[7] The RPD applied Ezokola v Canada (Citizenship and Immigration), 2013 SCC 40 [Ezokola], in which the Supreme Court stated that a person is excluded from refugee protection under Article 1F(a) for complicity in international crimes “if there are serious reasons for considering that he or she voluntarily made a knowing and significant contribution to the crime or criminal purpose of the group alleged to have committed the crime.” The RPD determined that the acts committed by the applicant as an employee of the Organization constitute in themselves voluntary, significant, and knowing contributions to the crimes against humanity committed by it. The RPD’s rationale for reaching this conclusion spans nearly 100 paragraphs. I summarize it below:

(i) Voluntary contribution: The RPD concluded that the applicant’s contributions were voluntary because he freely joined the Organization and agreed to climb its ranks during his long career. The RPD found the applicant’s allegation that if he had resigned from his position, he and his family would have been punished not to be credible.
(ii) Knowing contribution: The RPD concluded that the applicant’s contributions were knowing. It noted that although the applicant denied that he was personally aware of specific crimes committed by the Organization between 1990 and 2006, he was aware of the repressive nature of the security community for which he worked and of the fact that crimes against humanity were being committed.
(iii) Nature of the Organization: Before the RPD, the applicant submitted that he was part of the Organization’s economic division and, as such, did not participate in crimes committed by the Organization’s other divisions. The RPD concluded that the question of whether the applicant was part of an economic division was not determinative, because regardless of whether it was [translation] “economic” or not, the refugee protection claimant’s intelligence work was, by his own admission, very significant work that had consequences for important interests of the Burundian state. The SPR determined that the applicant’s allegation that an economic division was supposedly insulated from the crimes committed by the Organization was not credible, considering the Organization’s fundamentally violent nature.
(iv) The applicant’s positions: The RPD noted that the applicant had climbed the ranks during his career in the Organization and inferred from this fact that, as he progressed, he was likely to have become aware of the Organization’s criminal activities. The RPD also noted that the applicant spent 16 years in the Organization. In light of that fact, the evidence that the Organization’s criminal acts were widespread and that it was relatively small in size, the RPD concluded that the applicant must have had detailed knowledge of the crimes against humanity to which he was contributing because of his duties as an intelligence officer, on a balance of probabilities.
(v) Significant nature: The RPD found that the applicant’s contributions to the crimes against humanity committed by the Organization were significant. It noted that the information gathered by the applicant in the performance of his duties was, on a balance of probabilities, used to commit international crimes.

III. Decision under review

[8] In a decision of about twenty paragraphs in length, the RAD confirmed the RPD’s decision that the applicant could not receive protection from Canada, given that he was excluded under section 98 of the IRPA and Article 1F(a) of the Convention. The RAD found that the RPD did not make the errors of which it was accused, namely, that it failed to consider all the evidence before it and that it misapplied the principles of Ezokola.

[9] The RAD found that the RPD relied on the evidence before it and met the tests set out in Ezokola in making its decision. The RAD noted that it considered, on reading the written transcript of the RPD hearing and the evidence as a whole, including the documentary evidence, that the applicant’s arguments were unfounded. According to the RAD, the RPD did not err in finding that there were serious reasons to believe that the applicant was complicit in the crimes against humanity committed by the Organization between 1990 and 2006.

IV. Issues and standard of review

[10] The only issue in the case at hand is whether the RAD decision is reasonable. The applicant claims that the decision is unreasonable because of:

a) an analysis with too much deference to the previous decision, with the result that the RAD failed to conduct an independent analysis;
b) omissions by the RAD, which did not consider evidence showing that the applicant could not have been complicit in the crimes against humanity committed by the Organization during the period he worked for it; and
c) an unreasonable application of the case law regarding complicity in the commission of crimes against humanity, and in particular Ezokola.

[11] The applicable standard of review in the case at hand is reasonableness. There is a presumption that this standard applies when reviewing administrative decisions (Canada (Minister of Citizenship and Immigration) v Vavilov, 2019 SCC 65 [Vavilov] at para 25), and the exceptions to this presumption do not apply here (Vavilov at para 17).

[12] In order for the Court to be satisfied that a decision is unreasonable, the party contesting the decision must demonstrate that it is not “based on an intrinsically coherent and rational analysis” or “justified in relation to the facts and the law that constrain the decision maker” (Vavilov at para 85). The Court must be satisfied that there are sufficiently serious shortcomings in the decision such that superficial or peripheral shortcomings would be sufficient to set aside the decision (Vavilov at para 100). Moreover, the Court must consider the decision as a whole and avoid conducting a “line-by-line treasure hunt for error” (Vavilov at para 102).

V. Positions of the parties and analysis of the determinative issue of undue deference

[13] The applicant argues that the RAD did not conduct its own independent evaluation of the file, instead simply upholding all the RPD’s findings. In so doing, it apparently failed to fulfill its duty as a hybrid appeal tribunal, as described in Canada (Citizenship and Immigration) v Huruglica, 2016 FCA 93 at para 103 [Huruglica].

[14] The applicant contends that although the RAD indicated that it [translation] “conducted an independent analysis of the file” and that it “applied the correctness standard”, the body of its decision demonstrates the opposite. The applicant argues that the RAD, throughout its decision, merely repeats that the RPD did not make the errors of which it was accused, without giving reasons. The applicant adds that the mere fact that the RAD listed the Ezokola principles and stated that the RPD did not err does not demonstrate, in the RAD’s own analysis, why these principles were applied properly.

[15] The applicant submits that because of the endorsement of all the RPD’s findings, it is impossible to determine where deference to the RPD decision ends and where the RAD’s independent analysis begins (Jeyaseelan v Canada (Citizenship and Immigration), 2017 FC 278 at para 19 [Jeyaseelan]). In support of his argument, the applicant lists several passages where the RAD relies on the RPD and its analysis. The applicant argues that these passages are an indication of the excessively high level of deference that the RAD gave to the RPD decision (Jeyaseelan at para 20).

