Docket: IMM-1091-14
Citation:
2014 FC 952
[UNREVISED ENGLISH CERTIFIED
TRANSLATION]
Toronto, Ontario, October 8, 2014
Present: The Honourable Mr. Justice Martineau
BETWEEN:
|
AGHIAD ALYAFI
|
Applicant
|
and
|
THE MINISTER OF CITIZENSHIP
AND IMMIGRATION
|
Respondent
|
JUDGMENT AND REASONS
[1]
The applicant is challenging the legality of a
decision of the Refugee Appeal Division (RAD) of the Immigration and Refugee
Board (IRB) of Canada, which confirms a previous decision of the Refugee
Protection Division (RPD) denying him refugee protection. For the reasons that
follow, the application for judicial review is allowed by the Court.
[2]
The applicant is a Syrian national. In 2010, he
came to study in Canada. In 2013, with a temporary visa, valid until
January 31, 2015, he made a refugee claim. The fear of the applicant, who
is a Sunni, is based on the fact that he risks being forced to join an army
accused of participating in war crimes, that he is already perceived by the authorities
as a political activist because of his fundraising activities for victims of
the civil war and, finally, since his family is rich, that he risks being kidnapped
or extorted in Syria. On October 2, 2013, the RPD determined that the applicant
is neither a Convention “refugee” or a “person in need of protection”, under
sections 96 and 97 of the Immigration and Refugee Protection Act, SC
2001, c 27 (IRPA).
[3]
The applicant appealed this decision to the RAD.
[4]
On January 30, 2014, the sole RAD member designated
to hear the matter, Stephen J. Gallagher (member), dismissed the applicant’s appeal.
First, the member asked what standard of review should be applied to issues raised
on appeal by the applicant (RAD decision, paras 11 to 13). Since
these are questions of fact (the applicant’s credibility) or mixed fact and law
(determination of a generalized risk), he found that the RPD’s decision must be
reviewed using the standard of judicial review of reasonableness, as defined in
Dunsmuir v New Brunswick, 2008 SCC 9 at para 47 (Dunsmuir) and
Newton v Criminal Trial Lawyers’ Association, 2010 ABCA 399 at para 95.
[5]
Second, the member refused to hold an oral
hearing, being of the view that the new evidence provided by the applicant is not
determinative (paras 14 to 18). After which, he found that the RPD’s
conclusions are, in all respects, reasonable (para 32). In summary, the RPD
could reasonably conclude that the risk of being enrolled in the Syrian army
was speculative as long as the applicant continued to study in Canada (paras 19 to 23). The RPD’s findings of lack of personalized risk (paras 24 and
25) and implausibility that the applicant is perceived as a political activist
by the authorities (paras 26 to 34) are also reasonable. Therefore, the RAD dismissed
the applicant’s appeal, hence this application for judicial review.
[6]
Today, the applicant does not attack the member’s
refusal to hold an oral hearing or the reasonableness of the conclusion on the speculative
nature of the applicant’s enrollment in the Syrian army. This application for
judicial review instead relates to the scope on appeal of the review of evidence
in the RPD record that the RAD conducted under sections 110 and 111 of the
IRPA. According to the applicant, the interpretation of these provisions raises
a question of law, which must be reviewed on a standard of correctness (Budhai
v Canada (Attorney General), 2002 FCA 298 at para 22 and Canada (Attorney General) v Hunter, 2013 FCA 12 at para 4). In this case,
the member did not apply the appropriate mode of analysis. That is sufficient
to set aside the RAD decision. Alternatively, the findings regarding generalized
risk and implausibility are unreasonable.
[7]
The respondent argues that the scope of the
review conducted by the RAD on appeal arises from its expertise. In this case, the
respondent alleges that the Court should apply the standard of reasonableness to
the interpretation that the member made of the scope of the provisions in the
IRPA, because although it is a question of law, it is not a question of law of “central importance to the legal system as a whole” (Dunsmuir,
above at para 60; Alberta (Information and Privacy Commissioner) v
Alberta Teachers’ Association, 2011 SCC 61 at para 84 (Alberta
Teachers’ Association)). Moreover, the member did not commit any reviewable
error. Indeed, nothing requires the RAD to reassess the evidence presented by a
refugee claimant before the RPD, especially since the appeal on the record
prepared by the RPD is not a de novo appeal. In this case, both the
standard of judicial review of reasonableness and the standard on appeal of
palpable and overriding error lead to the same result, that the Court should
not intervene.
[8]
According to settled law from this Court, it seems
that the standard of correctness should be applied to the scope of the review
conducted by the RAD on appeal (Iyamuremye v Canada (Citizenship and
Immigration), 2014 FC 494 at para 20 (Iyamuremye); Garcia
Alvarez v Canada (Citizenship and Immigration), 2014 FC 702 at para 17
(Garcia Alvarez); Eng v Canada (Citizenship and Immigration),
2014 FC 711 at para 18 (Eng); Huruglica v Canada (Citizenship
and Immigration), 2014 FC 799 at paras 24 to 34 (Huruglica)).
