, Mr. Mason also invoked the Refugee Convention and the 1967 Protocol Relating to the Status of Refugees, Can. T.S. 1969 No. 29 (“Refugee Protocol”). The Court of Appeal declined to entertain this argument because it viewed this as a new issue that should have been raised before the IAD, and because “certain background documents and other instruments needed to understand any international obligations” were not in evidence (para. 74).

The Court of Appeal concluded that some elements of s. 34(1)(e)’s text, context, and purpose favoured the need for a nexus with national security or the security of Canada, while others did not. It stated that “the issue of legislative interpretation is best described as one where the issue is open to some debate” (para. 76).

In the result, the Court of Appeal ruled that it is reasonable to interpret s. 34(1) (e) of the IRPA in a manner that does not require proof of conduct with a nexus to “national security” or “the security of Canada”. The court therefore allowed the appeals, set aside the judgments of the Federal Court, and dismissed the applications for judicial review.

These appeals raise two issues: (1) What is the appropriate standard of review of the administrative decisions at issue? (2) Was that standard of review applied properly in these cases?

In what follows, I first address the applicable standard of review and explain why it is reasonableness. I then summarize Vavilov’s guidance on conducting reasonableness review and apply that guidance to these cases. I conclude that the administrative decision makers’ reasons involved failures of justification that made their decisions unreasonable.

The first issue concerns the appropriate standard of review. As this Court has noted, “[a] reviewing judge’s selection and application of the standard of review is reviewable for correctness” (Northern Regional Health Authority v. Horrocks, 2021 SCC 42, at para. 10, citing Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, at paras. 45‑47). An appellate court must “ste[p] into the shoes” of the lower court and focus on the administrative decision under review (Agraira, at para. 46, citing Merck Frosst Canada Ltd. v. Canada (Health), 2012 SCC 3, [2012] 1 S.C.R. 23, at para. 247, per Deschamps J., dissenting). On this approach, an appellate court “accords no deference to the reviewing judge’s application of the standard of review”, but rather “performs a de novo review of the administrative decision” (Horrocks, at para. 10, citing D. J. M. Brown, with the assistance of D. Fairlie, Civil Appeals (loose-leaf), at § 14:45). None of these principles was changed by this Court’s decision in Vavilov. As a result, the issue is whether the Federal Court selected the correct standard of review and applied it properly (Agraira, at para. 47).

Before this Court, the appellant Mr. Mason and the intervener the Canadian Association of Refugee Lawyers assert that the standard of review is correctness. Citing pre-Vavilov jurisprudence, they say that because the appeal to the Federal Court of Appeal involved a “serious question of general importance” certified under s. 74 (d) of the IRPA , it would be “incoherent” if the standard were not correctness (A.F., at para. 49, citing Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982, at para. 43). The certified question regime under s. 74 (d) of the IRPA provides a mechanism for the Federal Court to provide for a statutory appeal of a judicial review decision that raises a serious question that is dispositive of the appeal, transcends the interests of the parties, and raises an issue of broad significance or general importance (Lunyamila v. Canada (Public Safety and Emergency Preparedness), 2018 FCA 22, [2018] 3 F.C.R. 674, at para. 46; Canada (Immigration and Citizenship) v. Laing, 2021 FCA 194, at para. 11 (CanLII); see also Canada (Public Safety and Emergency Preparedness) v. XY, 2022 FCA 113, 89 Imm. L.R. (4th) 173, at para. 7). The appellant Mr. Mason and the intervener the Canadian Association of Refugee Lawyers say that the certified question regime “weighs in favour of a correctness standard” and “evinces a particular concern that questions of general importance be appropriately resolved” (A.F., at para. 49, citing Chieu v. Canada (Minister of Citizenship and Immigration), 2002 SCC 3, [2002] 1 S.C.R. 84, at para. 23; see also I.F., at paras. 4‑10).

I respectfully disagree with this submission. In my view, the standard of review of the administrative decisions is reasonableness. This conclusion rests on three propositions: (1) Vavilov established that the standard of review of the merits of an administrative decision is presumptively reasonableness; (2) no established exception to the presumption of review for reasonableness applies in these cases; and (3) the certified question regime under s. 74 (d) of the IRPA does not displace the presumption of reasonableness review and warrant recognition of a new category of correctness. I will address each point in turn.

In Vavilov, this Court established a presumption that when a court reviews the merits of an administrative decision, the standard of review is reasonableness (para. 16; Canada Post Corp. v. Canadian Union of Postal Workers, 2019 SCC 67, [2019] 4 S.C.R. 900, at para. 27). This presumption is rebutted in two types of situations, which together provide six categories of correctness review (Vavilov, at paras. 17 and 69; Society of Composers, Authors and Music Publishers of Canada v. Entertainment Software Association, 2022 SCC 30, at para. 40). The first type of situation is where the legislature has indicated that it intends a different standard or set of standards to apply, and the second type of situation is where the rule of law requires that the standard of correctness be applied (Vavilov, at para. 17). Vavilov thus sets out a “general rule” of reasonableness review, “subject to limited exceptions” (D. Mullan, “Reasonableness Review Post-Vavilov: An ‘Encomium for Correctness’ or Deference As Usual?” (2021), 23 C.L.E.L.J. 189, at p. 200).

The first situation, based on legislative intent, provides for two categories of correctness review: when the legislature explicitly prescribes the standard of review, and when it provides for an appeal from an administrative decision to a court, thus signalling that appellate standards of review apply (Vavilov, at para. 17; Canada Post, at para. 27).

As for the second situation, Vavilov set out three categories of questions that the rule of law requires to be reviewed on a standard of correctness: constitutional questions, general questions of law of central importance to the legal system as a whole, and questions related to the jurisdictional boundaries between two or more administrative bodies (para. 17; Canada Post, at para. 27).

At the time it was rendered, Vavilov thus recognized five categories of correctness review: (1) legislated standards of review; (2) statutory appeal mechanisms; (3) constitutional questions; (4) general questions of law of central importance to the legal system as a whole; and (5) questions related to the jurisdictional boundaries between two or more administrative bodies (paras. 17 and 69).

At the same time, Vavilov did not definitively foreclose the possibility of recognizing new categories of correctness “[i]n rare and exceptional circumstances . . . when applying reasonableness would undermine legislative intent or the rule of law in a manner analogous to the five correctness categories” already identified (Society of Composers, at para. 27; Vavilov, at para. 70). This Court recently recognized a sixth category of correctness review in Society of Composers, a case involving copyright royalties for works accessed online: (6) “when courts and administrative bodies have concurrent first instance jurisdiction over a legal issue in a statute” (para. 28). The Court noted that both the Copyright Board and the courts have concurrent first instance jurisdiction regarding the interpretation of the Copyright Act , R.S.C. 1985, c. C-42 , which signals “a legislative intent for judicial involvement” (para. 31) and highlights the need for consistent and definitive interpretation of the Copyright Act to maintain the rule of law (paras. 33‑35).

The presumption of reasonableness review and the limited circumstances in which it is rebutted provide a comprehensive framework for determining the standard of review. This framework brings simplicity, coherence, and predictability to the law on the standard of review, since reviewing courts need no longer conduct an unwieldy “contextual” inquiry to identify the appropriate standard of review, as had been the case in the period before Vavilov (Vavilov, at paras. 7 and 17).

None of the established exceptions to the presumption of reasonableness review applies in these cases.

First, the legislature has not explicitly prescribed the standard of review or provided a statutory appeal mechanism from an administrative decision to a court. The present cases proceeded to the Federal Court on applications for judicial review under s. 72(1) of the IRPA , which does not prescribe the standard of review.

Second, the standard of correctness is not required based on the rule of law. The proper interpretation of s. 34(1) (e) of the IRPA is not a “general question of law of central importance to the legal system as a whole” under Vavilov (paras. 58‑62). Such “general questions of law” require uniform and consistent answers because of their impact on the administration of justice as a whole or for other institutions of government (para. 59). Examples of general questions of law include questions with legal implications for many other statutes or for the proper functioning of the justice system as a whole (paras. 59‑61). It is not enough for the question to “touc[h] on an important issue” or to raise an issue of “wider public concern” (para. 61). Although the proper interpretation of s. 34(1)(e) is important for the affected persons and the proper administration of the IRPA , it does not affect the legal system or the administration of justice as a whole, have legal implications for many other statutes, or affect other institutions of government. Rather, the issues raised are particular to the interpretation of the conditions for inadmissibility under s. 34(1)(e). Moreover, the proper interpretation of s. 34(1)(e) is not a constitutional question or a question related to the jurisdictional boundaries between two or more administrative bodies, nor does it engage the correctness category recognized in Society of Composers. Thus, the rule of law does not require the Federal Court to review an administrative interpretation of s. 34(1)(e) for correctness.

Nor does the certified question regime under s. 74 (d) of the IRPA displace the presumption of reasonableness review and warrant a new category of correctness. As I will explain, the Federal Court’s certification of a question for appeal to the Federal Court of Appeal provides for a statutory appeal of the Federal Court’s decision to the Federal Court of Appeal, but it does not change the standard of review to be applied by either court.

To begin with, it is obvious that the Federal Court’s decision to certify a serious question of general importance under s. 74(d) does not affect the standard of review to be applied by the Federal Court itself on the application for judicial review under s. 72(1) of the IRPA . The Federal Court does not certify the question until it is rendering its judgment on the application for judicial review. Section 74(d) provides that an appeal to the Federal Court of Appeal may be made “only if, in rendering judgment, the [Federal Court] certifies that a serious question of general importance is involved and states the question”. In effect, the certified question procedure plays a gatekeeping role by requiring the matter to meet a threshold of importance to merit an appeal to the Federal Court of Appeal. As this Court has stated, the certified question “may be the ‘trigger’ by which an appeal is permitted”, but the “subject of the appeal is still the judgment itself, not merely the certified question” (Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 S.C.R. 909, at para. 44, citing Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817, at para. 12). Thus, the certified question regime does not amount to “rare and exceptional circumstances” in which the Federal Court applying reasonableness review would undermine legislative intent or the rule of law in a manner analogous to the existing correctness categories (Society of Composers, at paras. 27 and 41).

The issue is then whether the certification of a serious question of general importance requires correctness review by the Federal Court of Appeal or this Court, either because of a legislative intent to apply correctness review or because it is required by the rule of law. In my view, the answer is no.

As a matter of precedent, this Court has concluded in the immigration context that “[d]espite the presence of a certified question, the . . . standard of review is reasonableness” (Kanthasamy, at para. 44; see also Baker, at para. 62). Vavilov does not require that conclusion to be revisited. Certification of a question under s. 74 (d) of the IRPA does not signal that the legislature intended that appellate courts apply correctness review. As noted above, the Federal Court of Appeal’s task in an appeal from a decision of the Federal Court in an application for judicial review — including in an appeal based on a certified question — is to determine whether the Federal Court identified the appropriate standard of review, and then to decide whether it applied that standard properly (Agraira, at paras. 45‑47; Canada (Citizenship and Immigration) v. Galindo Camayo, 2022 FCA 50, 468 D.L.R. (4th) 358, at para. 38, citing Horrocks, at para. 10). The Federal Court of Appeal must “step into the shoes” of the Federal Court and apply the same standard of review that it should have applied (Galindo Camayo, at para. 38, citing Kanthasamy, at para. 44). Here, the Federal Court had to judicially review the administrative decisions for reasonableness. The Federal Court of Appeal, when stepping into the shoes of the Federal Court on the certified question, had to do likewise. As a result, the certified question regime neither rebuts the presumption of reasonableness, nor alters the Court of Appeal’s task when it hears appeals from first instance judicial review decisions.

What is more, recall that Vavilov itself involved an appeal to the Federal Court of Appeal on a certified question of general importance under s. 22.2 (d) of the Citizenship Act , R.S.C. 1985, c. C-29 , which provides — in terms materially identical to s. 74 (d) of the IRPA — that “an appeal to the Federal Court of Appeal may be made only if, in rendering judgment, the judge certifies that a serious question of general importance is involved and states the question”. This Court in Vavilov reviewed the administrative decision at issue for reasonableness, noting that there was “no indication that the legislature intended a standard of review other than reasonableness to apply” (para. 170).

