2021 FCA 156
A-415-19
The Minister of Citizenship and Immigration (Appellant)
v.
Earl Mason (Respondent)
A-37-20
The Minister of Citizenship and Immigration (Appellant)
v.
Seifeslam Dleiow (Respondent)
Indexed as: Canada (Citizenship and Immigration) v. Mason
Federal Court of Appeal, Stratas, Rennie and Mactavish JJ.A.—By videoconference, June 17; Ottawa, July 29, 2021.
Citizenship and Immigration — Exclusion and Removal — Inadmissible Persons — Appeals from Federal Court decisions allowing applications for judicial review of decisions rendered by Immigration and Refugee Board Immigration Appeal Division (IAD), Immigration Division (ID )— In these administrative decisions, ID, IAD had to interpret Immigration and Refugee Protection Act, s. 34(1)(e), which provides that permanent residents or foreign nationals are “inadmissible on security grounds” for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada” — More specifically, ID, IAD had to determine whether this provision applying only where there is connection to national security — Respondents (Mason, Dleiow) respectively said it did while in both cases appellant said it did not — Both ID, IAD agreed with appellant, stated that s. 34(1)(e) operates whether or not there is connection to national security — Respondents sought judicial review of these administrative decisions, submitting that decisions unreasonable, should be quashed — Federal Court agreed, quashed decisions — Whether Federal Court’s decisions reasonable — IAD in case involving respondent Mason concluded that inadmissibility under Act, s. 34(1)(e) not requiring that conduct have link to national security or security of Canada, that Parliament intended that provisions of s. 34(1) relate to security in broader sense — In reaching its conclusion, IAD was very much alive to essential elements of text, context, purpose of s. 34(1)(e), analyzing most important elements of each — IAD also examined context of other inadmissibility provisions in Act; examined in particular s. 36, which provides for inadmissibility upon conviction of criminal offence — In its view, context must come not just from immediate co-text but from overall scheme, object of Act — This non-blinkered approach to issues of context, IAD’s application of it to this case could not be said to be unreasonable — As for Federal Court, it wrongly concluded that IAD’s decision in Mason was unreasonable because it failed to appreciate context of other inadmissibility provisions in Act — What Federal Court did was akin to correctness review; erred in law — ID decision about respondent Dleiow following IAD decision in Mason — ID’s decision to follow IAD’s decision in Mason on meaning of Act, s. 34(1)(e) was reasonable — This accorded with case law as it existed at that time, as it exists today — Federal Court heard judicial review in Dleiow after having decided Mason — Federal Court quashed ID decision in Dleiow because of view having to follow Mason — Since Federal Court’s judgment in Mason had to be set aside, Federal Court`s judgment in Dleiow also having to be set aside — Appeals allowed.
Administrative Law — Judicial Review — Federal Court allowing applications for judicial review of decisions rendered by Immigration and Refugee Board Immigration Appeal Division (IAD), Immigration Division (ID) — In these administrative decisions, ID, IAD had to interpret Immigration and Refugee Protection Act, s. 34(1)(e) regarding inadmissibility on security grounds — More specifically, ID, IAD had to determine whether this provision applying only where there is connection to national security — Respondents (Mason, Dleiow) respectively said it did while in both cases appellant said it did not — Both ID, IAD agreed with appellant, stated that s. 34(1)(e) operates whether or not there is connection to national security — Respondents sought judicial review of these administrative decisions, submitted that decisions unreasonable, should be quashed — Federal Court agreed, quashed decisions — Question of how reviewing court goes about reasonableness review of administrators’ interpretations of legislative provisions squarely arising here — Jurisprudential stakes high: if reviewing courts not doing reasonableness review of administrators’ interpretations properly, they can inadvertently slip into correctness review, be reversed on appeal —Hillier v. Canada (Attorney General) tells reviewing courts to conduct themselves in way that gives administrators space legislator intended them to have, yet still hold them accountable — Although Federal Court in its decision involving respondent Mason was bound by decision in Hillier, did not cite or follow it — Rather, Federal Court struck out on its own, developing, applying its own approach to review of administrative interpretations of legislative provisions — In Mason, Federal Court said that reviewing court must accept administrator’s interpretation unless it can find some sort of “knock-out punch” that defeats it — This was problematic for two reasons: first, that one-size-fits-all approach disregarded particular context in which administrators operate, wording of their particular legislation; second, while Federal Court stated that reviewing court should not re-weigh competing factors that were before administrator, its proposed methodology caused it to do exactly that — It identified several provisions that it considered to be relevant to context of Act, s. 34(1)(e), analyzed that context for itself, applied this analysis as yardstick against IAD’s interpretation, then, measuring that interpretation with exactitude, found it wanting, quashed it.
These were two appeals from Federal Court decisions allowing applications for judicial review of decisions rendered by the Immigration Appeal Division (IAD) and the Immigration Division (ID) of the Immigration and Refugee Board. In these administrative decisions, the ID and the IAD had to interpret paragraph 34(1)(e) of the Immigration and Refugee Protection Act, which provides that permanent residents or foreign nationals are “inadmissible on security grounds” for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada”. More specifically, they had to determine whether this provision applies only where there is a connection to national security. Before the IAD and the ID, the respondents (Mason and Dleiow) respectively said that it did while in both cases the appellant said that it did not. Both the ID and IAD agreed with the appellant and stated that paragraph 34(1)(e) operates whether or not there is a connection to national security. The respondents sought judicial review of these administrative decisions, submitting that they were unreasonable and should be quashed. The Federal Court agreed and quashed the decisions.
The issue was whether the Federal Court’s decisions were reasonable.
Held, the appeals should be allowed.
The question of how a reviewing court goes about reasonableness review of administrators’ interpretations of legislative provisions squarely arose here: the interpretation of paragraph 34(1)(e) of the Act is somewhat complicated and the outcome was open to some debate. The starting point was the Supreme Court of Canada’s decision in Canada (Minister of Citizenship and Immigration) v. Vavilov, which had much to say but left some things unclear. And the jurisprudential stakes are high: if reviewing courts do not do reasonableness review of administrators’ interpretations properly, they can inadvertently slip into correctness review and be reversed on appeal. Fortunately, when reviewing administrators’ interpretations of legislation, reviewing courts need not slip from reasonableness into correctness. They can avoid that danger. Hillier v. Canada (Attorney General) tells reviewing courts to conduct themselves in a way that gives administrators the space the legislator intends them to have, yet still hold them accountable. Reviewing courts can do this by conducting 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. But the lay of the land is as far as they should go. They should not make any definitive judgments and conclusions themselves. Although the Federal Court in its decision involving the respondent Mason was bound by Hillier, it did not cite or follow it. Rather, it struck out on its own, developing and applying its own approach to the review of administrative interpretations of legislative provisions. In Mason, the Federal Court said that a reviewing court must accept an administrator’s interpretation unless it can find some sort of “knock-out punch” that defeats it. This was problematic for two reasons. First, it was a one-size-fits-all approach that disregarded the particular context in which administrators operate and the wording of their particular legislation. A default position of acceptance is not always appropriate. Context matters. Second, the Federal Court’s approach in Mason lures a reviewing court into correctness review in another way. While the Federal Court stated that a reviewing court should not re-weigh competing factors that were before an administrator, its proposed methodology caused it to do exactly that. In determining whether there is a “knock-out punch”, the reviewing court must conduct its own analysis of text, context and purpose and then examine whether the administrator has rebutted the reviewing court’s interpretation with sufficiently strong “counterpunches”. This is nothing more than the reviewing court fashioning its own yardstick to measure the administrator’s interpretation and interfering if the difference is too much. The Federal Court in Mason erred in that very way. It identified several provisions that it considered to be relevant to the context of paragraph 34(1)(e), analyzed that context for itself, applied this analysis as a yardstick against the IAD’s interpretation and then, measuring that interpretation with exactitude, found it wanting and quashed it.