[9]
Nevertheless, I do not believe that the choice
of the standard of correctness is determinative in this case, or that it is necessary
to define the exact scope of the review conducted by the RAD on appeal. Other
colleagues of the Court have ruled on this last issue. Serious questions of
general importance regarding the RAD’s jurisdiction and the role on appeal were
or are being certified in various matters, but it must be expected that it will
take several months before the Federal Court of Appeal or even the Supreme
Court of Canada makes a decision.
[10]
It is sufficient to conclude today that the interpretation
adopted by Member Gallagher is not an acceptable outcome in respect of the law,
as the appeal before the RAD is simply not a judicial review, which is a reviewable
error that is determinative in this case (Spasoja c Canada (Citoyenneté et Immigration),
2014 CF 913 at paras 3, 9, 11 and 47 (Spasoja)).
[11]
Paragraphs 110(1) and 111(1) and (2)
of the IRPA state:
110. (1) Subject to subsections (1.1) and (2), a
person or the Minister may appeal, in accordance with the rules of the Board,
on a question of law, of fact or of mixed law and fact, to the Refugee Appeal
Division against a decision of the Refugee Protection Division to allow or
reject the person’s claim for refugee protection.
|
110. (1) Sous réserve des paragraphes (1.1) et
(2), la personne en cause et le ministre peuvent, conformément aux règles de
la Commission, porter en appel — relativement à une question de droit, de
fait ou mixte — auprès de la Section d’appel des réfugiés la décision de la
Section de la protection des réfugiés accordant ou rejetant la demande d’asile.
|
…
|
[…]
|
111. (1)
After considering the appeal, the Refugee Appeal Division shall make one of
the following decisions:
(a) confirm the determination of the Refugee Protection Division;
(b) set aside the determination and substitute a determination
that, in its opinion, should have been made; or
(c) refer the
matter to the Refugee Protection Division for re-determination, giving the
directions to the Refugee Protection Division that it considers appropriate.
|
111. (1) La Section d’appel des réfugiés
confirme la décision attaquée, casse la décision et y substitue la décision
qui aurait dû être rendue ou renvoie, conformément à ses instructions, l’affaire
à la Section de la protection des réfugiés.
|
…
|
[…]
|
(2) The Refugee
Appeal Division may make the referral described in paragraph (1)(c) only if
it is of the opinion that
|
(2) Elle ne peut
procéder au renvoi que si elle estime, à la fois :
|
(a) the decision of
the Refugee Protection Division is wrong in law, in fact or in mixed law and
fact; and
|
a) que la décision
attaquée de la Section de la protection des réfugiés est erronée en droit, en
fait ou en droit et en fait;
|
(b) it cannot make
a decision under paragraph 111(1)(a) or (b) without hearing evidence that was
presented to the Refugee Protection Division.
|
b) qu’elle ne peut
confirmer la décision attaquée ou casser la décision et y substituer la
décision qui aurait dû être rendue sans tenir une nouvelle audience en vue du
réexamen des éléments de preuve qui ont été présentés à la Section de la
protection des réfugiés.
|
[12]
Although it is clear on reading the above provisions
that the RAD may set aside the RPD’s decision and substitute the decision that
should have been made, it must be noted that this does not settle the question of
the said [translation] “deference”
that an appeal tribunal should or should not give to findings of fact, or mixed
facts and law, by the court of first instance. After all, it was the RPD who
heard the witnesses and assessed the probative value of documentary evidence
submitted by the parties. Moreover, the RAD exercises a specialized
jurisdiction on appeal at least equal to that of the RPD at trial. Otherwise, the
creation of a specialized appeal tribunal for refugee determination would serve
no purpose.
[13]
Additionally, not all RPD decisions may be
subject to an appeal to the RAD. For example, even if a country is not part of
those that are excluded from an appeal, when the RPD refers in its decision to
no credible basis for the refugee claim (subsection 107(2) of the IRPA), there
cannot be an appeal before the RAD (paragraph 110(2)(c) of the IRPA).
Further, the RAD may enter new evidence in an appeal and decide to hold an oral
hearing in cases specified by Parliament (subsections 110(3) to (6) of the IRPA).
In this last case, it can probably be argued that it is a kind of de novo
appeal, a point that I do not have to rule on today.
[14]
The very concept of “deference” originated in
case law. When Parliament wants to limit the courts’ power of intervention, it
will adopt so-called “privative” clauses. It could also provide that a decision
may be set aside for one reason or another, as is the case at subsection 18.1(4)
of the Federal Courts Act, RSC 1985, c F-7 (FCA). However, if
the text of the Act allows it, the courts will not interpret the grounds for
review as standards of review and will have to show more or less deference,
depending on the nature of the issues raised by the parties: Dunsmuir, above,
and Canada (Citizenship and Immigration) v Khosa, [2009] 1 S.C.R. 339, 2009
SCC 12 (Khosa).
[15]
As the Supreme Court of Canada has already decided,
with or without a privative clause, a measure of deference is appropriate where
a particular decision has been allocated to an administrative decision-maker in
matters relating to its role, function and expertise (Dunsmuir, above). Some
deference is also owed, regardless of whether the court has had the advantage
of either receiving or not receiving a statutory direction, explicit or by
necessary implication. These general principles of judicial review are not displaced
by section 18.1 of the FCA, which deals essentially with grounds of review
of administrative action, and not of standards of review (Khosa, above).