Finally, recognizing a new correctness category here would conflict with Vavilov’s goal of simplifying and making more predictable the standard of review framework by providing only limited exceptions to reasonableness review (para. 47). Treating s. 74(d) as justifying correctness review would effectively reintroduce a “contextual” approach to the standard of review — with the certification of a serious question of general importance being a “contextual” factor suggesting correctness — and thus would revive the approach that Vavilov eliminated because it created “uncertainty” and was “unwieldy” (para. 7; see also P. Daly, “Unresolved Issues after Vavilov” (2022), 85 Sask. L. Rev. 89, at pp. 91-92 (Vavilov is “an exercise in simplification and clarification” that “excised” the “‘vexing contextual factors’ . . . from the standard of review selection exercise”.)).

I conclude that no exception to the presumption of reasonableness applies in these cases. As a result, the standard of review of the decisions of the IAD in Mr. Mason’s case and the ID in Mr. Dleiow’s case is reasonableness.

I now turn to the second issue in these appeals: whether the administrative decisions under review were reasonable. This section recaps Vavilov’s guidance on conducting reasonableness review and comments briefly on the methodology of reasonableness review conducted by the courts below. The following section then applies reasonableness review to the two administrative decisions at issue.

Vavilov provided extensive guidance on conducting reasonableness review of administrative decisions (paras. 73-142). Without canvassing every detail of that guidance, the main elements of reasonableness review can be summarized as follows.

Vavilov explained that the purpose of reasonableness review is “to give effect to the legislature’s intent to leave certain decisions with an administrative body while fulfilling the constitutional role of judicial review to ensure that exercises of state power are subject to the rule of law” (para. 82). Reasonableness review starts from a posture of judicial restraint and “a respect for the distinct role of administrative decision makers” (para. 13), arising from the legislature’s institutional design choice to give administrative decision makers rather than courts the jurisdiction to decide certain issues (para. 24). Reasonableness review also serves to “maintain the rule of law” (para. 2) and “to safeguard the legality, rationality and fairness of the administrative process” (para. 13). Thus, the purpose of reasonableness review is to uphold “the rule of law, while according deference to the statutory delegate’s decision” (Canada Post, at para. 29).

Vavilov noted that, given the deference owed to an administrative decision, reasonableness review is “methodologically distinct” from correctness review (para. 12). The Court explained that “[w]hat distinguishes reasonableness review from correctness review is that the court conducting a reasonableness review must focus on the decision the administrative decision maker actually made, including the justification offered for it, and not on the conclusion the court itself would have reached in the administrative decision maker’s place” (para. 15). Reasonableness review is thus concerned with both the administrator’s decision-making process and the outcome (paras. 83 and 87; see also Canada Post, at para. 29).

A decision will be unreasonable when the reasons “fail to provide a transparent and intelligible justification” for the result (para. 136). A reviewing court must therefore take a “reasons first” approach that evaluates the administrative decision maker’s justification for its decision (para. 84). It must “begin its inquiry into the reasonableness of a decision by examining the reasons provided with ‘respectful attention’ and seeking to understand the reasoning process followed by the decision maker to arrive at its conclusion” (para. 84, citing D. Dyzenhaus, “The Politics of Deference: Judicial Review and Democracy”, in M. Taggart, ed., The Province of Administrative Law (1997), 279, at p. 286). As noted by Professor David Mullan, the “reasons first” approach “underscores a commitment to deference” and requires that reasons are “the principal lens through which the exercise of reasonableness review takes place” (p. 202). Thus, as he explains, “the starting or focal point for the conducting of truly deferential reasonableness review should be the reasons provided by the decision-maker” (p. 215; see also Daly (2022), at pp. 108-10).

Under Vavilov’s “reasons first” approach, the reviewing court should remember that “the written reasons given by an administrative body must not be assessed against a standard of perfection”, and need not “include all the arguments, statutory provisions, jurisprudence or other details the reviewing judge would have preferred” (para. 91). The reviewing judge must read the administrator’s reasons “holistically and contextually” (para. 97), “in light of the history and context of the proceedings in which they were rendered”, including “the evidence before the decision maker, the submissions of the parties, publicly available policies or guidelines that informed the decision maker’s work, and past decisions of the relevant administrative body” (para. 94). Reasons must be read “in light of the record and with due sensitivity to the administrative regime in which they were given” (para. 103). Such factors may “explain an aspect of the decision maker’s reasoning process that is not apparent from the reasons themselves, or may reveal that an apparent shortcoming in the reasons is not, in fact, a failure of justification, intelligibility or transparency” (para. 94).

A reviewing court should also avoid engaging in “disguised correctness review”, or correctness in the guise of reasonableness (para. 294, per Abella and Karakatsanis JJ., concurring in the result; see also Wilson v. Atomic Energy of Canada Ltd., 2016 SCC 29, [2016] 1 S.C.R. 770, at para. 27, citing D. Mullan, “Unresolved Issues on Standard of Review in Canadian Judicial Review of Administrative Action — The Top Fifteen!” (2013), 42 Adv. Q. 1, at pp. 76‑81). Because “[t]he role of courts in these circumstances is to review”, they should, as a general rule, “refrain from deciding the issue themselves” (Vavilov, at para. 83 (emphasis in original)). A reviewing court should not create its “own yardstick and then use [it] to measure what the administrator did” (para. 83, and Canada Post, at para. 40, both citing Delios v. Canada (Attorney General), 2015 FCA 117, 100 Admin. L.R. (5th) 301, at para. 28). Nor should a reviewing court ask “what decision it would have made in place of that of the administrative decision maker, attempt to ascertain the ‘range’ of possible conclusions that would have been open to the decision maker, conduct a de novo analysis or seek to determine the ‘correct’ solution to the problem” (Vavilov, at para. 83; see also Canada Post, at para. 40). Rather, a “reviewing court must consider only whether the decision made by the administrative decision maker — including both the rationale for the decision and the outcome to which it led — was unreasonable” (Vavilov, at para. 83).

Finally, Vavilov cautioned that the “reasons first” approach is “not a ‘rubber-stamping’ process or a means of sheltering administrative decision makers from accountability” (para. 13). Instead, it is “a robust form of review” (para. 13; see also paras. 12, 67 and 138), one that highlights “the need to develop and strengthen a culture of justification in administrative decision making” (para. 2).

Vavilov identified two types of “fundamental flaws” indicating that an administrative decision is unreasonable: (1) a failure of rationality internal to the reasoning process; or (2) a failure of justification given the legal and factual constraints bearing on the decision (para. 101). A reviewing court need not categorize unreasonableness as falling into one category or another. They are simply a helpful way of describing how a decision may be unreasonable (para. 101).

A failure of rationality in the reasoning process arises if the decision is not rational or logical (paras. 102‑4). A decision is unreasonable if, “read holistically”, it “fail[s] to reveal a rational chain of analysis” or “reveal[s] that the decision was based on an irrational chain of analysis” (para. 103). A reviewing court “must be able to trace the decision maker’s reasoning without encountering any fatal flaws” in the decision maker’s “overarching logic” (para. 102). It must “be satisfied that the decision maker’s reasoning ‘adds up’” (para. 104).

A failure of justification in light of the legal and factual constraints bearing on the decision arises if the decision is not “justified in relation to the constellation of law and facts that are relevant to the decision” (para. 105). The legal and factual context “operate as constraints on the decision maker in the exercise of its delegated powers” (para. 105). The burden of justification varies with the circumstances, including the wording of the relevant statutory provisions, the applicable precedents, the evidence, the submissions of the parties, and the impact of the decision on the affected persons. The greater the interpretive constraints in a given case, the greater the burden of justification on the decision maker in deviating from those constraints (see M. Popescu, “L’arrêt Vavilov: à la recherche de l’équilibre perdu entre la primauté du droit et la suprématie législative” (2021), 62 C. de D. 567, at p. 603). Examples include the seven non-exhaustive constraints set out below. As was highlighted in Vavilov, “[t]hese elements are not a checklist for conducting reasonableness review, and they may vary in significance depending on the context. They are offered merely to highlight some elements of the surrounding context that can cause a reviewing court to lose confidence in the outcome reached” (para. 106).

Whether an interpretation of the governing statutory scheme is justified will “depend on the context, including the language chosen by the legislature in describing the limits and contours of the decision maker’s authority” (para. 110). Narrower and more precise language imposes a greater constraint on the decision maker, while “broad, open-ended or highly qualitative language” affords greater flexibility (para. 110). What matters is whether the decision maker has “properly justified its interpretation of the statute in light of the surrounding context” (para. 110).

As already noted, a court evaluating the reasonableness of an administrative decision on a question of statutory interpretation “does not undertake a de novo analysis of the question or ‘ask itself what the correct decision would have been’” (para. 116). Instead, the court “must examine the administrative decision as a whole, including the reasons provided by the decision maker and the outcome that was reached” (para. 116).

Although an administrative decision maker need not “engage in a formalistic statutory interpretation exercise in every case” (para. 119), its decision must be consistent with the “modern principle” of statutory interpretation, which focusses on the text, context, and purpose of the statutory provision. The decision maker must demonstrate in its reasons that it was alive to those essential elements (para. 120). The omission of a minor aspect of the text, context, or purpose is unlikely to undermine the decision as a whole: omissions are not “stand-alone grounds for judicial intervention” (para. 122). In each case, “the key question is whether the omitted aspect of the analysis causes the reviewing court to lose confidence in the outcome reached by the decision maker” (para. 122). For example, an administrative interpretation may well be unreasonable if it fails to consider the potentially harsh consequences of its interpretation of a statutory provision for a large class of individuals, and whether, in light of those consequences, the legislature would have intended the provision to apply in that way (paras. 191-92). And even if a decision does not explicitly consider the meaning of a relevant provision, the court may be able to discern the interpretation adopted from the record and evaluate whether it is reasonable (para. 123).

In interpreting a statute, an administrative decision maker may draw on its institutional expertise and experience and rely on considerations that a court would not have thought to employ, but which “enrich and elevate the interpretive exercise” (paras. 93 and 119; Canada Post, at para. 43). As Professor Audrey Macklin explains, courts should be “genuinely receptive to input beyond the usual techniques that courts use to discern text, context and purpose. These may include operational implications, alignment with broader statutory mandate, and so on” (“Seven Out of Nine Legal Experts Agree: Expertise No Longer Matters (in the Same Way) After Vavilov!” (2021), 100 S.C.L.R. (2d) 249, at p. 261). By being receptive to such factors, courts acknowledge that administrative decision makers have a role to play in elaborating the content of the schemes that they administer (Vavilov, at para. 108). Reasonableness review demands both that administrative decision makers demonstrate their expertise through their reasons and that judges pay “[r]espectful attention” to the ways in which their reasons reflect that expertise (para. 93; P. Daly, “Vavilov and the Culture of Justification in Contemporary Administrative Law” (2021), 100 S.C.L.R. (2d) 279, at pp. 285‑86).

Finally, a court may conclude during a reasonableness review that “the interplay of text, context and purpose leaves room for a single reasonable interpretation of the statutory provision, or aspect of the statutory provision” (Vavilov, at para. 124, citing Dunsmuir, at paras. 72‑76, and Nova Tube Inc./Nova Steel Inc. v. Conares Metal Supply Ltd., 2019 FCA 52). In such a case, although a court should “generally pause before definitively pronouncing upon the interpretation” of a statutory provision, the court may conclude that remitting the question to the administrative decision maker may serve no useful purpose (Vavilov, at para. 124). It must be stressed that the possibility of a single reasonable interpretation is not a starting point of reasonableness review, as this would be contrary to a “reasons first” approach. Rather, it is a conclusion that a reviewing court may draw as a result of a proper reasonableness review, as part of the court’s consideration of the appropriate remedy.