The IAD in Mason concluded that inadmissibility under paragraph 34(1)(e) of the Act does not require that the conduct have a link to national security or the security of Canada and that Parliament intended that the provisions of subsection 34(1) relate to security in a broader sense. In reaching its conclusion, the IAD was very much alive to the essential elements of the text, context and purpose of paragraph 34(1)(e), analyzing the most important elements of each. A reasoned explanation could be discerned from the reasons it gave and from some matters that could be implied. Given what was in the record and given what was argued, there were no omitted aspects that would cause a loss of confidence in the outcome reached by the decision maker. The IAD also examined the context of the other inadmissibility provisions in the Act. In particular, it examined section 36 of the Act, which provides for inadmissibility upon conviction of a criminal offence. It did not ignore the contextual argument of the respondent Mason. It refused to limit its analysis to just the other paragraphs of subsection 34(1). In its view, context must come not just from the immediate co-text but from the overall scheme and object of the Act. And the overall scheme of the relevant inadmissibility provisions was particularly relevant. This non-blinkered approach to issues of context and the IAD’s application of it to this case could not be said to be unreasonable. The IAD also reasoned that the mere fact some criminality could fall under paragraph 34(1)(e) did not require it to have a connection to national security. It noted that while there is some overlap, the two provisions do not cover the exact same types of behaviour. It found that the conduct captured by paragraph 34(1)(e), which speaks of the danger posed to the “lives and safety” of persons in Canada, is only a small subset of what would be considered serious criminality in section 36. And it found that sections 34 and 36 deal with two different matters: conduct and convictions respectively. These were plausible conclusions consistent with reasonableness. To the extent some possible elements of context did not appear in the IAD’s reasons in Mason, this did not cause loss of confidence in the outcome reached by the decision maker. The IAD touched upon only the most salient aspects of the text, context or purpose and was not required to deal with all aspects. To the extent that the IAD failed to mention some elements in its analysis of text, context and purpose, this was not a fundamental gap. In these circumstances, the IAD implicitly found that the preponderance of elements supported the appellant’s interpretation.
The Federal Court wrongly concluded that the IAD’s decision in Mason was unreasonable because it failed to appreciate the context of other inadmissibility provisions in the Act. What it did was akin to correctness review and, in some respects, it erred in law. The Federal Court found that one element of context outweighed all others. It looked at the consequences that flow from the Act’s sections 34 and 36 inadmissibility findings and concluded that, because somewhat harsher consequences flow from section 34 than section 36, paragraph 34(1)(e) must have a connection to national security. The Federal Court concluded that paragraph 34(1)(e)—without a connection to national security—would just be section 36 “lite” and would “render section 36 meaningless”. But this conclusion was based on the Federal Court’s incorrect assumption that the behaviour captured by paragraph 34(1)(e), “acts of violence”, and section 36, “offence[s] under an Act of Parliament”, is essentially the same. It is not. Section 36 applies to much more, including much non-violent, criminal behaviour. The Federal Court erred in other ways such as when it imposed its view over that of the IAD and when it gave little to no weight to the presumption against redundancy.
The ID’s decision involving the respondent Dleiow (Dleiow v. Canada (Public Safety and Emergency Preparedness) followed the IAD’s decision in Mason. Its decision came before the Federal Court judgment in Mason. The ID’s decision to do so was reasonable because, at that time, Mason could be said to be the most recent and persuasive authority on the meaning of paragraph 34(1)(e) of the Act. The ID considered the respondent Dleiow’s submission that the IAD wrongly decided Mason. In doing so, it did not regard the IAD’s decision in Mason or any other IAD decisions as binding on it. This accorded with the case law as it existed at that time and as it exists today. The Federal Court heard the application for judicial review in Dleiow after the Federal Court had decided Mason. It quashed the ID’s decision in Dleiow only because it was of the view that it had to follow the Federal Court’s judgment in Mason. Since the Federal Court’s judgment in Mason had to be set aside, the Federal Court’s judgement in Dleiow also had to be set aside.
STATUTES AND REGULATIONS CITED
Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44], s. 11(d).
Criminal Code, R.S.C., 1985, c. C-46, ss. 151–153, 155, 160, 163.1, 172.2, 334(a), 382, 382.1, 400.
Federal Courts Act, R.S.C., 1985, c. F-7, s. 18.3(1).
Immigration and Refugee Protection Act, S.C. 2001, c. 27, ss. 3, 34, 36, 153.
TREATIES AND OTHER INSTRUMENTS CITED
Protocol Relating to the Status of Refugees, 606 U.N.T.S. 267.
United Nations Convention Relating to the Status of Refugees, July 28, 1951, [1969] Can. T.S. No. 6.
CASES CITED
APPLIED:
Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, 441 D.L.R. (4th) 1, affg 2017 FCA 132, [2018] 3 F.C.R. 75; Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190; Hillier v. Canada (Attorney General), 2019 FCA 44, 431 D.L.R. (4th) 556, [2019] 2 F.C.R. D-3.
CONSIDERED:
Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 S.C.R. 909; Komolafe v. Canada (Citizenship and Immigration), 2013 FC 431, 16 Imm. L.R. (4th) 267; Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559.