[16]
To date, with respect to the case law of the RAD
and the Federal Court, on the scope of the review conducted on appeal by the RAD,
three different approaches to deference owed to the RPD can be noted: (1) the
standard of judicial review, or “reasonableness” (decision of Member Gallagher and
various other RAD decisions); (2) the standard of appellate review or of “palpable
and overriding error” (Garcia Alvarez, Eng and Spasoja, above);
and (3) a composite and variable standard of review resulting from the nature of
the claim before the RAD—characterized as a “hybrid appeal”—and the particular
nature of questions of fact, or of mixed fact and law, raised by an appellant (Huruglica,
above).
[17]
Let us consider the first approach. Deference
given to findings of fact or mixed fact and law by a reviewing Court is well known.
The following warning again applies: judicial review is not an appeal and is
conducted exclusively by superior courts. It is, above all, a review of the
lawfulness of the administrative tribunal’s decision, whereas the appeal relates
to the merit and opportunity to make a different decision from that of the tribunal
of first instance. Characterizing a decision as “unreasonable”, thus allows a court
of law, in the absence of a statutory appeal, to ensure that administrative
tribunal respects the law and that a palpable injustice will not be committed, so
that the legal remedy is to return the matter to the administrative
decision-maker and not to make the decision that should have been made.
[18]
The role of a reviewing court is by definition
limited. It is well settled by case law. Its analysis will be concerned with the
existence of justification, transparency and intelligibility within the
decision-making process, and also with whether the decision falls within a
range of possible, acceptable outcomes which are defensible in respect of the
facts and law (Dunsmuir, above at para 47). It even goes so far as
affording great deference to the interpretation that the administrative
tribunal may make of its home statute when it does not concern jurisdiction or
a question of law of central importance for the system. This does not refer to
a decision that another decision-maker with knowledge of the same facts and the
applicable law may have made, but only to a “reasonable” decision—even if it is
not the best decision in the circumstances and it opens itself to criticism. Further,
this is the test that this Court applies to findings of fact, and mixed fact
and law, of the RPD in cases where there was no appeal to the RAD (see for example
Vitalis v Canada (Citizenship and Immigration), 2014 FC 723 at paras 3-4;
Kotai v Canada (Citizenship and Immigration), 2013 FC 693 at para 10;
Wei v Canada (Citizenship and Immigration), 2012 FC 854 at paras 39-41).
It is also the standard of review applied by Member Gallagher in this file and by
other members of the RAD up to now. This first approach was rejected by the Court
as we will see below, but before examining my colleagues’ decisions, let us examine
the second approach, which is also applied throughout Canada by the general courts of appeal.
[19]
Indeed, given that the appeal is not a new trial,
it must be asked what is the standard of review applicable on appeals with
respect to the trial judge’s findings. The standard of review applicable to
pure questions of law is that of correctness, while following the standard of
review applicable to findings of fact, these findings can only be overturned if
it is established that the trial judge committed a palpable and overriding
error: Housen v Nikolaisen, [2002] 2 S.C.R. 235, 2002 SCC 33 (Housen).
Therefore, in Housen, the Supreme Court of Canada clearly indicated
that, in principle, it is not the role of appellate courts
to second-guess the weight assigned by a trial judge to
the various items of evidence and it is only when the inference-drawing process
itself is palpably in error that an appellate court can
interfere with the trial judge’s finding of fact. In
passing, the same high degree of deference should apply to the trial judge’s assessment
of the credibility of a witness and the other findings of fact (Housen, above
at paras 23 and 24).
[20]
More specific information as to the scope of the
standard of palpable and overriding error was later provided by the Supreme
Court of Canada in HL v Canada (Attorney General), 2005 SCC 25 (HL).
Justice Fish, writing for the majority, explained that the standard of palpable
and overriding error applies to all the trial judge’s findings of fact, including
those relating to credibility, the facts proven directly, the facts inferred
and the overall assessment of the evidence.
[21]
However, Justice Fish also raised reasonableness
in the implementation of the standard of review:
[55] … [A]n appellate court will not interfere
with the trial judge’s findings of fact unless it can plainly identify the
imputed error, and that error is shown to have affected the result.
[56] In my respectful view, the test is met as
well where the trial judge’s findings of fact can properly be characterized as “unreasonable”
or “unsupported by the evidence”. …
[A]s a matter of principle, it seems to me
that unreasonable findings of fact—relating to credibility, to primary or
inferred “evidential” facts, or to facts in issue—are reviewable on appeal
because they are “palpably” or “clearly” wrong. The same is true of findings
that are unsupported by the evidence. I need hardly repeat, however, that
appellate intervention will only be warranted where the court can explain why
or in what respect the impugned finding is unreasonable or unsupported
by the evidence. And the reviewing court must of course be persuaded that the
impugned factual finding is likely to have affected the result. (HL, above,
at paras 55-56; emphasis in original).
[22]
Justice Fish also noted that:
Not infrequently, different inferences may
reasonably be drawn from facts found by the trial judge to have been directly
proven. Appellate scrutiny determines whether inferences drawn by the judge
are “reasonably supported by the evidence”. If they are, the reviewing court
cannot reweigh the evidence by substituting, for the reasonable inference
preferred by the trial judge, an equally—or even more—persuasive inference of
its own (HL, above at para 74).