Statutory law, common law, and international law may operate as legal constraints on an administrative decision maker (paras. 111 and 114). An administrative decision will be unreasonable if it fails to justify a departure from binding precedents (para. 112). International law can also operate as an important constraint, arising from the presumption that legislation is presumed to operate in conformity with Canada’s international obligations and the values and principles of customary and conventional international law, or by informing whether a decision was a reasonable exercise of administrative authority (para. 114).

Absent exceptional circumstances, a reviewing court will defer to an administrative decision maker’s factual findings (para. 125). A reviewing court may intervene, however, if the decision is unreasonable: if it is not “justified in light of the facts” or when “the decision maker has fundamentally misapprehended or failed to account for the evidence before it” (para. 126).

An administrative decision maker’s reasons must “meaningfully account for the central issues and concerns raised by the parties” (para. 127). Reasons must be “responsive” to the parties’ submissions, because reasons are the “primary mechanism by which decision makers demonstrate that they have actually listened to the parties” (para. 127 (emphasis in original)). Although an administrative decision does not have to “respond to every argument or line of possible analysis” raised by the parties, “a decision maker’s failure to meaningfully grapple with key issues or central arguments raised by the parties may call into question whether the decision maker was actually alert and sensitive to the matter before it” (para. 128).

Administrative decision makers should be concerned with the general consistency of their decisions, even if they are not bound by their prior decisions in the same way that courts are bound by stare decisis (para. 129). A decision will be unreasonable if the reasons fail to meet the “justificatory burden” for departing from “longstanding practices or established internal authority” (para. 131).

Vavilov also explained that “[w]here the impact of a decision on an individual’s rights and interests is severe, the reasons provided to that individual must reflect the stakes” (para. 133). The principle of “responsive justification” means that if a decision has “particularly harsh consequences for the affected individual”, then “the decision maker must explain why its decision best reflects the legislature’s intention” (para. 133). An administrative decision may be unreasonable if it fails to grapple with particularly severe or harsh consequences for the affected individual (para. 134). An administrative decision maker’s reasons must “demonstrate that they have considered the consequences of a decision and that those consequences are justified in light of the facts and law” (para. 135).

Having set out Vavilov’s guidance on conducting reasonableness review, I now comment briefly on the approach to reasonableness review of the courts below.

The Federal Court in Mr. Mason’s case did not have the benefit of Vavilov, and therefore did not apply Vavilov’s “reasons first” approach in judicially reviewing the administrative decisions. The Federal Court essentially conducted its own freestanding interpretation of s. 34(1)(e) based on the provision’s text, context, and purpose (paras. 38‑51), and then evaluated the administrative decisions against that interpretation (paras. 52‑62). As the Court of Appeal noted, this involved the Federal Court “fashioning its own yardstick to measure the administrator’s interpretation and interfering if the difference is too much” (para. 24). This approach was eschewed in this Court’s decision in Vavilov.

But the Federal Court of Appeal also strayed from Vavilov’s methodology of reasonableness review. The Court of Appeal grafted onto Vavilov an extra step of “conducting a preliminary analysis of the text, context and purpose of the legislation just to understand the lay of the land before . . . examin[ing] the administrators’ reasons” (para. 17). The parties before this Court contended that this preliminary step is inconsistent with Vavilov. The respondent Minister of Citizenship and Immigration — who otherwise agreed with the Court of Appeal’s conclusion — submitted that the Court of Appeal’s approach “should not be adopted”, and urged that “[t]he focus in the reasonableness analysis needs to remain, as this Court has instructed, on the reasons of the decision-maker, and not on a range of potential conclusions to be determined by a reviewing court in the abstract” (R.F., at para. 54). I agree. Vavilov is clear that a reviewing court must start its analysis with the reasons of the administrative decision maker; starting with its own perception of the merits may lead a court to slip into correctness review.

I now turn to consider whether the administrative decisions under review reasonably interpreted s. 34(1) (e) of the IRPA as not requiring a nexus with national security or the security of Canada.

In reviewing the IAD’s reasons, I recall this Court’s instruction in Vavilov that a reviewing court should conduct reasonableness review mindful of the impact of the decision on the affected individual. The principle of “responsive justification” means that “[w]here the impact of a decision on an individual’s rights and interests is severe, the reasons provided to that individual must reflect the stakes” (para. 133). Here, the interpretation of s. 34(1)(e) will affect whether two individuals — one of whom has not been convicted of a criminal offence — could be deported from Canada. As this Court has noted, individuals facing deportation may experience “any number of serious life-changing consequences”, including dislocation or permanent separation from their family (R. v. Wong, 2018 SCC 25, [2018] 1 S.C.R. 696, at para. 72, per Wagner J. (as he then was), dissenting). The IAD’s reasons must reflect these stakes.

I begin by summarizing in greater detail the interpretation of s. 34(1)(e) in the reasons of the IAD in Mr. Mason’s case, which were followed by the ID in Mr. Dleiow’s case, before addressing what the appellants say are failures of justification in the IAD’s reasons.

The IAD in Mr. Mason’s case ruled that a person can be inadmissible under s. 34(1) (e) of the IRPA even without violent conduct linked to national security or the security of Canada. It decided the appeal based on the parties’ written submissions, and concluded that Parliament intended s. 34(1)(e) to “relate to security in a broader sense”, including by ensuring that “individual Canadians are secure from acts of violence that would or might endanger their lives or safety” (para. 37). The IAD’s chain of reasoning was as follows:

• Section 34(1)(e) cannot be read in isolation. The provision must be interpreted using the modern approach to statutory interpretation. The words used in s. 34(1)(e) must be assessed “in their entire context and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament” (para. 17, citing Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27, at para. 21).

• The other paragraphs of s. 34(1) all have a nexus to national security, which may be a signal of Parliament’s intent, but is not determinative. Context comes not just from “the immediate co-text, but from the overall scheme and object of the IRPA ” (para. 21). The inadmissibility provisions in Part 1, Division 4 of the IRPA are “particularly relevant” (para. 21).

• In at least one prior decision, X (Re), 2017 CanLII 146735 (I.R.B. (Imm. Div.)), in which a foreign national allegedly assaulted his former girlfriends in Canada but was not convicted of any offence in Canada, the ID interpreted the terms “security” and “security grounds” in s. 34(1)(e) as meaning “the security of Canada” or “national security”. But this interpretation is “not consistent with the presumption of consistent expression” (IAD reasons, at para. 23). The terms “security” and “security grounds” in s. 34 must have different meanings from the terms “the security of Canada” or “national security” used elsewhere in the IRPA , including in the phrase “being a danger to the security of Canada” in s. 34(1)(d), which would otherwise be redundant.

• The dictionary definition of “security”, which includes a “secure condition or feeling”, provides useful guidance (para. 25). This is a “broad definition” that “does not necessarily incorporate a national security element” (para. 25).

• Obiter comments in prior cases tend to support the Minister of Public Safety and Emergency Preparedness’ position on the interpretation of s. 34(1)(e) (para. 27 (emphasis added), citing El Werfalli v. Canada (Public Safety and Emergency Preparedness), 2013 FC 612, [2014] 4 F.C.R. 673, at para. 75, which stated that “[t]he provision of reasonable grounds to believe an organization may engage in terrorism in the future serves to maintain national security and public safety being the object of the subsection 34(1)”; para. 28 (emphasis added), citing Fuentes v. Canada (Minister of Citizenship and Immigration), 2003 FCT 379, [2003] 4 F.C. 249, at para. 62, which interpreted a predecessor provision, s. 19 of the Immigration Act, R.S.C. 1985, c. I-2, as directed at the “key concepts” of “subversion, terrorism, crimes against humanity, war crimes and ordinary crimes”).

• Section 36(1) (a) of the IRPA provides for inadmissibility based on the commission of a criminal offence in Canada, and requires a conviction. Section 34(1)(e) creates a different ground of inadmissibility based on conduct, “described in terms of the danger posed to the lives and safety of persons in Canada” (para. 33). Such conduct is “a small subset of what would be considered serious criminality in section 36 of the IRPA ”, and “is distinct from criminal law” (para. 33). Sections 34 and 36 “overlap but are distinct” (para. 38). Section 36 addresses criminal offences, while s. 34 addresses danger posed to the lives and safety of persons in Canada.

• Because inadmissibility is not a criminal sanction, it does not offend Canadian values to find a person inadmissible for acts that were “arguably criminal, but which did not lead to a criminal conviction” (para. 35). The conduct described in s. 34(1)(e) “is narrowly defined and anchored in terms of the danger posed to Canadians, not to criminal law” (para. 36).

Plainly, the IAD’s reasons applied several recognized techniques of statutory interpretation. The IAD adverted to the modern approach to statutory interpretation; attempted to read s. 34(1)(e) in the context of the rest of s. 34 and the broader context of the grounds of inadmissibility in Division 4 of the IRPA ; relied on a dictionary definition of “security”; considered the presumption of consistent expression to give the terms “security” and “security grounds” in s. 34(1)(e) a distinct meaning from the terms “the security of Canada” and “national security” used elsewhere in the IRPA ; and considered the only prior decision, a ruling of the ID, that had interpreted s. 34(1)(e) and was consistent with Mr. Mason’s position, as well as obiter statements from two Federal Court decisions that it viewed as supporting the Minister of Public Safety and Emergency Preparedness’ position. On balance, the IAD concluded that the Minister’s interpretive arguments outweighed Mr. Mason’s.

As I will elaborate, although the IAD considered several of Mr. Mason’s arguments, it failed to address significant legal constraints that he had raised in his written submissions to the IAD: (1) two points of statutory context, and (2) the broad consequences of its decision. The IAD’s decision also failed to address (3) constraints imposed by international law that s. 3(3) (f) of the IRPA requires to be considered in interpreting and applying the legislation. These omissions establish that the IAD’s decision was unreasonable.

The IAD failed to address two significant points of statutory context that Mr. Mason raised before the IAD in support of his argument that s. 34(1)(e) requires a nexus with national security or the security of Canada. Together, these omissions involve a failure of responsive justification.

First, Mr. Mason submitted before the IAD that s. 34(1)(e) requires a nexus with national security or the security of Canada partly because the ministerial relief from inadmissibility under s. 34 (security grounds) is narrower than that available under s. 36 (serious criminality and criminality). He claimed that this narrower relief suggests that s. 34 is a “grave” form of inadmissibility concerned with “national security” (A.R., vol. II, at p. 30).

Mr. Mason submitted as follows. If a foreign national is inadmissible on security grounds under s. 34, the relief available is narrow: the Minister of Public Safety and Emergency Preparedness may declare that the matters referred to in s. 34 “do not constitute inadmissibility in respect of the foreign national if they satisfy the Minister that it is not contrary to the national interest” (IRPA , s. 42.1(1)). But a foreign national inadmissible under s. 36 for being convicted of a criminal offence in Canada may seek broader relief: they may be granted ministerial relief from inadmissibility on humanitarian and compassionate grounds (s. 25(1)), and may not be found inadmissible if a record suspension for the conviction is in effect under the Criminal Records Act , R.S.C. 1985, c. C-47 (IRPA , s. 36(3) (b)).

Here, for example, because Mr. Mason was not convicted of a criminal offence and his alleged conduct would make him inadmissible under the IAD’s interpretation of s. 34(1)(e), he could seek relief from inadmissibility only if he satisfied the Minister of Public Safety and Emergency Preparedness that it is in the “national interest”. But had Mr. Mason been convicted of a criminal offence for the same conduct, he could have applied for ministerial relief from inadmissibility on humanitarian and compassionate grounds, and he would not have been inadmissible at all had he been convicted of a criminal offence and later pardoned under the Criminal Records Act .

Mr. Mason argued that because the relief available to persons inadmissible on security grounds under s. 34 is narrower than the relief available to persons inadmissible for criminality under s. 36, s. 34 was among the “grave forms of inadmissibility” and should “be interpreted in the context of national security” (A.R., vol. II, at p. 30).