REFERRED TO:
Entertainment Software Association v. Society of Composers, Authors and Music Publishers of Canada, 2020 FCA 100, [2021] 1 F.C.R. 374; Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27, (1998), 154 D.L.R. (4th) 193; Bell ExpressVu Limited Partnership v. Rex, 2002 SCC 42, [2002] 2 S.C.R. 559; Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54, [2005] 2 S.C.R. 601; Delios v. Canada (Attorney General), 2015 FCA 117, 472 N.R. 171; McLean v. British Columbia (Securities Commission), 2013 SCC 67, [2013] 3 S.C.R. 895; Bell Canada v. British Columbia Broadband Association, 2020 FCA 140, [2021] 3 F.C.R. 206; Chieu v. Canada (Minister of Citizenship and Immigration), 2002 SCC 3, [2002] 1 S.C.R. 84; Canadian Copyright Licensing Agency (Access Copyright) v. Canada, 2018 FCA 58, 422 D.L.R. (4th) 112, [2018] 4 F.C.R. D-3; Canadian Union of Public Employees, Local 301 v. Montreal (City), [1997] 1 S.C.R. 793, (1997), 144 D.L.R. (4th) 577; Construction Labour Relations v. Driver Iron Inc., 2012 SCC 65, [2012] 3 S.C.R. 405; Williams v. Canada (Public Safety and Emergency Preparedness), 2017 FCA 252, [2018] 4 F.C.R. 174; Canada v. Cheema, 2018 FCA 45, [2018] 4 F.C.R. 328; Canada (Attorney General) v. Utah, 2020 FCA 224, 455 D.L.R. (4th) 714; Suresh v. Canada (Minister of Citizenship and Immigration), 2002 SCC 1, [2002] 1 S.C.R. 3; El Werfalli v. Canada (Public Safety and Emergency Preparedness), 2013 FC 612, [2014] 4 F.C.R. 673; Fuentes v. Canada (Minister of Citizenship and Immigration), 2003 FCT 379, [2003] 4 F.C. 249; Canada (Attorney General) v. Distribution G.V.A. Inc., 2018 FCA 146; Canada (Attorney General) v. Bri-Chem Supply Ltd., 2016 FCA 257, [2017] 3 F.C.R. 123; Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association, 2011 SCC 61, [2011] 3 S.C.R. 654; ‘Namgis First Nation v. Canada (Fisheries and Oceans), 2019 FCA 149; Forest Ethics Advocacy Association v. Canada (National Energy Board), 2014 FCA 245, [2015] 4 F.C.R. 75; Association of Universities and Colleges of Canada v. Canadian Copyright Licensing Agency (Access Copyright), 2012 FCA 22, 428 N.R. 297; Bernard v. Canada (Revenue Agency), 2015 FCA 263, 9 Admin. L.R. (6th) 296; Bell Canada v. 7262591 Canada Ltd. (Gusto TV), 2016 FCA 123, 17 Admin. L.R. (6th) 175; Wilson v. Atomic Energy of Canada Limited, 2015 FCA 17, [2015] 4 F.C.R. 467.
AUTHORS CITED
Daly, Paul. “Waiting for Godot: Canadian Administrative Law in 2019”, 2019 CanLIIDocs 4436, online: ˂https://canlii.ca/t/t23p˃.
APPEALS from Federal Court decisions (Mason v. Canada (Citizenship and Immigration), 2019 FC 1251, [2020] 2 F.C.R. 3; Dleiow v. Canada (Citizenship and Immigration), 2020 FC 59) allowing applications for judicial review of decisions rendered by the Immigration Appeal Division (Mason v. Canada (Public Safety and Emergency Preparedness), 2019 CanLII 55171 (I.R.B.)) and the Immigration Division (Dleiow v Canada (Public Safety and Emergency Preparedness), 2019 CanLII 129531 (I.R.B.)) wherein the respondents were found to be inadmissible pursuant to paragraph 34(1)(e) of the Immigration and Refugee Protection Act based on the interpretation that this paragraph operates whether or not there is a connection to national security. Appeals allowed.
APPEARANCES
Helen Park, Tasneem Karbani and Ezra Park for appellant.
Erica Olmstead and Molly Joeck for respondent Earl Mason
Robert J. Kincaid for respondent Seifeslam Dleiow.
SOLICITORS OF RECORD
Deputy Attorney General of Canada for appellant.
Edelmann & Company, Vancouver, for respondent Earl Mason.
Robert J. Kincaid Law Corp., Vancouver, for respondent Seifeslam Dleiow.
The following are the reasons for judgment rendered in English by
[1] Stratas J.A.: Did the Immigration Appeal Division and the Immigration Division of the Immigration and Refugee Board of Canada adopt a reasonable interpretation of paragraph 34(1)(e) of the Immigration and Refugee Protection Act, S.C. 2001, c. 27? That is the common issue in these appeals.
[2] Paragraph 34(1)(e) provides that permanent residents or foreign nationals are “inadmissible on security grounds” for “engaging in acts of violence that would or might endanger the lives or safety of persons in Canada”.
[3] Does this provision apply only where there is a connection to national security? Before the Immigration Appeal Division and the Immigration Division, Mr. Mason and Mr. Dleiow, respectively, said yes. In both cases, the Minister said no.
[4] The Immigration Appeal Division and the Immigration Division agreed with the Minister. In their view, paragraph 34(1)(e) operates whether or not there is a connection to national security.
[5] Messrs. Mason and Dleiow sought judicial review of these administrative decisions. They submitted that they were unreasonable and should be quashed. The Federal Court agreed and quashed the decisions: 2019 FC 1251, [2020] 2 F.C.R. 3 (per Grammond J. in Mason [Mason v. Canada (Citizenship and Immigration)]) and 2020 FC 59 (per Barnes J. in Dleiow [Dleiow v. Canada (Citizenship and Immigration)], following Mason for reasons of comity). In both cases, the Federal Court stated a certified question for the consideration of this Court. The Minister now appeals.
[6] I would allow the appeals, set aside the judgments of the Federal Court and dismiss the applications for judicial review. The administrative decisions—in particular, their interpretation of paragraph 34(1)(e)—are reasonable.
[7] I direct that these reasons be filed in A-415-19 and a copy be filed in A-37-20.
A. Reasonableness review of administrators’ legislative interpretations
(1) The proper methodology
[8] How should a reviewing court go about reasonableness review of administrators’ interpretations of legislative provisions? The question squarely arises here: the interpretation of paragraph 34(1)(e) is somewhat complicated and the outcome is open to some debate.
[9] Of course, the starting point is Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653, 441 D.L.R. (4th) 1. Vavilov tells us much but it leaves some things unclear. And the jurisprudential stakes are high: if reviewing courts do not do reasonableness review of administrators’ interpretations properly, they can inadvertently slip into correctness review and be reversed on appeal.
[10] For over 13 years now, since Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190, Canadian courts have been reviewing the reasonableness of administrators’ interpretations of legislative provisions. In so doing, many have been affording administrators a margin of appreciation, especially where the provision admits of ambiguity. This remains the case today: see the summary in Entertainment Software Association v. Society of Composers, Authors and Music Publishers of Canada, 2020 FCA 100, [2021] 1 F.C.R. 374, at paragraphs 31–33.
[11] However, reviewing courts have not always found this easy. After all, both reviewing courts and administrators interpret legislative provisions using the same methodology, an analysis of their text, context and purpose: Vavilov, at paragraphs 117–118; on this, see also Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27, (1998), 154 D.L.R. (4th) 193; Bell ExpressVu Limited Partnership v. Rex, 2002 SCC 42, [2002] 2 S.C.R. 559; Canada Trustco Mortgage Co. v. Canada, 2005 SCC 54, [2005] 2 S.C.R. 601.
[12] This creates a danger: a reviewing court, thinking it is doing the same thing as an administrator, might be tempted to interpret legislative provisions itself and then apply its interpretation to see if the administrator has gotten it right. But that would not be reasonableness review. That would be nothing more than the reviewing court fashioning its own yardstick and then using it to measure what the administrator did: Delios v. Canada (Attorney General), 2015 FCA 117, 472 N.R. 171, at paragraph 28. That would be correctness review.
[13] Many courts have been vulnerable to this danger, even the Supreme Court itself. In the very decision that required reasonableness review for administrators’ interpretations, the Supreme Court told us that reasonableness review is deferential review: Dunsmuir, at paragraph 47. However, in case after case after Dunsmuir, it simply interpreted and applied legislation itself with no deference at all to administrators’ interpretations.