[23]
“Palpable and overriding” error was also defined
as follows by the Court of Appeal of Quebec in Parizeau c Barreau du Québec,
2011 QCCA 1498 at para 91 (Parizeau):
[translation]
Palpable and overriding error is an error that,
in its undeniability – and therefore not a difference of opinion on the
assessment of the evidence – determines
the outcome of the dispute in that the conclusion of the trier of fact, i.e. the
result of his or her decision, cannot hold, thus, ipso facto, making the
decision unreasonable.
[24]
The Court of Appeal of Quebec had previously provided
the following explanations in Regroupement des CHSLD Christ-Roy (Centre
hospitalier, soins longue durée) c Comité provincial des malades, 2007 QCCA
1068 at para 55:
[translation]
An error in the determination of a fact in
issue is only palpable if its immediacy or obviousness is clearly identified in
the reassessment of the relevant part of the evidence and a different finding
on this fact in issue ought to be mafa necessary basis in fact, thereby leading
to a false operative part of the trial decision and requiring the reworking of
this part for that reason.
[25]
According to my understanding, a palpable and
overriding error is thus a clear and discernible error that has an important influence
on the outcome of the dispute. I now come to the case law of this Court on the
scope of the RAD review on appeal.
[26]
The first decision on this topic was made on
May 26, 2014, in Iyamuremye, above. In this decision, Justice
Shore indicated that the role of the RAD is not that of judicial review, but
rather of appeal and that the RAD had to consider all the evidence and make its
own assessment (at paras 30-38). However, on questions of fact, Justice Shore indicates that a certain degree of deference is required, which corresponds
to the standard of “reasonableness” (at paras 2 and 40). It is also a question
of “palpable and overriding error” (at paras 1 and 39).
[27]
This is how Justice Shore summarizes his
thinking on the subject:
[39] The Court recognizes that it would be
absurd, and contrary to subsection 110(3), to task the RAD with re-examining,
for every instance, whether the claimants are in fact refugees or persons in
need of protection within the meaning of sections 96 and 97 of the IRPA. It
is clear from the case law that an appellate body cannot substitute its own
reasoning for that of a specialized tribunal of first instance, the tribunal of
fact, having the advantage of having heard viva voce testimony and with its
authority conferred by the Inquiries Act, unless the trial judge made a “palpable
and overriding error” that led to an erroneous result …
[40] In this case, the Court is entirely in
agreement with the RAD that the standard of review to be applied to findings of
fact made by the RPD is reasonableness. It is well established that an
appellate body must review the findings of a tribunal of first instance by
applying a correctness standard to findings that involve questions of law, and
applying a reasonableness standard to those involving questions of mixed fact
and law …
[41] That said, the Court finds that in
assessing the reasonableness of the decision, the RAD should, at the very
least, have reviewed the evidence that was presented before the RPD and
conducted an independent assessment of all of the evidence in order to
determine whether the RPD, on the basis of the facts and the conditions of the
country in question, had properly considered the evidence and reasonably
justified its conclusion (Dunsmuir, above; Newfoundland and Labrador
Nurses’ Union, above; Alberta Teachers’ Association, above). According
to this trio of judgments by the Supreme Court of Canada, the RAD cannot exempt
itself from considering the evidence as a whole. [Emphasis added]
[28]
In this first decision, Justice Shore avoided characterizing the new action before the RAD as an “appeal de novo”, such
as for appeals brought before the Immigration Appeal Division (IAD) under
section 67 of the IRPA. That said, he nevertheless finds that “this limitation in no way diminishes the jurisdiction conferred
upon the RAD to review the evidence that was before the RPD” (at para
35). It follows that to determine whether the RPD’s decision is “reasonable”, the
RAD must consider the evidence presented to the RPD and conduct “an independent assessment of all of the evidence” (at paras 3
and 41). In this regard, it is a variation of the third approach proposing a
composite and variable standard of review (Huruglica, above). I will
return to this vital point below, given that it is a finding of law that is not
shared by certain members of the Court, including Justice Roy, who wanted to avoid
the problem resulting from any [translation]
“blurring of lines” (Spasoja, above at para 21).
[29]
In Eng and Garcia Alvarez, above, decided
at the same time on July 17, 2014, Justice Shore returned to his analysis
of the law and this time he breathed new life into the standard of palpable and
overriding error that had only been raised in Iyamuremye, above, as we
have seen before. Basing himself on what the Court of Appeal of Quebec wrote in
Parizeau, above, and by analogy in the scheme of the IAD, Justice Shore reiterated
that the RAD cannot be assimilated to a judicial review body and therefore must
not apply the standard of judicial review (Eng, above at paras 23 to
28; Garcia Alvarez, above, at paras 22 to 27). While acknowledging that “an appeal before the RAD is not an appeal de novo”
(Garcia Alvarez, above at para 25; Eng, above at para 26),
Justice Shore explains further:
[29] … When analyzing a decision of the RPD,
the RAD must not merely determine whether it was made in a reasonable manner,
but, rather, analyze whether the RPD relied on a wrong principle of law or
misassessed the facts to the point of making a palpable and overriding error (Housen,
above).