The IAD’s reasons, read with sensitivity and in light of the record, did not address this important contextual argument. While perhaps not in itself determinative, this argument supported Mr. Mason’s position and imposed a significant legal constraint on the interpretation of s. 34(1)(e).

Second, Mr. Mason submitted before the IAD that s. 34(1)(e) requires a nexus with national security or the security of Canada partly because the Minister of Citizenship and Immigration must consider different criteria when conducting a “pre‑removal risk assessmen[t]” for a foreign national found inadmissible under s. 34 (security grounds) and under s. 36 (serious criminality and criminality) (A.R., vol. II, at pp. 30‑31). He submitted that in an assessment involving s. 36 the Minister must consider whether the person poses a danger to the public in Canada. In contrast, in an assessment involving s. 34 the Minister must consider whether the person poses a danger to the security of Canada. He claimed that this distinction supported his position that the security grounds under s. 34 require a nexus to national security or the security of Canada.

A pre-removal risk assessment is a process under the IRPA by which a person subject to a removal order may apply to the Minister of Citizenship and Immigration for protection, resulting in refugee protection or a stay of the removal order (ss. 112 and 114(1)). In any application for a pre-removal risk assessment, the Minister must consider the danger that the person would be subjected to torture, the risk to their life, and the risk they would be subjected to cruel and unusual treatment or punishment (s. 97). For persons found inadmissible under s. 36(1) for serious criminality, the Minister must also consider “whether they are a danger to the public in Canada” (s. 113(d)(i)). But for persons found inadmissible on security grounds under s. 34, the Minister must, in addition to the s. 97 factors, consider “whether the application should be refused because of the nature and severity of acts committed by the applicant or because of the danger that the applicant constitutes to the security of Canada” (s. 113(d)(ii)).

Mr. Mason argued before the IAD that these diverging considerations for pre-removal risk assessments suggest that Parliament contemplated that conduct captured by s. 36 poses a danger to the public in Canada, while conduct captured by s. 34 poses a danger to the security of Canada (A.R., vol. II, at pp. 30‑31). He claimed that this supported his position that the security grounds under s. 34 require a nexus to national security or the security of Canada.

Again, the IAD’s reasons did not address this important contextual argument, which, while not in itself determinative, supported Mr. Mason’s position and imposed a significant legal constraint on the interpretation of s. 34(1)(e).

The Federal Court accepted that the IAD failed to address these two important contextual arguments that Mr. Mason had raised (para. 53). The Federal Court of Appeal disagreed, stating that the IAD had considered these two arguments “implicitly” (para. 59). In its view, “[t]o the extent that the [IAD] failed to mention some elements in its analysis of text, context and purpose, this was not a fundamental gap” (para. 59). The Court of Appeal said that “although one can quibble that certain elements of text, context and purpose were not mentioned in the reasons”, the court was “confident from the quality of the [IAD]’s overall reasoning that it considered them to be outweighed by other elements” (para. 59).

I respectfully disagree with the Court of Appeal. I see no basis to conclude that the IAD considered these two important points of statutory context, even implicitly. Mr. Mason expressly raised both points as core planks supporting his position. The IAD’s failure to address them, while addressing other points, casts into doubt whether it was alert and sensitive to these issues (Vavilov, at paras. 127‑28). Reasons are the primary mechanism for the IAD to demonstrate that it actually listened to Mr. Mason (para. 127). In my view, the IAD’s reasons did not address — far less meaningfully grapple with — two key arguments that Mr. Mason had raised. The IAD’s reasons therefore failed to meet Vavilov’s standard of responsive justification (para. 127).

The IAD also failed to address Mr. Mason’s submission that interpreting s. 34(1)(e) without a nexus to national security or the security of Canada would result in two broad consequences, which he claimed ran afoul of the principle of statutory interpretation that a legislature does not intend to produce absurd consequences (Rizzo, at para. 27; Vavilov, at para. 120). The IAD’s failure to address these two broad consequences involves another failure of responsive justification.

First, Mr. Mason submitted that interpreting s. 34(1)(e) without a nexus to national security or the security of Canada would mean that “any act of violence against another individual would presumably result in one of the most grave forms of inadmissibility” (A.R., vol. II, at p. 31). The acts of violence caught by s. 34(1)(e) would, he submitted, extend from “domestic altercations” to “bar fights and schoolyard fights” (p. 31). I note that unlike a criminal conviction triggering inadmissibility under s. 36, which must be proven beyond a reasonable doubt, the facts triggering inadmissibility under s. 34 are subject to the much lower standard of “facts for which there are reasonable grounds to believe that they have occurred, are occurring or may occur” (IRPA , s. 33). Thus, s. 34(1)(e) would extend to any “acts of violence” — past, present, or future — that “would or might” “endanger the lives or safety” of “persons in Canada”. Mr. Mason submitted that it would be absurd for such acts to be captured by s. 34(1)(e).

Unlike the Federal Court of Appeal, I do not read the IAD’s reasons as having addressed this argument. The Federal Court of Appeal interpreted the IAD’s reasons as interpreting “safety” as “something approaching the level of a threat to life, not just minor harm” (para. 57). The court based this conclusion on the IAD’s observation that s. 34(1)(e) is “narrowly defined and anchored in terms of the danger posed to Canadians, not to criminal law”. But the IAD made this statement in explaining that immigration consequences under the IRPA are distinct from criminal sanctions and are subject to different evidentiary thresholds; it did not say that s. 34(1)(e) applied only to a small class of especially serious violent conduct approaching a threat to life. As the IAD explained:

. . . immigration consequences under the IRPA are not criminal sanctions. Criminal law and the IRPA have different objects. The IRPA operates under a different scheme including, for example, a lower evidentiary threshold. A person can be found inadmissible for conduct even if they have not been convicted criminally for that conduct. The conduct described in paragraph 34(1)(e) is narrowly defined and anchored in terms of the danger posed to Canadians, not to criminal law. [para. 36]

With respect, the Court of Appeal effectively buttressed the IAD’s reasons to provide a justification that the IAD did not itself provide, contrary to Vavilov’s direction that “it is not open to a reviewing court to disregard the flawed basis for a decision and substitute its own justification for the outcome” (para. 96). This Court warned that “[t]o allow a reviewing court to do so would be to allow an administrative decision maker to abdicate its responsibility to justify to the affected party, in a manner that is transparent and intelligible, the basis on which it arrived at a particular conclusion” (para. 96).

As for the second broad consequence, Mr. Mason submitted to the IAD that if s. 34(1)(e) is interpreted without a nexus to national security or the security of Canada, it would do an “end-run around the limitations under s. 36(3)(e) with respect to youth offences” (A.R., vol. II, at pp. 31‑32). Section 36(3) (e) of the IRPA provides that young persons who are permanent residents or foreign nationals found guilty under the Young Offenders Act , R.S.C. 1985, c. Y-1 , or who received a youth sentence under the Youth Criminal Justice Act , S.C. 2002, c. 1 , cannot be found inadmissible under s. 36(1) (serious criminality) or s. 36(2) (criminality). In effect, under s. 36(3)(e), Parliament exempted young persons from inadmissibility for most criminal offences. But on the IAD’s interpretation of s. 34(1)(e), young persons can be found inadmissible for any acts of violence that would or might engage the lives or safety of persons in Canada, even without a nexus to national security or the security of Canada, and even without a criminal conviction.

Neither the IAD nor the Federal Court of Appeal addressed this incompatibility of the IAD’s interpretation of s. 34(1)(e) with the scheme for the inadmissibility of young persons under the IRPA . This omission was also not a minor aspect of s. 34(1)(e)’s interpretive context, and should have been addressed.

Finally, the IAD failed to address the legal constraints imposed by international law on its interpretation of s. 34(1)(e). As I will explain, the IAD’s interpretation allows foreign nationals to be returned to countries where they may face persecution, contrary to Canada’s non-refoulement obligation in Article 33 of the Refugee Convention. By contrast, interpreting s. 34(1)(e) as requiring a nexus with national security or the security of Canada means that a removal order would not breach Article 33. Although this argument was not presented to the IAD, the IAD was required by its home statute to interpret and apply the IRPA in a manner that complies with Canada’s international human rights obligations, including Canada’s non-refoulement obligation under Article 33 of the Refugee Convention.

Vavilov highlighted that international law may be an “important constraint on an administrative decision maker”, including through the presumption of statutory interpretation that “legislation is presumed to operate in conformity with Canada’s international obligations” (para. 114). Canada has ratified both the 1951 Refugee Convention and the 1967 Refugee Protocol (Németh v. Canada (Justice), 2010 SCC 56, [2010] 3 S.C.R. 281, at para. 17). These international human rights instruments to which Canada is a party trigger the interpretive presumption of conformity with international law.

The presumption of conformity with international law assumes added force when interpreting the IRPA , because Parliament has made its “presumed intent to conform to Canada’s international obligations explicit” through two provisions of the IRPA (B010 v. Canada (Citizenship and Immigration), 2015 SCC 58, [2015] 3 S.C.R. 704, at para. 49). First, s. 3(2)(b) of the IRPA expressly identifies one of the IRPA ’s objectives as being “to fulfil Canada’s international legal obligations with respect to refugees and affirm Canada’s commitment to international efforts to provide assistance to those in need of resettlement”. Indeed, this Court has described the IRPA as the “main legislative vehicle for implementing Canada’s international refugee obligations” (Németh, at para. 21). Second, s. 3(3) (f) of the IRPA instructs courts and administrative decision makers to construe and apply the IRPA in a manner that “complies with international human rights instruments to which Canada is signatory” (B010, at para. 49). This Court has stated that “[t]here can be no doubt that the Refugee Convention is such an instrument, building as it does on the right of persons to seek and to enjoy asylum from persecution in other countries as set out in art. 14 of the Universal Declaration of Human Rights, G.A. Res. 217 A (III), U.N. Doc. A/810, at 71 (1948)” (para. 49). As a result, the Refugee Convention is “determinative of how the IRPA must be interpreted and applied, in the absence of a contrary legislative intention” (de Guzman v. Canada (Minister of Citizenship and Immigration), 2005 FCA 436, [2006] 3 F.C.R. 655, at para. 87; B010, at para. 49).

The centrepiece of the Refugee Convention is contained in the provisions relating to “expulsion and return” (Németh, at para. 18). Article 33, which has been expressly incorporated into the IRPA (s. 115), bars the expulsion or return of a refugee, by any means, to a country where they are at risk of persecution, unless they are found to pose a danger to the security of the host country or are convicted of a serious crime. Article 33 of the Refugee Convention provides:

Article 33

Prohibition of Expulsion or Return (“Refoulement”)

1. No Contracting State shall expel or return (“refouler”) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion.

2. The benefit of the present provision may not, however, be claimed by a refugee whom there are reasonable grounds for regarding as a danger to the security of the country in which he is, or who, having been convicted by a final judgment of a particularly serious crime, constitutes a danger to the community of that country.

Article 33(1) enshrines the principle of non-refoulement, which has been described as “the cornerstone of the international refugee protection regime”, and which, generally, “prohibits the direct or indirect removal of refugees to a territory where they run a risk of being subjected to human rights violations” (Németh, at paras. 18‑19). Article 33(2), which operates as a limited exception to the principle of non-refoulement in Article 33(1), allows a person to be removed in exceptional circumstances: when there are reasonable grounds for regarding the person as a danger to the security of the country in which they are, or when the person is convicted of a serious crime and is a danger to the community of that country (see Febles v. Canada (Citizenship and Immigration), 2014 SCC 68, [2014] 3 S.C.R. 431, at para. 25). Article 42 of the Refugee Convention further stipulates that ratifying states may not make reservations to the non-refoulement protections of Article 33 (Németh, at para. 18). The principle of non-refoulement is generally recognized as a norm of customary international law (see Prosecutor v. Germain Katanga, ICC-01/04-01/07, Decision on the application for the interim release of detained Witnesses, 1 October 2013 (Trial Chamber II), at para. 30; Zaoui v. Attorney-General (No. 2), [2005] 1 N.Z.L.R. 690 (C.A.), at paras. 34-35; S. E. Lauterpacht and D. Bethlehem, “The scope and content of the principle of non-refoulement: Opinion”, in E. Feller, V. Türk and F. Nicholson, eds., Refugee Protection in International Law: UNHCR’s Global Consultations on International Protection (2003), 87, at paras. 193-253; H. Lambert, “Customary Refugee Law”, in C. Costello, M. Foster and J. McAdam, eds., The Oxford Handbook of International Refugee Law (2021), 240, at pp. 242-49; and United Nations High Commissioner for Refugees, Advisory Opinion on the Extraterritorial Application of Non-Refoulement Obligations under the 1951 Convention relating to the Status of Refugees and its 1967 Protocol (2007), at paras. 14-16).