[14] In immigration cases, like the ones now before us, this mismatch was particularly frequent: see cases cited in Vavilov v. Canada (Citizenship and Immigration), 2017 FCA 132, [2018] 3 F.C.R. 75, at paragraph 37. Some members of the Supreme Court even started to worry that lower courts were being told to “do as we say, not what we do”: Kanthasamy v. Canada (Citizenship and Immigration), 2015 SCC 61, [2015] 3 S.C.R. 909, at paragraph 112.
[15] Fortunately, when reviewing administrators’ interpretations of legislation, reviewing courts need not slip from reasonableness into correctness. They can avoid that danger: Hillier v. Canada (Attorney General), 2019 FCA 44, 431 D.L.R. (4th) 556, [2019] 2 F.C.R. D-3, at paragraphs 13–17.
[16] Hillier begins by reminding reviewing courts of three basic things they should appreciate when conducting reasonableness review. First, in many cases, administrators may have a range of interpretations of legislation open to them based on the text, context and purpose of the legislation. Second, in particular cases, administrators may have a better appreciation of that range than courts because of their specialization and expertise. And, third, the legislation—the law on the books that reviewing courts must follow—gives administrators the responsibility to interpret the legislation, not reviewing courts.
[17] For these reasons, Hillier tells reviewing courts to conduct themselves in a way that gives administrators the space the legislator intends them to have, yet still hold them accountable. Reviewing courts can do this by conducting 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. But the lay of the land is as far as they should go. They should not make any definitive judgments and conclusions themselves. That would take them down the road of creating their own yardstick and measuring the administrator’s interpretation to make sure it fits.
[18] Instead, Hillier recommends (at paragraph 16) that a reviewing court should “focus on the administrator’s interpretation, noting what the administrator invokes in support of it and what the parties raise for or against it”, trying to understand where the administrator was coming from and why it ruled the way it did: Hillier, at paragraph 16.
[19] Under this approach, the reviewing court does not act in an “external” way, i.e., “arrive at a definitive conclusion about the best way to read the statutory provision under review before considering how the [administrator’s] interpretation matched up with [the] preferred reading”. Rather, as Professor Daly has observed, the reviewing court acts in an “internal” way, i.e., “a relatively cursory examination of the provision at issue, with a view to analyzing the robustness of the [administrator’s] interpretation”. See Paul Daly, “Waiting for Godot: Canadian Administrative Law in 2019” [2019 CanLIIDocs 4436] (online: https://canlii.ca/t/t23p, at page 11).
[20] By necessary implication, Vavilov supports the Hillier approach. Vavilov warns us that even though reviewing courts are accustomed in other contexts to interpret legislative provisions themselves, when conducting reasonableness review of administrative interpretations they should avoid that. Reviewing courts must not “ask how they themselves would have resolved [the] issue”, “undertake a de novo analysis”, “ask itself what the correct decision would have been” or “[decide] the issue themselves”: Vavilov, at paragraphs 75, 83 and 116. In other words, reviewing courts must not “make [their] own yardstick and then use that yardstick to measure what the administrator did”: Vavilov, at paragraph 83, citing Delios, at paragraph 28. Instead, reviewing courts must exercise “judicial restraint” and respect “the distinct role of administrative decision makers”: Vavilov, at paragraph 75. They are to do this by examining the administrator’s reasons with “ʻrespectful attention’” and by “seeking to understand the reasoning process”: Vavilov, at paragraph 84.
(2) The methodology followed by the Federal Court in Mason
[21] Although the Federal Court in Mason was bound by Hillier, it did not cite or follow it. Rather, it struck out on its own, developing and applying its own approach to the review of administrative interpretations of legislative provisions.
[22] The Federal Court in Mason said that a reviewing court must accept an administrator’s interpretation unless it can find some sort of “knock-out punch” that defeats it. This is problematic for two reasons.
[23] First, it is a one-size-fits-all approach that disregards the particular context in which administrators operate and the wording of their particular legislation. A default position of acceptance is not always appropriate. As this Court emphasized in many pre-Vavilov cases, as the Supreme Court occasionally noted in many pre-Vavilov cases, and as the Supreme Court later emphasized in Vavilov, context matters. For example, in some contexts, particularly where the wording of legislation is narrow and clear, the administrator may be highly constrained in the interpretations it can reach: see, e.g., McLean v. British Columbia (Securities Commission), 2013 SCC 67, [2013] 3 S.C.R. 895.
[24] Second, the approach of the Federal Court in Mason lures a reviewing court into correctness review in another way. While the Federal Court says (at paragraph 25) that a reviewing court should not re-weigh competing factors that were before an administrator, its proposed methodology causes it to do exactly that. In determining whether there is a “knock-out punch”, the reviewing court must conduct its own analysis of text, context and purpose and then examine whether the administrator has rebutted the reviewing court’s interpretation with sufficiently strong “counterpunches”. This is nothing more than the reviewing court fashioning its own yardstick to measure the administrator’s interpretation and interfering if the difference is too much.
[25] As we shall see, the Federal Court in Mason erred in that very way. It identified several provisions that it considered to be relevant to the context of paragraph 34(1)(e), analyzed that context for itself, applied this analysis as a yardstick against the Immigration Appeal Division’s interpretation and then, measuring that interpretation with exactitude, found it wanting and quashed it.
(3) More on reasonableness review: Vavilov’s guidance
[26] The Supreme Court in Vavilov confirmed that reasonableness review has two components: an assessment of the “outcome” and the “reasoning process”: Vavilov, at paragraph 83.
[27] In these appeals, Messrs. Mason and Dleiow submit that the Immigration Appeal Division and the Immigration Division, respectively, fell well short on both of these components of reasonableness. They attack the administrative interpretations of paragraph 34(1)(e) as unacceptable and indefensible and, thus, unreasonable. They also attack the administrators’ reasoning as incomplete and deficient and, thus unreasonable.
(a) The outcome reached by the administrator
[28] The outcome reached by an administrator must be within the constraints imposed by matters such as the legislative wording (including the nature of the decision maker and the decision), the evidence adduced and the submissions of the parties.
[29] The majority of the Supreme Court has described in considerable detail how the various contextual factors in Vavilov, acting as hard or soft constraints, can affect the ambit of decision-making that will pass muster under reasonableness review. This Court has offered further guidance in Entertainment Software Association, at paragraphs 26–36. Not much more need be said.
(b) The reasoning of the administrator
[30] The majority of the Supreme Court in Vavilov also describes this component of reasonableness in much detail. However, it breaks up the detail into a number of pieces sprinkled throughout the reasons. Collecting the pieces and consolidating them advances clarity.
[31] Vavilov tells us that a reviewing court conducting reasonableness review of an administrative decision must investigate whether a reasoned explanation can be discerned. That explanation can be expressly or impliedly in the administrator’s reasons but, as we shall see, it can also be outside the reasons.