[30] “Palpable and overriding error” is often
used interchangeably with the “clearly wrong” or “unreasonable” decision test.
However, this is the appellate-level standard of intervention that a
specialized appeal tribunal such as the RAD must apply when reviewing a
decision and not the judicial review standard of reasonableness. Even though
there are similarities, these standards are different. [Emphasis added]
[30]
Therefore, it may be said that Justice Shore is no longer “entirely in agreement with the RAD that the
standard of review to be applied to findings of fact made by the RPD is
reasonableness” (Iyamuremye, above at para 40). The result is
that the RAD had committed a reviewable error by using the standard of review of
reasonableness rather than that of palpable and overriding error. In summary, according
to Justice Shore, the RAD must conduct its own assessment of all the evidence, but
in doing so, it still owes some deference to the RPD’s findings of fact, because
it is the tribunal of first instance and it has the advantage hearing testimony
viva voce (Eng, above at para 34; Garcia Alvarez, above at
para 33). In passing, in Garcia Alvarez and Eng, above, Justice Shore noted that:
The idea that the RAD may substitute an
original decision by a determination that should have been rendered without
first assessing the evidence is completely inconsistent with the purpose of the
IRPA and the case law dealing with the virtually identical wording of
subsection 67(2) (Garcia Alvarez, above at para 33; Eng,
above at para 34)
[31]
However, there is a different way to address the
scope of the RAD’s appellate review. For the purposes of being well understood,
I spoke earlier of “a third approach”. In Huruglica, above, made on August
22, 2014, Justice Phelan had previously arrived at the same result as Justice Shore in Eng and Garcia Alvarez, above: “[The RAD]
should have done more than address the decision from the perspective of “reasonableness”.
Therefore, the matter will have to be referred back” (para 56). Like Justice Shore, Justice Phelan noted that the RAD is not a judicial review body; therefore,
it must not limit itself to assessing the reasonableness of the RPD’s decision
(paras 35 to 49). And if a comparison must be made with other administrative
appeal schemes, that of the IAD is the most relevant (paras 50 to 53). In this
sense, the thinking of Justice Phelan concurs with that of Justice Shore.
[32]
I come to the key element of Justice Phelan’s reasoning.
According to my colleague, the RAD’s power of intervention in relation to the
RPD’s findings of fact is more important than the mere power to intervene when
there is a palpable and overriding error:
[54] Having concluded that the RAD erred in
reviewing the RPD’s decision on the standard of reasonableness, I have further
concluded that for the reasons above, the RAD is required to conduct a hybrid
appeal. It must review all aspects of the RPD’s decision and come to an
independent assessment of whether the claimant is a Convention refugee or a
person in need of protection. Where its assessment departs from that of the
RPD, the RAD must substitute its own decision.
[55] In conducting its assessment, it can
recognize and respect the conclusion of the RPD on such issues as credibility
and/or where the RPD enjoys a particular advantage in reaching such a
conclusion but it is not restricted, as an appellate court is, to intervening
on facts only where there is a “palpable and overriding error”. (Huruglica,
above at paras 54-55).
[33]
It is this more nuanced approach that was
recently endorsed by Justice Locke in two decisions made on September 10, 2014:
Njeukam v Canada (Citizenship and Immigration), 2014 FC 859 (Njeukam),
and Yetna v Canada (Citizenship and Immigration), 2014 FC 858 (Yetna).
Justice Locke stated that:
[14] Except in cases where the credibility of
a witness is critical or determinative or when the RPD has a particular benefit
from the RAD to draw a specific conclusion, the RAD must not give any deference
to the analysis of the evidence made by the RPD: see Huruglica, at
paras 37 and 55. The RAD has as much expertise as the RPD and maybe more
with respect to the analysis of the relevant documents and the representations
from the parties. (Njeukam, above at para 14; see also Yetna,
above at para 17).
[34]
In short, we may speak above of a composite and
variable standard of review flowing from a “hybrid appeal” and the particular
nature of the questions of fact, or mixed fact and law. For example, in Njeukam,
above, where it was a pure question of credibility (the RPD’s non-credibility
finding was based on the testimony of the refugee claimant herself), the RAD had
to show deference (Njeukam, above at paras 18-20). Therefore, the
use of the standard of reasonableness was not determinative; Justice Locke dismissed
the application for judicial review. However, in Yetna, above, Justice Locke
determined that, since some findings of the RPD on the credibility of the
applicant were not based only on her testimony, but also on evidence in the
record, the RPD enjoyed no advantage over the RAD when it made its findings and
the RAD had to reconsider the evidence on this topic (Yetna, above at paras 21-25).
The use of the standard of reasonableness proved fatal and the application for
judicial review was allowed by the Court.
[35]
I pause here before continuing this analysis of case
law. The hearings in Njeukam and Yetna took place on July 15
and 16, 2014, before the Eng and Garcia Alvarez decisions were
published. The parties did not argue that the decisions of Justice Shore or the standard of palpable and overriding error should be applied. In Yetna,
counsel for the applicant instead argued the standard of the “wrong” decision. However,
the hearings also took place before Huruglica, which Justice Locke applied
with different results, was decided. Nevertheless, nothing indicates that
Justice Locke requested or received additional arguments from the parties regarding
Eng, Alvarez Garcia or Huruglica.