The IAD’s interpretation allows a foreign national found inadmissible under s. 34(1)(e) to be subject to refoulement contrary to Article 33(1) of the Refugee Convention. On the IAD’s interpretation, a foreign national can be deported to persecution once they are found inadmissible under s. 34(1)(e), without a finding that the person poses a danger to the security of Canada or even if they have not been convicted of a serious offence. Such a person would be entitled to the benefit of Article 33(1) of the Refugee Convention, as the exceptions under Article 33(2) would not apply: on the IAD’s approach to inadmissibility under s. 34(1)(e), there need not be “reasonable grounds” to regard the foreign national as a “danger to the security” of Canada, or for them to have been “convicted by a final judgment of a particularly serious crime”.

A person facing such a removal order would not be protected from refoulement under Article 33(1) through the IRPA ’s pre-removal risk assessment process. Although such a person may apply to the Minister of Citizenship and Immigration for a pre-removal risk assessment (IRPA , s. 112), the grounds that the Minister must consider in assessing the risk to the person if the removal order is not stayed do not include those under Article 33(1) of the Refugee Convention. The Minister does not consider whether the person is a “Convention refugee”, that is, whether the person has a well-founded fear of persecution for reasons of race, religion, nationality, membership in a particular social group, or political opinion (s. 96). Instead, the Minister need consider only (1) whether the application for a stay of the removal order should be refused because of the nature and severity of acts committed by the person or because of the danger that the person constitutes to the security of Canada, and (2) whether the person’s removal would subject them to a danger of torture, to a risk to their life, or to a risk of cruel and unusual treatment or punishment (ss. 97(1) and 113(d)(ii)). Thus, as fairly conceded by the respondent Minister, the Minister does not consider the forms of persecution described in Article 33(1) (R.F., at para. 106). And while there are several “safety valves” under the IRPA that provide discretionary exemptions from the application of the general ineligibility rules under the IRPA (see Canadian Council for Refugees v. Canada (Citizenship and Immigration), 2023 SCC 17, at paras. 43-48), none of these discretionary exemptions detract from the concern that the IAD’s interpretation of s. 34(1)(e) would, as a general rule, allow for a removal order without protection from refoulement, contrary to Article 33(1) of the Refugee Convention.

As a result, interpreting s. 34(1) (e) of the IRPA as not requiring a nexus with national security or the security of Canada — as did the IAD — could subject persons to refoulement contrary to Article 33(1) of the Refugee Convention. By contrast, interpreting s. 34(1)(e) as requiring a nexus with national security or the security of Canada would trigger the exception in Article 33(2) to the ability to claim the protection of Article 33(1) of the Refugee Convention, and thus a removal order in such circumstances would not breach Canada’s obligation of non-refoulement.

The respondent Minister contends that s. 115(1) and (2) of the IRPA operate as a safeguard against refoulement and “fulfil[l] Canada’s international obligations by ensuring that it is only in exceptional cases that a Convention refugee or a protected person will lose the benefit of non-refoulement and be removed from Canada” (R.F., at para. 125). Section 115(1) prohibits removal of certain persons to persecution, and s. 115(2) provides for limited exceptions to that prohibition:

Protection

115 (1) A protected person or a person who is recognized as a Convention refugee by another country to which the person may be returned shall not be removed from Canada to a country where they would be at risk of persecution for reasons of race, religion, nationality, membership in a particular social group or political opinion or at risk of torture or cruel and unusual treatment or punishment.

Exceptions

(2) Subsection (1) does not apply in the case of a person

(a) who is inadmissible on grounds of serious criminality and who constitutes, in the opinion of the Minister, a danger to the public in Canada; or

(b) who is inadmissible on grounds of security, violating human or international rights or organized criminality if, in the opinion of the Minister, the person should not be allowed to remain in Canada on the basis of the nature and severity of acts committed or of danger to the security of Canada.

According to the Minister, if a person inadmissible under s. 34(1)(e) is subject to an enforceable removal order, s. 115(1) and (2) will prevent their refoulement except in limited circumstances, and so s. 34(1)(e) need not be interpreted as requiring a link to national security or the security of Canada.

I disagree with the Minister’s submission for two reasons. First, while the Minister is correct that the s. 115(2) exceptions apply in limited circumstances, they nonetheless allow refoulement of persons inadmissible under s. 34(1)(e) in circumstances outside the scope of the Article 33(2) exceptions. As discussed above, Article 33(2) provides for exceptions to the non-refoulement principle for persons reasonably regarded as a danger to the security of the host country, or who are convicted of a particularly serious crime and are a danger to the community. But s. 115(2) countenances refoulement outside those circumstances: it only requires that the Minister of Citizenship and Immigration be of the opinion that the person should not be allowed to remain in Canada on the basis of (1) the nature and severity of acts committed, or (2) danger to the security of Canada. It thus permits the Minister to allow refoulement based on “the nature and severity” of acts not linked to “the security of the country”.

Second, s. 115(1) affords protection to a narrower set of persons than are protected from refoulement under the Refugee Convention. Section 115(1) only affords protection to a person recognized as having refugee status: “[a] protected person or a person who is recognized as a Convention refugee by another country to which the person may be returned”. By contrast, Article 33 affords protection to “a refugee”, and thus does not require a prior recognition of refugee status (Refugee Convention, Article 1; Refugee Protocol, Article 1; G. v. G., [2021] UKSC 9, [2022] A.C. 544, at para. 81 (“The obligation not to refoule an individual arises by virtue of the fact that their circumstances meet the definition of ‘refugee’, not by reason of the recognition by a contracting state that the definition is met.”)). As a result, if s. 34(1)(e) applies to conduct not linked to national security or the security of Canada, s. 115(1) allows the Minister of Citizenship and Immigration to subject refugee claimants to refoulement contrary to Article 33(1) of the Refugee Convention.

That neither Mr. Mason nor Mr. Dleiow is a refugee claimant does not detract from this conclusion. The Refugee Convention imposes an important legal constraint on the interpretation of s. 34(1)(e) generally, irrespective of whether the specific foreign national subject to deportation is a refugee claimant.

The Federal Court of Appeal declined to consider this argument because it had not been made to the IAD, and because it said that “certain background documents and other instruments needed to understand any international obligations are not in evidence before us” (paras. 73‑74). The court did not say what documents and instruments might be missing.

In any event, the role of the Refugee Convention in constraining the interpretation of the IRPA is a question of law, one that Parliament by s. 3(3)(f) has expressly directed a court or administrative decision maker to consider. The IAD’s failure to consider this issue did not involve the omission of a “minor aspect of the interpretive context” (Vavilov, at para. 122). Rather, it involved the omission of the principle of non-refoulement — “the cornerstone of the international refugee protection regime”, and a critical legal constraint on interpretation of the IRPA , one that Parliament has decreed must be considered in construing and applying the IRPA . This crucial omission made the IAD’s decision unreasonable.

The IAD’s reasons in Mr. Mason’s case failed to address critical points of statutory context and the broad consequences of its interpretation of s. 34(1) (e) of the IRPA , all of which Mr. Mason had highlighted in his written submissions to the IAD. These omissions were significant, not minor; they involved a failure of “responsive justification” by the IAD. The IAD also failed to interpret and apply s. 34(1)(e) in compliance with Canada’s obligation of non-refoulement under the Refugee Convention, a matter that Parliament has decreed it must consider. Cumulatively, these omissions rendered the IAD’s decision unreasonable.

The ID’s decision in Mr. Dleoiw’s case, which simply followed the IAD’s interpretation of s. 34(1) (e) of the IRPA in Mr. Mason’s case, was unreasonable for the same reasons.

Vavilov cautioned that although a court conducting a reasonableness review is not tasked with determining the “correct” interpretation of a disputed statutory provision, it may become clear in the course of conducting a judicial review that the relevant constraints bearing on the decision “so overwhelmingly” favour one interpretation that there is room for only one reasonable interpretation of the provision at issue (para. 124, citing approvingly Nova Tube, at para. 61 (CanLII), per Laskin J.A.). This Court noted that, based on the applicable constraints, a particular outcome may be “inevitable” (Vavilov, at para. 142). In such a case, while “a court should generally pause before definitively pronouncing upon the interpretation of a provision entrusted to an administrative decision maker”, it would serve no useful purpose to remit the matter to the original decision maker (para. 124). The Court reached this conclusion not as the starting point of its reasonableness review, but rather as a result of a proper reasonableness review when considering the appropriate remedy.

That is also the case here. The relevant legal constraints cumulatively point overwhelmingly to only one reasonable interpretation of s. 34(1)(e) — the provision requires a nexus to national security or the security of Canada. This interpretation is supported by the placement of s. 34(1)(e) amid other “security grounds” in s. 34, all of which have a link to national security or the security of Canada. And although the IAD noted that some interpretive constraints point in the other direction, the two critical points of statutory context that the IAD failed to consider, and especially, the legal constraints imposed by international law, overwhelmingly support the opposite conclusion: s. 34(1)(e) can be invoked to render a person inadmissible only when their “acts of violence that would or might endanger the lives or safety of persons in Canada” have a nexus with national security or the security of Canada.

As a result, the decision of the IAD in Mr. Mason’s case and of the ID in Mr. Dleiow’s case were unreasonable and should be quashed. Because the Minister has not alleged that Mr. Mason or Mr. Dleiow engaged in acts of violence with a link to national security or the security of Canada, s. 34(1)(e) of the IRPA does not provide a legal basis for the inadmissibility of either person. And because the Minister has not alleged any other basis for their inadmissibility, there is no need to remit either decision to the ID or IAD for redetermination.

I would allow the appeals, set aside the judgments of the Federal Court of Appeal and allow the applications for judicial review. In Mr. Mason’s appeal, I would quash the IAD decision thus restoring the ID decision. In Mr. Dleiow’s appeal, I would quash the ID decision and deportation order. Since neither appellant requested costs, I would make no order as to costs.

The following are the reasons delivered by

Côté J. —

I agree with my colleague’s disposition of these appeals. Inadmissibility under s. 34(1) (e) of the Immigration and Refugee Protection Act , S.C. 2001, c. 27 (“IRPA ”), requires a nexus between the relevant act of violence and with national security or the security of Canada (see paras. 11 and 121). However, I would review the Immigration Appeal Division’s (“IAD”) interpretation of s. 34(1)(e) on a standard of correctness, as submitted by the appellant Mr. Earl Mason and the interveners the Canadian Association of Refugee Lawyers and the Canadian Council for Refugees.

In Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, this Court held that reviewing courts should derogate from the presumption of reasonableness review where required by (1) a clear indication of legislative intent or (2) the rule of law (para. 10). In my view, the rule of law requires — and Parliament intended for appellate courts to provide — definitive, correct answers to legal questions certified under s. 74 (d) of the IRPA . By definition, these are questions that transcend the interests of the parties and raise issues of broad significance within Canada’s immigration and refugee protection scheme (see Lunyamila v. Canada (Public Safety and Emergency Preparedness), 2018 FCA 22, [2018] 3 F.C.R. 674, at para. 46).

In Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982, this Court noted that the certified question regime would be “incoherent” if the standard of review were anything other than correctness (para. 43). This is exemplified by the companion appeals before us. The IAD’s interpretation of s. 34(1)(e) in Mr. Mason’s case, subsequently applied to Mr. Dleiow, would significantly expand the grounds on which foreign nationals or permanent residents may be deported from Canada. It would allow foreign nationals to be returned to countries where they may face persecution, in a manner contrary to Canada’s obligations under the Convention Relating to the Status of Refugees, Can. T.S. 1969 No. 6 (see Jamal J.’s reasons, at paras. 104‑17). Parliament did not intend for appellate courts, as the Federal Court of Appeal did in this case, to defer to such interpretations where they may be “reasonable”, but are nonetheless wrong in law (see Pushpanathan, at para. 43).

To be consistent with the principles and framework set out in Vavilov, I would recognize a new category of correctness review: when appellate courts decide a “serious question of general importance” certified under s. 74 (d) of the IRPA .

Prior to Vavilov, this Court was inconsistent in its approach to certified questions. In Pushpanathan, the majority of this Court held that the certified question regime under s. 83(1) of the former Immigration Act, R.S.C. 1985, c. I‑2 — the predecessor to s. 74 (d) of the IRPA — would be “incoherent if the standard of review were anything other than correctness” (para. 43). Writing for the majority, Bastarache J. noted that the only way in which the certification procedure could be given its explicitly articulated scope was if courts were permitted to substitute their own answers for those of the Immigration and Refugee Board in respect of questions of general importance.

Under the pre‑Vavilov contextual approach, this Court departed from the approach taken in Pushpanathan in some subsequent cases, though not in most. In Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817, the majority of this Court applied the intermediate standard of review of reasonableness simpliciter, distinguishing Pushpanathan based on the “highly discretionary and fact‑based nature” of the Minister’s decision to refuse to exempt the appellant in that case from deportation on humanitarian and compassionate grounds (Baker, at paras. 61‑62). Importantly, however, the Court answered the certified question before it definitively:

Simpson J. certified the following as a “serious question of general importance” under s. 83(1) of the Immigration Act: “Given that the Immigration Act does not expressly incorporate the language of Canada’s international obligations with respect to the International Convention on the Rights of the Child, must federal immigration authorities treat the best interests of the Canadian child as a primary consideration in assessing an applicant under s. 114(2) of the Immigration Act?”

. . .

The certified question asks whether the best interests of children must be a primary consideration when assessing an applicant under s. 114(2) and the Regulations. The principles discussed above indicate that, for the exercise of the discretion to fall within the standard of reasonableness, the decision‑maker should consider children’s best interests as an important factor, give them substantial weight, and be alert, alive and sensitive to them. [Emphasis added; emphasis in original deleted; paras. 9 and 75.]

The Court also applied a standard of review of reasonableness in Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, on the basis that the Minister’s decision under the former s. 34(2) of the IRPA was discretionary (para. 50). In Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 S.C.R. 909, the majority of this Court held that the fact that the reviewing judge “considered the question to be of general importance” was “relevant, but not determinative” of the standard of review (para. 44). “Despite the presence of a certified question, the appropriate standard of review” in that case was reasonableness (ibid., citing Baker, at para. 62).

However, as the Canadian Association of Refugee Lawyers notes in its factum, Agraira and Kanthasamy are outliers. The Court applied a standard of correctness in Chieu v. Canada (Minister of Citizenship and Immigration), 2002 SCC 3, [2002] 1 S.C.R. 84, at para. 26, and Hilewitz v. Canada (Minister of Citizenship and Immigration), 2005 SCC 57, [2005] 2 S.C.R. 706, at para. 71. In other cases, the Court gave definitive answers to certified questions of interpretation either without addressing the standard of review (see Ezokola v. Canada (Citizenship and Immigration), 2013 SCC 40, [2013] 2 S.C.R. 678, at paras. 6‑9; Febles v. Canada (Citizenship and Immigration), 2014 SCC 68, [2014] 3 S.C.R. 431, at paras. 6 and 60) or after finding that it was unnecessary to resolve the issue (B010 v. Canada (Citizenship and Immigration), 2015 SCC 58, [2015] 3 S.C.R. 704, at paras. 26 and 76; Tran v. Canada (Public Safety and Emergency Preparedness), 2017 SCC 50, [2017] 2 S.C.R. 289, at paras. 23, 53 and 56).

Relying on pre‑Vavilov authorities, my colleague says that this Court has “concluded in the immigration context” that the standard of review for certified questions is reasonableness (para. 51, citing Kanthasamy and Baker). With respect, I disagree.

First, this Court did not endorse or even cite Kanthasamy in Vavilov. It relied on Baker in Vavilov, but for reasons unrelated to the determination of the standard of review.

Second, prior to Vavilov, this Court consistently provided definitive answers to certified questions of statutory interpretation (see, e.g., Pushpanathan, at paras. 75‑76; Baker, at para. 75; Chieu, at para. 90; Ezokola, at paras. 6‑9; Febles, at para. 60; Hilewitz, at para. 71; B010, at para. 76; Tran, at para. 56; see also Vavilov v. Canada (Citizenship and Immigration), 2017 FCA 132, [2018] 3 F.C.R. 75, at para. 37; Kanthasamy v. Canada (Citizenship and Immigration), 2014 FCA 113, [2015] 1 F.C.R. 335, at para. 33). In Kanthasamy, the majority of this Court arguably did the same. Indeed, prior to considering the standard of review, Abella J., who wrote the majority reasons, engaged in a lengthy interpretive exercise with respect to s. 25(1) of the IRPA (see paras. 10‑41). In dissent, Moldaver J. (Wagner J. (as he then was) concurring) lamented that the majority had adopted a “do as we say, not what we do” approach to reasonableness review:

In particular, I am concerned that my colleague has not given the Officer’s reasons the deference which, time and again, this Court has said they deserve. In her reasons, she parses the Officer’s decision for legal errors, resolves ambiguities against the Officer, and reweighs the evidence. Lest we be accused of adopting a “do as we say, not what we do” approach to reasonableness review, this approach fails to heed the admonition in Newfoundland and Labrador Nurses — that reviewing courts must be cautious about substituting their own view of the proper outcome by designating certain omissions in the reasons to be fatal (para. 17). As is the case with every other court, this Court has no licence to find an officer’s decision unreasonable simply because it considers the result unpalatable and would itself have come to a different result. [Emphasis added; para. 112.]

Following Kanthasamy, a unanimous Court held in Tran that an administrative interpretation of the “serious criminality” provisions set out in s. 36(1) (a) of the IRPA could not stand under either standard of review (para. 23). The Court did not endorse or cite Kanthasamy and, again, provided definitive answers to the two questions of general importance certified by the Federal Court:

1. Is a conditional sentence of imprisonment imposed pursuant to the regime set out in ss. 742 to 742.7 of the Criminal Code a “term of imprisonment” under s. 36(1) (a) of the IRPA ?

― No.

2. Does the phrase “punishable by a maximum term of imprisonment of at least 10 years” in s. 36(1) (a) of the IRPA refer to the maximum term of imprisonment available at the time the person was sentenced or to the maximum term of imprisonment under the law in force at the time admissibility is determined?

― It refers to the maximum term of imprisonment available at the time of the commission of the offence. [para. 56]

In sum, the weight of authority confirms that prior to Vavilov, this Court repeatedly answered certified questions definitively and without deference to administrative decision makers. I therefore disagree that, “[a]s a matter of precedent”, reasonableness review applies (Jamal J.’s reasons, at para. 51).

In any event, Vavilov overtook prior jurisprudence (para. 143; see Society of Composers, Authors and Music Publishers of Canada v. Entertainment Software Association, 2022 SCC 30, at para. 25). The jurisprudence discussed above must be analyzed in light of the principles and framework set out in Vavilov.

In Vavilov, this Court recognized five categories of correctness review, based on either legislative intent (legislated standards of review and statutory appeal mechanisms) or the rule of law (constitutional questions, general questions of law of central importance to the legal system as a whole, and questions regarding jurisdictional boundaries between administrative bodies) (paras. 17 and 69). However, it was expressly stated in Vavilov that the categories of correctness review are not closed (para. 70). And indeed, in Society of Composers, this Court recognized a new category: “. . . when courts and administrative bodies have concurrent first instance jurisdiction over a legal issue in a statute. Applying correctness to these issues accords with legislative intent and promotes the rule of law” (para. 28).

The issue in Vavilov reached the Federal Court of Appeal by way of a certified question under s. 22.2 (d) of the Citizenship Act , R.S.C. 1985, c. C‑29 . The Federal Court of Appeal provided a definitive answer to that question:

The proper stated question and my proposed answer to it are as follows:

Question: Are the words “other representative or employee [in Canada] of a foreign government” found in paragraph 3(2) (a) of the Citizenship Act limited to foreign nationals [falling within these words] who [also] benefit from diplomatic privileges and immunities?

Answer: Yes. [para. 90]

This Court upheld the Federal Court of Appeal’s decision quashing the Registrar of Citizenship’s decision to cancel Mr. Vavilov’s citizenship certificate (para. 194). The majority of the Court concluded that Mr. Vavilov’s status was governed by s. 3(1) (a) of the Citizenship Act and that he was a Canadian citizen (para. 196).

While I acknowledge that our Court applied a standard of review of reasonableness to the Registrar’s decision, I would not read Vavilov as decisive of the standard of review for certified questions under the IRPA moving forward. Let me explain.

First, this Court did not discuss the issue of certified questions in Vavilov. As noted in Society of Composers, when this Court “wanted to reject the possibility of a certain correctness category, it did so expressly” (para. 42, citing Vavilov, at paras. 71‑72). In my view, the standard of review for certified questions — certainly in the unique context of the IRPA , as explained below — remains an open question post‑Vavilov.

I would note that the Federal Court of Appeal does not view Vavilov as dispositive on this point. To the contrary, in Canada (Citizenship and Immigration) v. Galindo Camayo, 2022 FCA 50, 468 D.L.R. (4th) 358, the court noted that correctness review of certified questions would “seem to gain greater credence now that the Supreme Court has held that statutory standards can have a bearing on the standard of review” (para. 41 (emphasis added)). Writing for the court, Mactavish J.A. observed that in Vavilov itself, the Federal Court of Appeal gave a “precise answer, akin to a correctness review answer”, to the certified question before it, an approach that this Court effectively ratified in dismissing the appeal (para. 43). The Federal Court of Appeal in Galindo Camayo appears to have applied a standard of reasonableness based on Kanthasamy, not Vavilov (see para. 42).

Second, the certified question in Vavilov arose under different legislation, the Citizenship Act . In the separate and unique context of the IRPA , multiple different ministers, government departments, and agencies, as well as Canada’s largest administrative tribunal (the Immigration and Refugee Board) are charged with independently administering the statutory scheme. In many cases, these separate decision makers are required to interpret the same statutory provisions. While it is beyond the scope of these appeals to exhaustively canvass the scheme of the Citizenship Act , the statutory certification process has been widely used under the IRPA to settle divergent interpretations or disagreements on legal issues of general importance (see Huruglica v. Canada (Citizenship and Immigration), 2016 FCA 93, [2016] 4 F.C.R. 157, at para. 28).

Finally, to say that Vavilov is determinative and that the standard of review for certified questions is reasonableness would contradict the Vavilov framework itself. As I explain below, reasonableness review of certified questions under the IRPA is inconsistent with both Parliament’s intent and the rule of law.

In s. 74(d) of the IRPA , Parliament has provided for an “exceptional” appeal (see Pushpanathan, at para. 43) to the Federal Court of Appeal for legal questions certified as “serious question[s] of general importance”. This indicates legislative intent for judicial involvement and a desire to subject these particular questions, as distinct from all others arising under the IRPA more broadly, to appellate standards of review (see Society of Composers, at para. 30; Vavilov, at para. 36). Parliament’s institutional design choice should be respected by the courts.