[32] Reasons of administrators are to be “read holistically and contextually” in “light of the record and with due sensitivity to the administrative regime in which they were given”: Vavilov, at paragraphs 97 and 103. But the basis for a decision may also be implied from the circumstances, including the record, previous decisions of the administrator and related administrators, the nature of the issue before the administrator and the submissions made: Vavilov, at paragraphs 94 and 123; and see, e.g., Bell Canada v. British Columbia Broadband Association, 2020 FCA 140, [2021] 3 F.C.R. 206. For this reason, the failure of the reasons to mention something explicitly is not necessarily a failure of “justification, intelligibility or transparency”: Vavilov, at paragraphs 94 and 122. In reviewing administrators’ reasons, a reviewing court is allowed to “connect the dots on the page where the lines, and the direction they are headed, may be readily drawn”: Komolafe v. Canada (Citizenship and Immigration), 2013 FC 431, 16 Imm. L.R. (4th) 267, at paragraph 11; Vavilov, at paragraph 97.
[33] From express or implied reasons, the reviewing court must be able to discern an “internally coherent and rational chain of analysis” that the “reviewing court must be able to trace” and must be able to understand on “critical point[s]”: Vavilov, at paragraphs 85 and 102–103. The reasoning must be “rational and logical” without “fatal flaws in its overarching logic”: Vavilov, at paragraph 102.
[34] One consideration in assessing whether a reasoned explanation has been given is whether the reasons are “responsive” to the submissions made by the parties in the sense that they “meaningfully account for the central issues and concerns raised by the parties” or “meaningfully grapple with key issues or central arguments raised by the parties”, i.e. to “assur[e] [the] parties that their concerns have been heard”, demonstrate that they “have actually listened to the parties” [italics in original] and were “actually alert and sensitive to the matter before it”: Vavilov, at paragraphs 127–128.
[35] In some cases, the requirement of a reasoned explanation is higher:
…. Where the impact of a decision on an individual’s rights and interests is severe, the reasons provided to that individual must reflect the stakes. The principle of responsive justification means that if a decision has particularly harsh consequences for the affected individual, the decision maker must explain why its decision best reflects the legislature’s intention.
(Vavilov, at paragraph 133.) As well, a failure to grapple with the consequences of a decision should be considered: Vavilov, at paragraph 134, citing Chieu v. Canada (Minister of Citizenship and Immigration), 2002 SCC 3, [2002] 1 S.C.R. 84.
[36] In order to interfere, a reviewing court needs to find a “fundamental gap” in express or implied reasoning, a “fail[ure] to reveal a rational chain of analysis”, a “flawed basis”, a finding that the “decision is based on an unreasonable chain of analysis” or “an irrational chain of analysis”, unintelligibility in the sense that “the reasons read in conjunction with the record do not make it possible to understand the decision maker’s reasoning on a critical point” or “clear logical fallacies, such as circular reasoning, false dilemmas, unfounded generalizations or an absurd premise”: Vavilov, at paragraphs 96 and 103–104. These problems must be on a key point, “sufficiently central” or “significant” such that they point to “sufficiently serious shortcomings in the decision”: Vavilov, at paragraph 100. They must be “more than merely superficial or peripheral to the merits of the decision”: Vavilov, at paragraph 100.
[37] In Vavilov, the Supreme Court tells us that we should not be too hasty to find these sorts of flaws. Vavilov’s requirement of a reasoned explanation cannot be applied in a way that transforms reasonableness review into correctness review. If reviewing courts are too fussy and adopt the attitude of a literary critic all too willing to find shortcomings, they will be conducting correctness review, not reasonableness review. That would return us to the bad old days in the 1960’s and 1970’s when reviewing courts would come up with any old excuse to strike down decisions they disliked—and often did: see Canadian Copyright Licensing Agency (Access Copyright) v. Canada, 2018 FCA 58, 422 D.L.R. (4th) 112, [2018] 4 F.C.R. D-3, at paragraphs 61–65.
[38] Is silence on a particular point a “fundamental gap” that warrants intervention by the reviewing court? Not necessarily. For one thing, the administrator’s reasons, read alone or in light of the record in a holistic and sensitive way, or the record before the administrator, or one of its previous decision might lead the reviewing court to find that an implicit finding must have been made.
[39] Vavilov also reminds us that reviewing courts “must remain acutely aware” that we cannot expect administrators “to deploy the same array of legal techniques that might be expected of a lawyer or judge” and that “ʻ[a]dministrative justice’ will not always look like ‘judicial justice’”: Vavilov, at paragraphs 92 and 119. To expect otherwise is to overly judicialize administrative processes, threatening their efficiency and potentially undermining the very reasons why the legislator entrusted this jurisdiction to the administrator in the first place: see, e.g., Canadian Union of Public Employees, Local 301 v. Montreal (City), [1997] 1 S.C.R. 793, (1997), 144 D.L.R. (4th) 577, at paragraph 39. This consideration is relevant to the appeals here, as members of the Immigration Division and Immigration Appeal Division need not be lawyers: Immigration and Refugee Protection Act, section 153.
[40] Overall, Vavilov instructs us that reasons “must not be assessed against a standard of perfection” and administrators should not be held to the “standards of academic logicians”: Vavilov, at paragraphs 91 and 104. An administrator’s reasons may “ʻnot include all the arguments, statutory provisions, jurisprudence or other details the reviewing judge would have preferred’” but that is not a basis on its own to set the decision aside: Vavilov, at paragraph 91. Rather, “a reviewing court must ultimately be satisfied that the [administrator’s] reasoning ‘adds up’”: Vavilov, at paragraph 104.
[41] Turning specifically now to legislative interpretation, what is said above about administrative decisions in general applies here too. Among other things, reviewing courts should be aware that the administrator may have made implied findings on issues of legislative interpretation. For example, suppose an administrator finds that almost all of the elements of text, context and purpose support a particular legislative interpretation. Its non-mention of a couple of elements put forward by a party is not necessarily a fundamental gap that is fatal. The reviewing court might be able to conclude that the administrator implicitly found that the preponderance of elements supported its view of the matter—in other words, that although certain matters were not mentioned in the reasons, they were considered and rejected by it or were found to be outweighed by other matters. And even where elements of the analysis are left out and, in the whole scheme of things, the omissions are minor, the decision is “not … undermine[d] … as a whole” and must stand: Vavilov, at paragraph 122.
[42] According to Vavilov, at paragraphs 120–122, where a party alleges that an administrator’s legislative interpretation is unreasonable in outcome or unreasonably omitted a relevant element in the process of legislative interpretation, the questions are whether the administrative decision:
• is “alive to [the] essential elements” of text, context and purpose: Vavilov, at paragraph 120; in this regard, a decision is “not required ‘to explicitly address all possible shades of meaning’ of a given provision” and may not have to “dwell on each and every signal of statutory intent”; instead, it is allowed “to touch upon only the most salient aspects of the text, context or purpose”: Vavilov, at paragraph 122; Construction Labour Relations v. Driver Iron Inc., 2012 SCC 65, [2012] 3 S.C.R. 405, at paragraph 3.