[36]
This application for judicial review was heard
on September 24, 2014. To my knowledge, the most recent decision on the
scope of the RAD’s appellate review was that made a day earlier in Spasoja,
above. Justice Roy favoured the application of the standard of palpable and
overriding error and set aside the first and third approaches described above. After
reviewing the legislative scheme governing the RAD, he found that it leaves no
room for the RAD to show deference to the RPD:
[translation]
The legislative scheme does not give any indication
that deference was considered by Parliament. Rather, we are facing a scheme where
if there is, for example, an error of fact, the matter must be returned to the RPD
if a reassessment of the evidence before the RPD is required for a decision. There
is no room for deference in such a scheme. (Spasoja, above at para 24).
[37]
As for the ultimate result, Justice Roy stated
that he shared the view of Justice Phelan in Huruglica, above, and the
finding that Justice Shore made in Garcia Alvarez and Eng, above
(Spasoja at para 12). Like Justice Shore, Justice Roy drew a parallel
between the RAD’s scheme and the legislative scheme reviewed in Parizeau,
above, and, after analyzing the legislative debates relating to the creation of
the RAD, he found that it exercises an appellate function and jurisdiction, not
a role of judicial review. However, Justice Roy indicates that an appeal does
not mean a new trial or a reconsideration of the matter in its entirety and
that the standard applicable during any appeal should also be applicable to RAD
appeals:
[translation]
[39] If the appeal discussed in sections 110 and
111 of the Act must be dealt with as an appeal and not a quasi-judicial review,
this does not mean that it will be an opportunity for a new trial or a
reconsideration of the matter in its entirety. The Court of Appeal of Quebec makes a very attractive proposition in Parizeau, above, that the appeal of an
administrative decision before another administrative tribunal should be treated
like any other appeal:
[81] The Supreme Court and our court have
consistently repeated the following instruction: the appeal tribunal may, in
principle, correct any error of law in the decision under appeal or any palpable
and overriding error in the determination of the facts or in the application of
law (if it was correctly determined) to the facts. This standard is just as
valid for the appeals filed with administrative tribunals and the standard for
intervention developed in matters of judicial appeal may certainly be transposed
to the quasi-judicial appeal, with reservations and modifications required by
the particular law of each case and the general rules of administrative law.
The
error of fact must be palpable and overriding to succeed on appeal. The standard
of correctness prevails for questions of law. I do not see why it should not be
so in an administrative appeal.
[40] My colleague Justice Phelan would have preferred
in Huruglica, above, to apply the standard of reasonableness to questions
of credibility (para 37). With respect, I am still concerned with the blurring
of lines. It seems to be preferable to focus on the standard of palpable and
overriding error on appeals on questions of fact. There is nothing a new in
proposing that an appeal tribunal show deference when a body whose decision is
being appealed flows from considerable discretion such as assessing credibility.
The law is clear: the RAD does not hear witnesses except in very exceptional
and specific cases. The credibility to be given to the witnesses heard by the RPD
is its responsibility and the RAD, on appeal, must show deference (Lensen v
Lensen, [1987] 2 S.C.R. 672; R v Burke, [1996] 1 S.C.R. 474). (Spasoja,
above, at paras 39-40; emphasis added).
[38]
Therefore, Justice Roy found that the standards
applicable by the RAD are the standards applicable on appeal, i.e. the standard
of palpable and overriding error for the RPD’s findings of fact, while the RAD must
correct any error of law by the RPD that is determinative.
[39]
It is not difficult to imagine that the choice
of standard of review may have a direct impact on the result of the appeal. Moreover,
there are currently three approaches. Only the first was categorically rejected
by the Court. This leaves two others that seem to wish to compete. Therefore,
both counsel who argued this matter before the Court were particularly bothered
by the Court’s questions.
[40]
By relying in particular on Justice Phelan’s
decision in Huruglica, above, counsel for the applicant argued that, in
applying the standard of reasonableness, Member Gallagher had not fulfilled his
mandate since he did not conduct his own legal analysis of the facts and the evidence.
It argued that it was also what the Court had decided in Garcia Alvarez,
Eng and Huruglica, above, even if the judges had referred to different
tests.
[41]
Counsel for the respondent indicated that his mandate
was clear: the Court’s prior case law simply should not be followed. Because of
its expertise in the field, it was reasonable for the RAD to apply the standard
of judicial review of reasonableness. Full stop. Counsel for the respondent also
argued that Justice Shore, in Iyamuremye, above, had erred in law by
indicating that the RAD had to conduct an independent assessment of all the evidence.
That was also the case with Justice Phelan, in Huruglica, above, when he
decided that the RAD was required to hear the matter as a “hybrid appeal”. In
the alternative, counsel for the respondent argued that the standard of “palpable
and overriding error” should be applied, if the first approach, i.e. the reasonableness
standard of judicial review , is not used.