I do not dispute that s. 74(d) provides for statutory appeals following judicial review decisions. If Parliament had provided for statutory appeals from administrative decisions directly, certified questions would fit within the existing Vavilov category. Nonetheless, as this Court noted in Society of Composers, the presumption of reasonableness review no longer applies where the legislature expressly involves the court in the administrative scheme:

Reasonableness is the standard of review that, in most instances, gives best effect to legislative intent. When the legislature has granted exclusive jurisdiction to an administrative decision maker, courts presume that the legislature wanted that decision maker to operate without undue judicial interference: Vavilov, at para. 24.

When the legislature expressly involves the court in the administrative scheme, this presumption no longer applies. That is why legislated standards of review and statutory appeal mechanisms give rise to a correctness standard of review. Such statutory features indicate legislative intent for judicial involvement and a desire to subject those decisions to appellate standards of review: Vavilov, at para. 36. [Emphasis added; paras. 29‑30.]

As my colleague acknowledges, the certified question regime under the IRPA is a “statutory mechanism” through which an appeal is provided for “in certain circumstances”, i.e. based on the nature and importance of the legal question at issue (para. 9). Through s. 74(d), Parliament does not “exclude the courts but rather makes them part of the enforcement machinery” in certain circumstances (Vavilov, at para. 36, citing Seneca College of Applied Arts and Technology v. Bhadauria, [1981] 2 S.C.R. 181, at p. 195). In my view, this strongly rebuts the presumption of reasonableness review for this particular type of legal question, in a manner analogous to the existing categories of correctness review.

Prior to Vavilov, the Federal Court of Appeal also adopted the practice of providing the “definitive answer to a certified question on a point of statutory interpretation” (see Kanthasamy, at para. 35). This is consistent with Parliament’s express intent for appellate courts to decide certified questions. In Pushpanathan, this Court discussed Parliament’s intent as follows:

First, s. 83(1) would be incoherent if the standard of review were anything other than correctness. The key to the legislative intention as to the standard of review is the use of the words “a serious question of general importance” . . . . The general importance of the question, that is, its applicability to numerous future cases, warrants the review by a court of justice. Would that review serve any purpose if the Court of Appeal were obliged to defer to incorrect decisions of the Board? Is it possible that the legislator would have provided for an exceptional appeal to the Court of Appeal on questions of “general importance”, but then required that despite the “general importance” of the question, the court accept decisions of the Board which are wrong in law, even clearly wrong in law, but not patently unreasonable? [Emphasis in original; para. 43.]

The Court then noted that, for the certified question regime to be given its explicitly legislated scope, appellate courts must be able to answer them correctly:

The only way in which s. 83(1) can be given its explicitly articulated scope is if the Court of Appeal — and inferentially, the Federal Court, Trial Division — is permitted to substitute its own opinion for that of the Board in respect of questions of general importance. [Emphasis added; para. 43.]

I could not agree more. For serious questions of general importance arising under the IRPA , Parliament did not intend courts to be forced to defer to administrative decisions that may be “reasonable”, but are wrong in law. As with s. 83(1) of the former Immigration Act, the only way s. 74 (d) of the IRPA can be given its explicitly legislated scope is if appellate courts are permitted to substitute their own opinion in respect of serious questions of general importance.

The Federal Court of Appeal has struggled with this exact issue since Vavilov. In Galindo Camayo, Mactavish J.A. lamented the “misfit between answering [a] certified question properly and conducting reasonableness review” (para. 41):

However, the fact that we have certified questions before us gives rise to an awkward situation. Certified questions generally raise questions of law, including, as in this case, questions of statutory interpretation. However, the questions, as phrased by the Federal Court, require a yes or no answer. This invites correctness review by this Court. That said, as described above, this Court is required to engage in reasonableness review on questions of statutory interpretation. This creates the possibility that, in some cases, this Court may find the [Refugee Protection Division’s] interpretation of a statutory provision to be reasonable, yet this Court may say something entirely different in providing its own view of the matter in answering the certified question — something that the Supreme Court expressly tells us not to do . . . . [Emphasis added; para. 40.]

The court’s solution to this problem in Galindo Camayo was to reformulate the certified question to ask whether a particular statutory interpretation was reasonable:

In this case, the second and third questions, as stated, call for a correctness response. I would therefore amend them to ask whether the particular statutory interpretation or approach suggested by the question is or is not reasonable. [Emphasis added; para. 44.]

This is also what the Federal Court did in Mr. Mason’s case, in an attempt to incorporate reasonableness as the applicable standard of review:

The parties, however, each proposed their version of the question referring to the “correctness” of the IAD’s interpretation. I thus rephrase the question as follows, incorporating reasonableness as the applicable standard of review:

Is it reasonable to interpret section 34(1) (e) of the Immigration and Refugee Protection Act , S.C. 2001, c. 27 , in a manner that does not require proof of conduct that has a nexus with “national security” or “the security of Canada?”

(2019 FC 1251, [2020] 2 F.C.R. 3, at para. 70)

In my view, this clearly distorts Parliament’s intent in s. 74(d). I do not accept that Parliament intended courts to reformulate certified questions, rather than answering them correctly. Where appellate courts conclude that there is only one reasonable interpretation of a disputed statutory provision, I struggle to see any meaningful difference between that interpretation and the correct interpretation of the statutory provision in question. Where there may be multiple “reasonable” interpretations, appellate courts are faced with the prospect of upholding decisions that incorrectly determine questions of law — for example, the impact of Canada’s international law obligations on the interpretation of certain sections of the IRPA (see Jamal J.’s reasons, at para. 117).

In this case, the IAD’s failure to consider this issue at all may have been unreasonable (see Jamal J.’s reasons, at para. 117). But what if the IAD had considered the issue and reached a reasonable, though incorrect, conclusion? As Bastarache J. warned in Pushpanathan, appellate courts applying reasonableness review would be forced to defer to that decision. In this case, it would result in the deportation of Mr. Mason and Mr. Dleiow from Canada. In future cases, it could result in foreign nationals being returned to countries where they may face persecution, in a manner contrary to Canada’s non‑refoulement obligations (Jamal J.’s reasons, at paras. 104 and 109).

In my view, this is untenable, and contrary to Parliament’s express intent for serious questions of general importance certified under s. 74 (d) of the IRPA to be reviewed and determined by appellate courts.

The presumption of reasonableness review must also give way to the importance of maintaining the rule of law, which requires that certified questions be answered consistently and definitively (see Society of Composers, at para. 33; Vavilov, at para. 53). I say this for two reasons.

First, the rule of law demands a “singular, determinate and final answer” (Vavilov, at para. 32) to a question certified as a serious question of general importance under the IRPA . In Lunyamila, the Federal Court of Appeal reiterated the criteria for certification under s. 74(d):

The question must be a serious question that is dispositive of the appeal, transcends the interests of the parties and raises an issue of broad significance or general importance. [Emphasis added; para. 46.]

(See also Canada (Immigration and Citizenship) v. Laing, 2021 FCA 194, at para. 11 (CanLII); Canada (Public Safety and Emergency Preparedness) v. XY, 2022 FCA 113, 89 Imm. L.R. (4th) 173, at para. 7).

A question whose answer turns on the unique facts of the case will not be certified (Lunyamila, at para. 46, citing Mudrak v. Canada (Minister of Citizenship and Immigration), 2016 FCA 178, 43 Imm. L.R. (4th) 199). By definition, then, certified questions concern issues of broad significance or general importance within Canada’s immigration and refugee protection regime. In my view, these are exactly the types of questions for which the rule of law demands consistent and definitive answers — and for which the risk of arbitrariness is unacceptable.

In Vavilov, our Court accepted that legal incoherence is antithetical to the rule of law (para. 72). While the Court rejected “persistent discord within an administrative body” as a standalone category of correctness review, this was based on the ability of a “more robust” form of reasonableness to guard against the risk of arbitrariness:

We are not persuaded that the Court should recognize a distinct correctness category for legal questions on which there is persistent discord within an administrative body. In Domtar Inc. v. Quebec (Commission d’appel en matière de lésions professionnelles), [1993] 2 S.C.R. 756, this Court held that “a lack of unanimity [within a tribunal] is the price to pay for the decision-making freedom and independence given to the members of these tribunals”: p. 800; see also Ellis‑Don Ltd. v. Ontario (Labour Relations Board), 2001 SCC 4, [2001] 1 S.C.R. 221, at para. 28. That said, we agree that the hypothetical scenario suggested by the amici curiae — in which the law’s meaning depends on the identity of the individual decision maker, thereby leading to legal incoherence — is antithetical to the rule of law. In our view, however, the more robust form of reasonableness review set out below, which accounts for the value of consistency and the threat of arbitrariness, is capable, in tandem with internal administrative processes to promote consistency and with legislative oversight (see Domtar, at p. 801), of guarding against threats to the rule of law. [Emphasis added; para. 72.]

There are two issues with even a “robust” form of reasonableness review in the context of certified questions. First, the “internal administrative processes” referenced in Vavilov, in this context, depend on appellate courts resolving disagreements on legal issues of general importance. As the Federal Court of Appeal noted in Huruglica, this process has been welcomed by the IAD and the Refugee Protection Division (“RPD”):

. . . for many years, the Federal Court resorted to the certification process under paragraph 74(d) to settle divergent interpretations or disagreements on legal issues of general importance. This Court’s providing the correct answer to certified questions appears to have been welcomed, particularly by the IAD and the RPD, who saw it as helpful in carrying out their functions. [Emphasis added; para. 28.]

In other words, the certified question process is the very way in which internal disputes are resolved within this administrative scheme, for a specific subset of legal questions. This category of questions may be defined with precision (see Society of Composers, at para. 39). As Paul Daly notes, “the unique features of [Canada’s] immigration regime could allow for correctness review where questions have been certified without having unfortunate consequences in other areas of law” (Certified Questions, References and Reasonableness: Canada (Citizenship and Immigration) v. Galindo Camayo, 2022 FCA 50, April 8, 2022 (online)). In the immigration context, the certified question procedure is “tailor‑made to achieve correctness review on questions of law” (ibid.).

Second, the risk of arbitrariness may be acceptable in the context of decisions regarding the extent of an income replacement indemnity during a temporary plant closure (as in Domtar Inc. v. Quebec (Commission d’appel en matière de lésions professionnelles), [1993] 2 S.C.R. 756) or alleged violations of a provincial collective agreement (as in Ellis‑Don Ltd. v. Ontario (Labour Relations Board), 2001 SCC 4, [2001] 1 S.C.R. 221) to use the two examples referred to in Vavilov (para. 72). It is not acceptable when the identity of the individual decision maker is what determines who is permitted to remain in Canada, as in these companion appeals, or in the context of other serious questions of general importance under the IRPA .

A number of scholars and several interveners in these appeals emphasize the fundamental importance of certified questions, the potential consequences for affected individuals, and the corresponding need for courts to provide correct and definitive answers in this context (see, e.g., J. C. Y. Liew, “The Good, the Bad, and the Ugly: A Preliminary Assessment of Whether the Vavilov Framework Adequately Addresses Concerns of Marginalized Communities in the Immigration Law Context” (2020), 98 Can. Bar Rev. 398, at p. 425; G. Heckman and A. Khoday, “Once More unto the Breach: Confronting the Standard of Review (Again) and the Imperative of Correctness Review when Interpreting the Scope of Refugee Protection” (2019), 42 Dal. L.J. 49, at pp. 62‑68 and 82; I.F., Canadian Association of Refugee Lawyers, at paras. 20‑21; I.F., Canadian Council for Refugees, at paras. 14‑20; I.F., United Nations High Commissioner for Refugees, at paras. 21‑24). I agree with my colleague that the potential consequences of a decision are relevant to the substance of reasonableness review (see para. 69). But those consequences are also relevant to the determination of the standard of review, and to the imperative of providing legally correct answers in certain circumstances.