• contains an important “omitted aspect”, i.e., a consideration that cannot be seen in the reasons and cannot be implied, whose importance is so great that it “causes the reviewing court to lose confidence in the outcome reached by the decision maker”: Vavilov, at paragraph 122;
• is “consistent with the text, context and purpose of the provision”: Vavilov, at paragraph 120;
• is genuine, i.e., is not tendentious, is not expedient and is not result-oriented: Vavilov, at paragraphs 120–121. On this, see also Williams v. Canada (Public Safety and Emergency Preparedness), 2017 FCA 252, [2018] 4 F.C.R. 174, at paragraphs 41–52; Canada v. Cheema, 2018 FCA 45, [2018] 4 F.C.R. 328, at paragraphs 73–86; Hillier, above, at paragraphs 18 and 24–27; Canada (Attorney General) v. Utah, 2020 FCA 224, 455 D.L.R. (4th) 714, at paragraph 15 (all in the context of courts but equally applicable to administrators).
[43] I shall now conduct reasonableness review in accordance with the foregoing principles.
B. The decisions of the Immigration Appeal Division in Mason and the Immigration Division in Dleiow were reasonable
[44] Sections 34 and 36 of the Immigration and Refugee Protection Act read as follows:
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).
…
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.
Application
(3) The following provisions govern subsections (1) and (2):
(a) an offence that may be prosecuted either summarily or by way of indictment is deemed to be an indictable offence, even if it has been prosecuted summarily;
(b) inadmissibility under subsections (1) and (2) may not be based on a conviction in respect of which a record suspension has been ordered and has not been revoked or ceased to have effect under the Criminal Records Act, or in respect of which there has been a final determination of an acquittal;
(c) the matters referred to in paragraphs (1)(b) and (c) and (2)(b) and (c) do not constitute inadmissibility in respect of a permanent resident or foreign national who, after the prescribed period, satisfies the Minister that they have been rehabilitated or who is a member of a prescribed class that is deemed to have been rehabilitated;
(d) a determination of whether a permanent resident has committed an act described in paragraph (1)(c) must be based on a balance of probabilities; and
(e) inadmissibility under subsections (1) and (2) may not be based on an offence
(i) designated as a contravention under the Contraventions Act,
(ii) for which the permanent resident or foreign national is found guilty under the Young Offenders Act, chapter Y-1 of the Revised Statutes of Canada, 1985, or
(iii) for which the permanent resident or foreign national received a youth sentence under the Youth Criminal Justice Act.
(1) The decision of the Immigration Appeal Division in Mason
[45] In Mason [Mason v. Canada (Public Safety and Emergency Preparedness), 2019 CanLII 55171 (I.R.B.)], the Immigration Appeal Division concluded as follows (at paragraphs 37–38):
Inadmissibility under paragraph 34(1)(e) does not require that the conduct have a link to national security or the security of Canada. Parliament intended that the provisions of subsection 34(1) relate to security in a broader sense. That includes ensuring that individual Canadians are secure from acts of violence that would or might endanger their lives or safety.
Section 36 of the [Immigration and Refugee Protection Act] creates a class of inadmissibility for serious criminality, requiring, for offences in Canada, a conviction. [Paragraph] 34(1)(e) creates a class of inadmissibility for engaging in acts of violence, criminal or not, that would or might endanger the lives or safety of persons in Canada. The two grounds of inadmissibility overlap but are distinct.
[46] In reaching its conclusion, the Immigration Appeal Division in Mason was, in the words of Vavilov, at paragraph 120, very much “alive to [the] essential elements” of the text, context and purpose of paragraph 34(1)(e), analyzing the most important elements of each. In my view, for the reasons that follow, a reasoned explanation can be discerned from the reasons it gave and from some matters that can be implied. As will be explained, given what was in the record and given what was argued, there are no “omitted aspects” that would cause a “[loss of] confidence in the outcome reached by the decision maker”: Vavilov, at paragraph 122.
[47] The Minister urged the Immigration Appeal Division to consider only the text of paragraph 34(1)(e). The Minister contended that the grammatical and ordinary sense of the text of the paragraph supports his position. The Immigration Appeal Division reasonably replied, supported by the authorities on legislative interpretation, that “[t]hat [the Minister’s] approach is not sufficient” because “paragraph [34(1)(e)] cannot be read in isolation” (at paragraph 20).
[48] Other indicia of reasonableness are peppered throughout the Immigration Appeal Division’s express reasons. For example, the Immigration Appeal Division directed its attention (at paragraph 22) to the terms “ʻsecurity’” and “ʻsecurity grounds’” in subsection 34(1) and reviewed the jurisprudence of the Immigration Division in this case and one other case. It noted (at paragraph 23) that elsewhere in the Act, when Parliament intends “ʻsecurity of Canada’” and “ʻnational security’”, it uses those terms, not the general term “ʻsecurity’”. Relying on the legally recognized “presumption of consistent expression”, it reasoned that if Parliament intended paragraph 34(1)(d) to have a national security nexus it would have used those terms (at paragraph 23). It also noted that if Mr. Mason’s interpretation were adopted, the use of the term “ʻsecurity of Canada’” in paragraph 34(1)(d) would be redundant (at paragraph 23). For further assistance, the Immigration Appeal Division also looked to the dictionary definition of “ʻsecurity’”, just as many courts do (at paragraph 25).
[49] On purpose, the Immigration Appeal Division was appropriately cautious. It noted paragraphs 3(1)(h) and (i) of the Immigration and Refugee Protection Act supported its interpretation (at paragraph 39) but once again exercised caution in light of governing Supreme Court law (at paragraph 34). It reasonably and appropriately drew upon the Supreme Court’s decision in Agraira v. Canada (Public Safety and Emergency Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, at paragraph 78 on how to reconcile the 11 purposes in section 3 of the Immigration and Refugee Protection Act. In that decision, the Supreme Court held that paragraphs 3(1)(h) and (i) must be interpreted in light of the other objectives in section 3. It held (at paragraph 34) that collectively these are “the values of a democratic state … committed to protecting the fundamental values of its Charter and its history as a parliamentary democracy”. Drawing upon this, the Immigration Appeal Division asked itself (at paragraph 35) whether “finding someone inadmissible for acts of violence that were arguably criminal, but which did not lead to a criminal conviction, contrary to Canadian values, the fundamental values of the Charter and our history as a parliamentary democracy?” It answered that question in the negative (at paragraphs 35 and 39).
[50] Viewed from the eyes of a second-guessing reviewing court, this approach seems a little loose, perhaps even a little overwrought. But this approach came from the Supreme Court in Agraira. The Immigration Appeal Division cannot be found unreasonable for following the approach of our highest court.
[51] As Vavilov encourages it to do, the Immigration Appeal Division also considered the potentially constraining effect of judicial decisions. It noted (at paragraph 10) that there were no judicial decisions directly on point. But (at paragraph 26) it drew upon some of the words of the Supreme Court in Suresh v. Canada (Minister of Citizenship and Immigration), 2002 SCC 1, [2002] 1 S.C.R. 3 concerning subsection 34(2) as a further indicator that more than just national security was covered. And it also drew (at paragraphs 27–28) upon the Federal Court’s statements in El Werfalli v. Canada (Public Safety and Emergency Preparedness), 2013 FC 612, [2014] 4 F.C.R. 673, at paragraph 75 and Fuentes v. Canada (Minister of Citizenship and Immigration), 2003 FCT 379, [2003] 4 F.C. 249, at paragraph 62 but, on a sensitive reading of them, plausibly dismissed certain statements as obiter and, indeed, they were arguably so. This shows that it considered these authorities carefully. Overall, it took on the important aspects of the interpretive issue before it and grappled with them.