[42]
As we said above, standards of review are judicial
creations. Nonetheless, once established, they must be respected by the courts just
as any other rule of law. And what gives the rule of law its precedence is its universality:
It applies equally to all; there is no place for judicial or administrative
discretion. Further, following the principle of judicial comity and, unless certain
exceptions apply, a judge of this Court should not deviate from decisions made
by his or her colleagues to avoid creating a situation of uncertainty in the
law. It would be expected that the principle of judicial comity is particularly
important in immigration matters, since under the IRPA, decisions of this Court
may uniquely be subject to an appeal to the Federal Court of Appeal if a question
of general importance is certified. Therefore, it is desirable to have some consistency
in the Court’s decisions. Yes, the judge can make the law, but when each judge
makes his or her own law, the rule law that is no longer applied withers. To
use descriptive language, the law loses weight and this unbearable lightness of
being makes it irrelevant, leaving more room than is needed for administrative
or legal discretion.
[43]
Today, I feel as though I face an Olympic dilemma:
I am asked to choose between two contradictory approaches (the first approach having
been eliminated) adopted by colleagues of the Court. As Justice Wilson stated
in Re Hansard Spruce Mills Ltd, [1954] 4 DLR 590 (BCSC), the statements
of which were recalled by this Court in Alfred v Canada (Minister of
Citizenship and Immigration), 2005 FC 1134 at para 15:
I have no power to overrule a brother Judge, I
can only differ from him, and the effect of my doing so is not to settle but
rather to unsettle the law, because, following such a difference of opinion,
the unhappy litigant is confronted with conflicting opinions emanating from the
same Court and therefore of the same legal weight.
[44]
In Almrei v Canada (Citizenship and
Immigration), 2007 FC 1025, Justice Lemieux summarized the possible
exceptions to the principle of judicial comity:
[61] The principle of judicial comity is
well-recognized by the judiciary in Canada. Applied to decisions rendered by
judges of the Federal Court, the principle is to the effect that a
substantially similar decision rendered by a judge of this Court should be
followed in the interest of advancing certainty in the law …
[62] There are a number of exceptions to the
principle of judicial comity as expressed above they are:
1. The existence of
a different factual matrix or evidentiary basis between the two cases;
2. Where the issue
to be decided is different;
3. Where the
previous condition failed to consider legislation or binding authorities that
would have produced a different result, i.e., was manifestly wrong; and;
4. The decision it
followed would create an injustice.
[45]
I repeat: the principle of judicial comity aims therefore
to prevent the creation of conflicting lines of jurisprudence and to encourage
certainty in the law. Generally, a judge should follow a decision on the same
question of one of his or her colleagues, unless the previous decision differs in
the facts, a different question is asked, the decision is clearly wrong or the application
of the decision would create an injustice. Judicial comity requires much humility
and mutual respect. If the rule of law does not tolerate arbitrariness, judicial
comity, its loyal companion, relies on reason and the good judgement of each
person. Failing a final judgment from the highest court, respect for the other’s
opinion can speak volumes. In short, judicial comity is elegance incarnate in
the person of the magistrate who respects the value of precedents.
[46]
From the point of view of the facts, this case does
not call for a reconsideration of the principles of law applicable or for a new
interpretation of sections 110 and 111 of the IRPA. The issue is always the
same question of law or of jurisdiction relating to scope of the appellate
review conducted by the RAD. Further, it cannot be said that Justices Shore, Phelan or Roy did not thoroughly canvas the relevant law or case law. However, they
seemed to interpret the law differently and contradictorily, as we saw above. Let
us begin with the simplest. The learned counsel for the respondent did not
succeed in persuading me that the exclusion by my colleagues of the first approach
discussed above is clearly wrong in law. For the same reasons as my colleagues,
I conclude that in applying the standard of judicial review of reasonableness, Member
Gallagher committed a reviewable error in law: the RAD is an appellate body, not
a judicial review body and it must not assess its role based on criteria of judicial
review. The matter must be returned to the RAD. Now, let us discuss the rest. I
chose the middle path, that of wisdom: allow the RAD to apply the second or third
approach as long as the question of the scope of the appellate review of the
RPD’s decisions has not been settled by a final judgment by the Federal Court
of Appeal or the Supreme Court of Canada. Quite simply.
[47]
In his appeal before the RAD, I note that the
applicant disputes the findings of generalized risk and implausibility. Counsel
have taken before me contradictory positions regarding the qualification and treatment
to be given to these two findings of fact. There is no agreement on either side
of the very notion of credibility, or on the applicable test for the inferences
that may be drawn by the RPD from the documentary evidence. The difficulty
today is that neither counsel was really able to explain to the Court how the application
of the test of palpable and overriding error or of hearing the matter as a
hybrid appeal translate in practice in this case. Because of its expertise, as
a specialized appeal tribunal, I feel that it is the kind of question that the RAD
can easily settle and should have the chance to decide, preferably before the Federal
Court of Appeal or the Supreme Court of Canada do (without having perhaps the benefit
of knowing the RAD’s point of view).