As Professors Heckman and Khoday note, where different decision makers adopt different interpretations of the scope of refugee protection under the IRPA , one claimant may receive protection while another presenting an identical claim may be returned to their country of origin to face persecution. Given the “momentous impact” of these decisions, “the existence of divergent interpretations of these key provisions is arbitrary and antithetical to the rule of law, which demands that the inconsistencies be immediately resolved through correctness review” (p. 68).

Relatedly, correctness review serves to ease the burden on lay applicants, many of whom will be facing life‑changing immigration consequences. The reasonableness of an administrative decision depends, in large part, on the evidence before the decision maker and the submissions made by the parties (Vavilov, at para. 94; see Jamal J.’s reasons, at para. 61). Not all immigration or refugee applicants will have the resources or knowledge to make sophisticated contextual and interpretive arguments. In the context of certified questions, even a robust form of reasonableness review may be insufficient to guard against the risk, and the consequences, of arbitrariness (see Vavilov, at para. 192).

While not all certified questions concern issues of admissibility or deportation, I am satisfied that the rule of law nonetheless demands consistent and definitive answers to all questions that are properly certified. I am unable to tolerate the risk of arbitrariness in this context.

Second, questions certified under s. 74(d) will, by definition, have implications beyond the immediate parties. In addition to their potential impacts on Canada’s international obligations, they may have impacts on criminal law or on other legislation.

For example, the Citizenship Act allows the Minister to commence an “action”, i.e., in the Federal Court, to have a person declared “inadmissible on security grounds, on grounds of violating human or international rights or on grounds of organized criminality” under s. 34 , 35 , or 37 of the IRPA :

Inadmissibility

10.5 (1) On the request of the Minister of Public Safety and Emergency Preparedness, the Minister shall — in the originating document that commences an action under subsection 10.1(1) on the basis that the person obtained, retained, renounced or resumed his or her citizenship by false representation or fraud or by knowingly concealing material circumstances, with respect to a fact described in section 34 , 35 or 37 of the Immigration and Refugee Protection Act other than a fact that is also described in paragraph 36(1)(a) or (b) or (2)(a) or (b) of that Act — seek a declaration that the person who is the subject of the action is inadmissible on security grounds, on grounds of violating human or international rights or on grounds of organized criminality under, respectively, subsection 34(1) , paragraph 35(1) (a) or (b) or subsection 37(1) of the Immigration and Refugee Protection Act .

The effect of s. 10.5(1) of the Citizenship Act is therefore to create a form of shared jurisdiction between courts and administrative decision makers over inadmissibility on grounds of security (s. 34), human or international rights violations (s. 35), or organized criminality (s. 37). The implications of this were not argued by the parties in these appeals, including in light of the new category of correctness review recognized by this Court in Society of Composers. However, it is clear that a determination of inadmissibility under s. 34, 35, or 37 — as distinct from inadmissibility on grounds of criminality (s. 36) or other less serious grounds set out in ss. 38 to 41 — has effects beyond the IRPA . Reasonableness review does not adequately safeguard against the need for consistency and legal coherence in this context.

Further, as this Court held in Vavilov, certain issues require a high degree of interpretive consistency to ensure conformity with Canada’s international obligations (para. 192). While neither Mr. Mason nor Mr. Dleiow is a Convention refugee, the IAD’s interpretation of s. 34(1)(e) could foreseeably impact Canada’s future compliance with its non‑refoulement obligations (see Jamal J.’s reasons, at paras. 104‑17). My colleague views the IAD’s “fail[ure] to address the legal constraints imposed by international law” as unreasonable (para. 104). 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. Correctness review of certified questions eliminates such concerns and ensures respect for Canada’s international and treaty obligations.

The answers to certified questions under the IRPA may also have impacts on criminal law (see, e.g., Tran, at paras. 39‑42). In such situations, a lack of clarity on the scope of inadmissibility under s. 34 may prevent accused persons from making informed decisions about whether to enter into a plea agreement. As Wagner J. (as he then was) noted in R. v. Wong, 2018 SCC 25, [2018] 1 S.C.R. 696 (dissenting, but not on this point),

[c]ollateral consequences that affect the accused person’s fundamental interests could have a more significant impact on the accused than the criminal sanction itself. As a result, it may be essential for an accused to be aware of such consequences in order to enter an informed guilty plea. This is particularly true in the immigration context, in which an accused may be exposed to a collateral consequence as serious as deportation. People who are to be deported may experience any number of serious life‑changing consequences. They may be forced to leave a country they have called home for decades. They may return to a country where they no longer have any personal connections, or even speak the language, if they emigrated as children. If they have family in Canada, they and their family members face dislocation or permanent separation. [para. 72]

In sum, certified questions tend to have significant consequences for the justice system as a whole or for other institutions of government (Vavilov, at para. 59). In many cases, this will arguably place them within the existing category of general questions of law of central importance to the legal system as a whole. If not, the combined effect of their broader significance and Parliament’s intent that they be decided by appellate courts justifies a departure from the presumption of reasonableness review.

To be clear, this conclusion would not change the standard of review applicable to the vast majority of administrative decisions under the IRPA . Decisions of the Immigration and Refugee Board, as well as those made by various ministers, government departments, and agencies, would continue to be reviewed on a standard of reasonableness in most cases. It is only a small and distinct subset of legal questions — those certified as raising issues of broad significance or general importance within the statutory scheme — for which the rule of law, in addition to Parliament’s intent, mandates correctness review.

Applying a standard of review of correctness, I agree with my colleague that inadmissibility under s. 34(1)(e) requires a nexus between the relevant act of violence and with national security or the security of Canada (para. 121). Together with the reasons he identifies, I would briefly emphasize the following points of statutory context.

First, the government’s own explanation of the IRPA describes s. 34 as a “national security” provision:

This provision makes a person inadmissible to Canada for reasons of national security, including espionage, subversion, and terrorism. This provision clearly states that permanent residents and foreign nationals are inadmissible on security grounds for engaging in terrorism or for being a member of an organization that engages in terrorism. The facts that constitute inadmissibility under this provision include facts arising from omissions and those for which there are reasonable grounds to believe that they have occurred, are occurring or may occur. Other inadmissible grounds relating to security include being a danger to the security of Canada and engaging in acts of violence that would or might endanger the lives or safety of persons in Canada. [Emphasis added.]

(Citizenship and Immigration Canada, Bill C‑11: Clause by Clause Analysis (September 2001), at pp. 31‑32)

Second, the serious nature of inadmissibility on security grounds (see Jamal J.’s reasons, at paras. 86‑97) is reinforced by several additional distinctions in the statutory scheme. Under s. 36(1), both foreign nationals and permanent residents may be inadmissible on grounds of serious criminality. However, only foreign nationals may be inadmissible on grounds of criminality (s. 36(2)). The difference between “criminality” and “serious criminality”, for offences committed in Canada, is that serious criminality requires a conviction for an offence punishable by a maximum term of imprisonment of at least 10 years or for an offence for which a sentence of more than 6 months’ imprisonment has been imposed (s. 36(1)(a)). A permanent resident convicted of an offence falling short of these requirements cannot be found inadmissible under s. 36(2).

The IAD’s interpretation of s. 34(1)(e) eliminates this carefully legislated distinction. A permanent resident who is charged with an act of violence that would not qualify as serious criminality under s. 36(1) could instead be found inadmissible under s. 34(1)(e), even if the alleged act did not result in a conviction.

Further, s. 64(1) of the IRPA denies a right of appeal to permanent residents or foreign nationals found inadmissible on grounds of security under s. 34, among other serious categories of inadmissibility:

No appeal for inadmissibility

64 (1) No appeal may be made to the Immigration Appeal Division by a foreign national or their sponsor or by a permanent resident if the foreign national or permanent resident has been found to be inadmissible on grounds of security [(s. 34)], violating human or international rights [(s. 35)], serious criminality [(s. 36(1))] or organized criminality [(s. 37)].

Thus, a foreign national convicted of assault, an act of violence, and found inadmissible by the Immigration Division (“ID”) on grounds of criminality under s. 36(2) may appeal that finding to the IAD. If the same conduct could ground inadmissibility under s. 34(1)(e) — as the IAD held in this case — the foreign national would lose the ability to appeal. That person would be better off being convicted, and found inadmissible under s. 36(2), than if their charges were dropped or stayed and they were instead found inadmissible under s. 34(1)(e).

Clearly, there is potential overlap between “acts of violence” contemplated in s. 34(1)(e) and “criminality” addressed in s. 36. The implication of the respective treatment of these sections throughout the IRPA is that inadmissibility on security grounds, under s. 34, is more serious than inadmissibility on grounds of criminality. As Grammond J. noted in Mr. Mason’s case, the IAD’s interpretation of s. 34(1)(e) “brings under the most severe category of inadmissibility a vast range of conduct that includes acts that are below the thresholds set by section 36” (F.C. reasons, Mason, at para. 50). Given the careful wording of s. 36, this cannot have been Parliament’s intention.

Third, I would return to s. 10.5(1) of the Citizenship Act , which also distinguishes between facts described in ss. 34, 35, and 37 of the IRPA and those described in s. 36. This, too, reinforces the fact that inadmissibility under s. 34 is considered among the gravest forms of inadmissibility and that the section should be interpreted as applying only to acts of violence with a nexus to national security.

This conclusion is consistent with the only prior interpretations of s. 34(1)(e) and its predecessor, s. 19(1)(g) of the Immigration Act. In X (Re), 2017 CanLII 146735 (I.R.B. (Imm. Div.)), Member King held that a series of common assaults could not ground inadmissibility under s. 34(1)(e):

I conclude that paragraph 34(1)(e) cannot be interpreted to include the type of one‑on‑one violent acts that exist in this case. While assaults against individuals are undesirable, they cannot be considered to be a threat to the safety of persons in Canada and the security of Canadian society, as contemplated by this section of the IRPA . [para. 42]

Member King also distinguished the circumstances in X (Re) from those before the Federal Court in Moumdjian v. Canada (Security Intelligence Review Committee), [1999] 4 F.C. 624 (C.A.), which were “more obviously related to the security of Canada” and dealt with a conspiracy to assassinate a Turkish diplomat in Canada (paras. 77‑78; see also Al Yamani v. Canada (Solicitor General), [1996] 1 F.C. 174 (T.D.)).

For these reasons, in addition to those identified by my colleague and by Grammond J. in the Federal Court, I would conclude that inadmissibility under s. 34(1)(e) requires a nexus between the relevant act of violence and with national security or the security of Canada. However, it remains the task of administrative decision makers under the IRPA to apply this interpretation going forward, including determining which acts of violence may indeed qualify as a threat to national security or the security of Canada.

This Court has made it clear that the focus of reasonableness review is on “the decision the administrative decision maker actually made”, not on the conclusion the court would have reached in the decision maker’s place (Vavilov, at para. 15; see Jamal J.’s reasons, at para. 8). Yet in the context of certified questions under the IRPA , this Court has repeatedly provided definitive, correct answers to disputed questions of statutory interpretation, including when applying a reasonableness standard of review.

My colleague does so again in this case. While I agree that the IAD’s interpretation of s. 34(1)(e) was unreasonable, I do not accept that Parliament intended for courts to defer to reasonable but legally incorrect answers to this or other certified questions. Invariably, these questions transcend the interests of the parties and raise issues of broad significance and general importance within Canada’s immigration and refugee protection regime. These are the exact types of questions for which the rule of law demands — and Parliament expressly intended for appellate courts to provide — correct answers (see Vavilov, at paras. 10 and 69‑70). As required by the principles and framework set out in Vavilov, I would recognize certified questions under the IRPA as a new category of correctness review moving forward.

In the result, I agree with my colleague’s disposition (para. 123). I would allow the appeals, set aside the decisions of the Federal Court of Appeal and allow the applications for judicial review. In Mr. Mason’s appeal, I would quash the IAD decision thus restoring the ID decision. In Mr. Dleiow’s appeal, I would quash the ID decision and deportation order.