[16] The respondent argues that RAD conducted its own analysis of the evidence to determine that there are serious reasons to believe that the applicant voluntarily made a knowing and significant contribution to crimes against humanity committed by the Organization.

[17] I can only agree with the applicant. The RAD’s reasons lead me to doubt strongly that it properly exercised its role as a hybrid appeal tribunal, as defined in Huruglica.

[18] I also note that the respondent hardly addressed this first issue of independence in its memorandum, focusing instead on the applicant’s other two arguments cited above. I pointed this out to the respondent at the hearing and asked whether the respondent’s counsel had more to say on this subject. She did not challenge the weaknesses in the RAD’s decision as to the endorsement of the RPD’s entire line of reasoning. I will discuss why this was a determinative error.

[19] When sitting on appeal from an RPD decision, the RAD must apply the correctness standard (Huruglica at para 78). Although an appeal to the RAD is not a de novo appeal, it is well established that the RAD must conduct its own analysis of the file to determine whether its intervention is required (Huruglica at paras 79 and 103). The RAD owes no deference to the RPD decision, except in the exceptional case where it is recognized that the RPD had a meaningful advantage with respect to assessing the credibility of viva voce evidence (Huruglica at para 70). Such a meaningful advantage has not been recognized in the case at hand (see also Rozas del Solar v Canada (Citizenship and Immigration), 2018 FC 1145, [2019] 2 FCR 597 at para 106).

[20] Simply because RAD agrees with the RPD’s findings does not necessarily mean that it did not conduct an independent analysis of the file (Ademi v Canada (Citizenship and Immigration), 2021 FC 366 at para 28). However, I am of the view that the RAD did not provide “cogent reasons which enables this Court upon judicial review to determine what evidence the RAD itself found persuasive and trace the path of its reasoning” (Allen v Canada (Citizenship and Immigration), 2015 FC 994 at para 18).

[21] I referred to the length of the two decisions. As a reminder, the RPD decision contains almost one hundred paragraphs, while that of the RAD contains only about twenty. The RAD decision is in fact brief and essentially a repetition of the RPD’s findings and reasoning and does not respond to the substantive arguments of the case raised by the applicant’s counsel in his detailed memorandum.

[22] Specifically, in its Decision, at paragraphs 7 to 10, the RAD lists the arguments the applicant raised on appeal. At paragraphs 12 to 15, it refers to the Ezokola principles regarding complicity in the commission of crimes against humanity. At paragraphs 16 to 20, it reviews the analysis conducted by the RPD, notes some evidence that the RPD reviewed, and states that having read the evidence on the record, it is of the opinion that the RPD did not make the errors of which it was accused by the applicant.

[23] Although it mentions that it followed the principles set out in Huruglica, I do not find a single passage in the RAD Decision where it sets out its analysis of the file and draws its own conclusions in relation to the applicable legal principles. Having read the RAD decision, it appears to me that it has instead analyzed the RPD’s analysis and relied on its conclusions.

[24] An illustration of this approach is the applicant’s listing of approximately twenty RAD references to the RPD’s findings, the evidence that the RPD considered, and even direct quotes from the RPD decision. The RAD’s systematic use of language such as [translation] “the RPD relied on,” “the RPD took into account,” and “the RPD found that” throughout its analysis does not reflect an independent analysis; failure to conduct an independent analysis is a reviewable error (Jeyaseelan at para. 21); Khachatourian v Canada (Citizenship and Immigration), 2015 FC 182 at para 33; Denis v Canada (Citizenship and Immigration), 2018 FC 1182 at para 39).

[25] In addition, I would add that, in my opinion, the RAD’s reasons are insufficient to satisfy Vavilov’s requirements of justification, transparency, and intelligibility. The RAD lists the applicant’s numerous arguments in the first part of its decision but does not refer to any of the applicant’s particular arguments in the subsequent parts of the Decision and fails to explain the reasons why it rejected all his arguments.

[26] I would like to reiterate that the fact that a decision maker failed to meaningfully grapple with the main arguments raised by the parties calls into question whether the decision maker was actually alert and sensitive to them (Vavilov at para 128). The RAD ought to have explained, in light of the facts and applicable law, why, for example, the applicant was not simply [translation] “associated” with the Organization and why, in its view, his duties as an intelligence officer constituted voluntary, significant, and knowing contributions to the crimes against humanity committed by the Organization. The RAD did not do so, relying on the RPD’s findings on this matter. I note that the RAD did the same with the other reasons that the applicant raised: It did not address them independently, merely upholding the RPD’s analysis. The RAD Decision is a mere recital and a blessing of the previous decision. As the Supreme Court stated at para 86 of Vavilov:

In short, it is not enough for the outcome of a decision to be justifiable. Where reasons for a decision are required, the decision must also be justified, by way of those reasons, by the decision maker to those to whom the decision applies. While some outcomes may be so at odds with the legal and factual context that they could never be supported by intelligible and rational reasoning, an otherwise reasonable outcome also cannot stand if it was reached on an improper basis.

[27] It is entirely possible that the RAD, even if it had presented reasons supporting an independent analysis, would have reached the same outcome as the RPD. However, it is not the role of a reviewing court, in applying the standard of reasonableness, to usurp the functions of an administrative tribunal by making the decision that, in its view, should have been made (Vavilov at para 98).

[28] In light of the above, I am of the view that the matter should be remitted to the RAD for redetermination. Since the tribunal failed to exercise its independence, the decision is unreasonable, and this first issue is determinative. There is no need to consider the other issues raised by the applicant.