[52] The Immigration Appeal Division also examined the context of the other inadmissibility provisions in the Act. In particular, it examined (at paragraphs 31–33 and 38) section 36, which provides for inadmissibility upon conviction of a criminal offence.
[53] This is where Mr. Mason submits that the Immigration Appeal Division’s decision was unreasonable. He says it ignored his contextual argument. On this, he is wrong.
[54] The Immigration Appeal Division did consider (at paragraphs 30–33) this context. It refused to limit its analysis to just the other paragraphs of subsection 34(1). In its view, “[c]ontext must come not just from the immediate co-text, but from the overall scheme and object of the [Immigration and Refugee Protection Act]” and “the overall scheme of the inadmissibility provisions in Part I, Division 4 of the [Immigration and Refugee Protection Act] is particularly relevant” (at paragraph 21). This non-blinkered approach to issues of context and the Immigration Appeal Division’s application of it to this case cannot be said to be unreasonable.
[55] The Immigration Appeal Division also reasoned that the mere fact some criminality could fall under paragraph 34(1)(e) did not require it to have a connection to national security. It noted that while there is some overlap, the two provisions do not cover the exact same types of behavior. In its view (at paragraph 33), the conduct captured by paragraph 34(1)(e), which speaks of the danger posed to the “lives and safety” of persons in Canada, is only a “small subset of what would be considered serious criminality in section 36”. And it found (at paragraph 33) that sections 34 and 36 deal with two different matters, conduct and convictions respectively. These are plausible conclusions consistent with reasonableness.
[56] The Immigration Appeal Division rejected the submission that, because section 36 required proof of a conviction for common criminality, paragraph 34(1)(e) may not apply to behavior that could fall under section 36. It noted (at paragraph 32) that there are other provisions in the Immigration and Refugee Protection Act that cover criminal behavior but permit inadmissibility without proof of a conviction.
[57] And the Immigration Appeal Division rejected the submission that paragraph 34(1)(e) would have absurd consequences if it did not have a connection to national security. It found (at paragraph 36) that paragraph 34(1)(e) would not be absurdly broad because the conduct captured by it is “narrowly defined”. This can only mean that it interpreted “safety” as something approaching the level of a threat to life, not just minor harm. And it accurately found (at paragraphs 35–36) that inadmissibility under section 34 is not governed by criminal law or contrary to paragraph 11(d) of the Charter [Canadian Charter of Rights and Freedoms]. None of these findings can be said to be unreasonable.
[58] Applying the words of Vavilov, at paragraph 42 above, I conclude that to the extent some possible elements of context do not appear in the Immigration Appeal Division’s reasons in Mason, this does not cause “[loss of] confidence in the outcome reached by the decision maker”: Vavilov, at paragraph 122. The Immigration Appeal Division “touch[ed] upon only the most salient aspects of the text, context or purpose” and was not required to deal with all aspects: Vavilov, at paragraph 122. It was “alive to [the] essential elements” of text, context and purpose, indeed very much so: Vavilov, at paragraph 120.
[59] To the extent that the Immigration Appeal Division failed to mention some elements in its analysis of text, context and purpose, this was not a fundamental gap. In these circumstances and consistent with paragraph 41 above, I conclude that the Immigration Appeal Division implicitly found that the preponderance of elements supported the Minister’s interpretation. In other words, although one can quibble that certain elements of text, context and purpose were not mentioned in the reasons, I am confident from the quality of the Immigration Appeal Division’s overall reasoning that it considered them to be outweighed by other elements.
[60] Contrary to the foregoing, the Federal Court concluded that the Immigration Appeal Division’s decision in Mason was unreasonable because it failed to appreciate the context of other inadmissibility provisions in the Immigration and Refugee Protection Act.
[61] I disagree with the Federal Court. What it did was akin to correctness review and, in some respects, it erred in law.
[62] The Federal Court found that one element of context outweighed all others. It looked at the consequences that flow from sections 34 and 36 inadmissibility findings, and concluded that, because somewhat harsher consequences flow from section 34 than section 36, paragraph 34(1)(e) must have a connection to national security (see paragraphs 39–51). The Federal Court concluded that paragraph 34(1)(e)—without a connection to national security—would just be section 36 “lite” and would “render section 36 meaningless” (at paragraph 53).
[63] But this conclusion is based on an incorrect assumption. The Federal Court assumed that the behaviour captured by paragraph 34(1)(e), “acts of violence”, and section 36, “offence[s] under an Act of Parliament”, is essentially the same. It is not. Section 36 applies to much more, including much non-violent, criminal behaviour. Even if paragraph 34(1)(e) is not connected to national security, section 36 plays a meaningful role in the inadmissibility provisions of the Act.
[64] Even “serious criminality”, the harshest criminality finding in the Immigration and Refugee Protection Act, includes a host of non-violent offences. Non-violent offences with 10-year maximum penalties are strewn throughout the Criminal Code, R.S.C., 1985, c. C-46: for example, white-collar crime (sections 382, 382.1 and 400), non-violent sexual offences (sections 151–153, 155, 160 and 172.2), theft (paragraph 334(a)), and child pornography (section 163.1). And many, many others too: I have found at least 31 additional, separate Criminal Code sections setting out offences that are covered by even the strictest interpretation of section 36. In section 36, read in light of the Criminal Code, Parliament has chosen to make a whole host of socially undesirable behaviour grounds for inadmissibility. By comparison, paragraph 34(1)(e) is the narrow provision, applying just to “acts of violence”.
[65] The Federal Court erred in other ways when it imposed its view over that of the Immigration Appeal Division. It gave little to no weight to the presumption against redundancy (at paragraphs 56–57): Canada (Attorney General) v. Distribution G.V.A. Inc., 2018 FCA 146, at paragraph 35. It asserted (at paragraph 49) that paragraph 34(1)(e) could have a broad scope, ignoring the internal limits in that paragraph. And (at paragraph 49) it tested its preferred interpretation against the facts in Dleiow [Dleiow v. Canada (Public Safety and Emergency Preparedness), 2019 CanLII 129531 (I.R.B.)] and concluded that they supported its conclusion.
[66] On this last-mentioned point, a court interpreting legislation itself can verify its interpretation by looking at its effects: Williams, at paragraph 52. But a reviewing court should not normally do this because it is not interpreting the provision itself: it is reviewing the administrator’s interpretation. And, in any event, in this case, the Federal Court’s assessment based on the facts the administrator found in Dleiow—serial and severe incidents of domestic violence—is open to question. At the very least, it is an open question whether an interpretation of paragraph 34(1)(e) that prevents a finding of inadmissibility to Canada in circumstances such as those in Dleiow accords with the purposes of section 3 of the Immigration and Refugee Protection Act.
(2) The decision of the Immigration Division in Dleiow
[67] The Immigration Division in Dleiow followed the Immigration Appeal Division decision in Mason. Its decision came before the Federal Court judgment in Mason. Its decision to follow the Immigration Appeal Division decision in Mason was reasonable because, at that time, it could be said to be the most recent and persuasive authority on the meaning of paragraph 34(1)(e).
[68] The Immigration Division considered Mr. Dleiow’s submission that the Immigration Appeal Division wrongly decided Mason. It rejected the submission (at paragraph 11). In doing so, it did not regard the Immigration Appeal Division’s decision in Mason or any other Immigration Appeal Division decisions as binding on it. This accords with the case law as it existed at that time and as it exists today: see, e.g., Vavilov, at paragraphs 129–132; see also Canada (Attorney General) v. Bri-Chem Supply Ltd., 2016 FCA 257, [2017] 3 F.C.R. 123, at paragraphs 33–51 and cases cited therein. However, the Immigration Division did recognize the value of certainty and consistency in decision-making. So it applied (at paragraphs 12–14) a threshold it said it had to meet before departing from the Immigration Appeal Division’s decision in Mason. This was a defensible approach. Finally, it concluded (at paragraph 15) that the threshold was not met. Thus, it adopted the Immigration Appeal Division’s interpretation in Mason. It concluded that no connection to national security is required under paragraph 34(1)(e).
[69] Then the Immigration Division in Dleiow applied paragraph 34(1)(e) to the facts before it. It found that Mr. Dleiow’s pattern of domestic violence constituted “acts of violence that would or might endanger the lives or safety of persons in Canada”. Mr. Dleiow does not challenge this finding in his notice of application for judicial review or his notice of appeal. Overall, the decision of the Immigration Division in Dleiow, highly contingent on the decision in Mason, is reasonable.
[70] The Federal Court heard the application for judicial review in Dleiow after the Federal Court had decided Mason. It quashed the Immigration Division in Dleiow only because it felt it had to follow the Federal Court’s judgment in Mason. As the Federal Court’s judgment in Mason must be set aside, so must the judgment of the Federal Court in Dleiow be set aside.
[71] Before closing, I wish to offer some final observations.
[72] First, not all possible arguments on legislative interpretation were put to these administrators. For example, in their written submissions to the Immigration Appeal Division, neither Mr. Mason nor Mr. Dleiow invoked the Refugee Convention, namely the United Nations Convention Relating to the Status of Refugees, signed at Geneva on July 28, 1951, [[1969] Can. T.S. No. 6], and the Protocol Relating to the Status of Refugees, signed at New York on January 31, 1967, [606 U.N.T.S. 267]. As a result, neither administrator considered whether the Refugee Convention was relevant to the interpretation of paragraph 34(1)(e) and, if so, how.
[73] Mr. Mason attempted to invoke the Refugee Convention in argument before us. But in this Court that is a new issue and we should not entertain it: Alberta (Information and Privacy Commissioner) v. Alberta Teachers’ Association, 2011 SCC 61, [2011] 3 S.C.R. 654, at paragraphs 23–26. It goes to the merits of the interpretation of paragraph 34(1)(e). That issue should be made to the merits-deciders under this legislative regime, in particular the Immigration Appeal Division, not a reviewing court or a court sitting in appeal from a reviewing court: ‘Namgis First Nation v. Canada (Fisheries and Oceans), 2019 FCA 149; Forest Ethics Advocacy Association v. Canada (National Energy Board), 2014 FCA 245, [2015] 4 F.C.R. 75.
[74] As well, certain background documents and other instruments needed to understand any international obligations are not in evidence before us. This is because they were not placed in evidence before the administrators. The forum for the introduction of evidence is the proceeding before the administrators, not a reviewing court and not a court sitting in appeal from a reviewing court: Association of Universities and Colleges of Canada v. Canadian Copyright Licensing Agency (Access Copyright), 2012 FCA 22, 428 N.R. 297, at paragraphs 14–20; Bernard v. Canada (Revenue Agency), 2015 FCA 263, 9 Admin. L.R. (6th) 296, at paragraphs 13–28; Bell Canada v. 7262591 Canada Ltd. (Gusto TV), 2016 FCA 123, 17 Admin. L.R. (6th) 175, at paragraphs 7–11.
[75] Different arguments and new supporting evidence may be placed before later administrators considering the interpretation of paragraph 34(1)(e). Later administrators are not bound by the decisions of the administrators in Mason and Dleiow. They may decide differently as long as their decision is reasonable and they give a reasoned explanation for departing from earlier decisions: Vavilov, at paragraphs 129–132; see also Canada (Attorney General) v. Bri-Chem Supply Ltd. [cited above], at paragraphs 33–51 and cases cited therein. Rival administrative interpretations of legislation can co-exist under the reasonableness standard though persistent discord can cause serious concerns about consistency and the rule of law: Vavilov, at paragraph 129; see also the concerns expressed by this Court in Wilson v. Atomic Energy of Canada Limited, 2015 FCA 17, [2015] 4 F.C.R. 467.
[76] Consistent with Hillier, I have not performed my own interpretation of paragraph 34(1)(e). But, again consistent with Hillier, I have gained an appreciation of the interpretive landscape in order to conduct reasonableness review. From this, I observe that some elements of text, context and purpose concerning paragraph 34(1)(e) favour the administrative interpretations reached in these cases, while others may not. Here, the issue of legislative interpretation is best described as one where the issue is open to some debate.
[77] To avoid the prospect of duelling administrative interpretations of paragraph 34(1)(e)—and all the uncertainty, inconsistent application and unfairness that might result—administrators tempted to reach a different interpretation may wish to follow another route. At any stage during proceedings, a “federal board, commission or other tribunal”, such as the Immigration Appeal Division, may “refer any question or issue of law … to the Federal Court for hearing and determination”: subsection 18.3(1) of the Federal Courts Act, R.S.C., 1985, c. F-7. Needless to say, in such a reference, the Federal Courts would not have to defer to any administrative decision-making, could receive all necessary evidence and submissions, and could pronounce the correct state of the law.
[78] Here, we have a major, stand-alone issue of legislative interpretation that arises on the facts of these cases. It is purely legal, requiring an examination of text, context and purpose. Unlike some administrators such as the National Energy Board in Forest Ethics, above, the Immigration and Refugee Board does not have any particular expertise that might contribute to the analysis of text, context and purpose. In the future, a member of the Immigration and Refugee Board could well conclude that it is entirely appropriate to refer this issue to the Federal Court for resolution once and for all.
[79] In making these observations, I am not expressing or implying any agreement or disagreement with the interpretation adopted by the administrators in the two cases before us.
C. Conclusion and proposed disposition
[80] For the foregoing reasons, the decision of the Immigration Appeal Division in Mason and the decision of the Immigration Division in Dleiow concerning the interpretation of paragraph 34(1)(e) of the Immigration and Refugee Protection Act are reasonable. The Federal Court in Mason and the Federal Court in Dleiow should not have set them aside.
[81] I would answer the certified question in each appeal as follows:
Q.: Is it reasonable to interpret para. 34(1)(e) of the Immigration and Refugee Protection Act in a manner that does not require proof of conduct that has a nexus with “national security” or the “security of Canada”?
A. Yes.
[82] I would allow the appeals, set aside the judgments of the Federal Court dated October 2, 2019 and January 16, 2020 in files IMM-1645-19 and IMM-4199-19 respectively and, making the judgment the Federal Court should have made in each file, dismiss the applications for judicial review.
Rennie J.A.: I agree.
Mactavish J.A.: I agree.