[48]
In Xie v Canada (Minister of Employment and
Immigration) (1994) 75 FTR 125 (FC) (Xie), the Court stated that
even if it had the jurisdiction to refer the case back for determination in
accordance with such directions as it considers to be appropriate, the Court should
leave the specialized tribunals the right to exercise their jurisdiction and
make decisions on merit based on the evidence before them (at para 18). Therefore,
it is only in exceptional cases that the Court will give directions equivalent to
a directed verdict and this power will only be rarely used when the issue is of
a factual nature (Xie, above at para 18; Canada (Minister of Human
Resources Development) v Rafuse, 2002 FCA 31 at para 14).
[49]
Moreover, the RAD has the power to standardize
the law by decisions that have the value of precedent. In this regard, section
163 of the IRPA provides that before all divisions of the Immigration and
Refugee Board (IRB), matters are heard by a single member unless the Chairperson
considers it necessary to designate a three-member panel. According to the IRB
Web site, when it concerns the RAD, the Chairperson will constitute a
three-member panel if at least one of the following criteria is met:
• The appeal raises unusually complex or
emerging legal issues.
• The appeal raises an issue in an area in
which there is significant divergence or inconsistency in decision-making at
either the RAD or the RPD.
• The appeal raises a serious question of
general importance.
• The appeal raises an issue that may have
a significant impact on practice and procedure at either the RAD or the RPD.
• Any other relevant circumstances exist
that make it appropriate that a three-member panel be designated.
[50]
Member Gallagher, who made the impugned decision
in January 2014, cannot be criticized for not considering the case law of the
Court cited above. It simply did not exist. However, in this file and in the
future the RAD should reassess the scope of the review that it conducts on
appeal in light of the precedents of the Court, which propose two different approaches.
That said, nothing prevents the Chairperson of the IRB to constitute a three-member
panel of the RAD to hear the case, which could confer the value of precedent on
the RAD decision under paragraph 171(c) of the IRPA, and on the approach
used.
[51]
For practical reasons, if we wish to be pragmatic,
our Court may have to resolve—so long as the Federal Court of Appeal and the
Supreme Court of Canada have not ruled on the issue—not to intervene in judicial
review when the RAD applies the second or third approach. That is what the
Court had resolved to do in the citizenship appeals while the citizenship judges
could opt for either of the three approaches developed by Justices Muldoon and
Reed, and Associate Chief Justice Thurlow (as he then
was) (Lam v Canada (Minister of Citizenship and Immigration),
1999 CanLII 7776 (FC), 164 FTR 177, 87 ACWS (3d) 432). In this case, since it
is impossible for the RAD to simultaneously follow both approaches, there would
be chaos if each judge of this Court adopted one of the two approaches and
referred back to the RAD the decisions that applied the other approach instead.
[52]
In our hierarchical justice system, the RAD must
give effect to the judgments of this Court and it should wait for the urbi
et orbi benediction of the Federal Court of Appeal or of the Supreme Court
of Canada itself, before imposing any desire to apply the reasonableness standard
of judicial review in any appeal of an RPD decision. Prudence is required, I do
not believe that it would be wise on my part to speculate on this topic, or to express
a final opinion regarding the scope of the RAD’s appellate review. If it can be
said that the first approach became moot (the standard of judicial review of reasonableness
was rejected by the Court), the person who could predict today which of the
other two approaches would finally prevail would be quite clever; especially
since the situation could again change as other judges choose to rule on the
issue.
[53]
In conclusion, whether it is a question of
jurisdiction or a question of interpretation falling within the RAD’s specialized
expertise, in this case, the applicant did not have the appeal that he is
allowed under the IRPA since the RAD chose to apply a standard of review of reasonableness
corresponding to an application for judicial review, such that this application
for judicial review must be allowed and the matter returned to the RAD for redetermination
of the appeal (Spasoja, above at paras 2, 3 and 47).
[54]
Counsel for the respondent proposes that the
Court certify the following question: [translation] “Considering
the legislative framework of the RAD, what is the scope of the review conducted
by the RAD on appeal when it considers an appeal of an RPD decision? Specifically,
in what circumstances must the RAD show deference to the RPD’s findings?”
[55]
Counsel for the applicant objects to the
certification of a question. Indeed, the question of the scope of the review
conducted by the RAD on appeal is essentially a question of jurisdiction or of
law and we must avoid a multiplicity of appeals raising the same questions.
[56]
In this case, the same question was proposed by the
respondent to Justice Phelan in Huruglica, above, and after the Court’s
verification, he certified it in the following words: [translation] “What
is the scope of the review conducted by the Refugee Appeal Division when it considers
an appeal of a decision of the Refugee Protection Division?” The
certification of the question proposed by counsel for the respondent is thus redundant.
Counsel for the respondent did not persuade me in this case that it would be in
the interest of justice to certify the question proposed above or that the
particular factual background of this case justifies the multiplicity of
concurrent appeals that arise from substantially the same questions of law or
of jurisdiction regarding the interpretation of the scope of the review
conducted by the RAD on appeal.
[57]
As the Federal Court of Appeal stated in Varela
v Canada (Citizenship and Immigration), 2009 FCA 145, the provision
preventing any appeal to the Federal Court of Appeal unless a question is
certified “fits within a larger scheme designed to ensure
that a claimant’s right to seek the intervention of the courts is not invoked
lightly, and that such intervention, when justified, is timely” (at para
23). Certifying the same question in all RAD files subject to judicial review with
this Court would certainly not allow “timely”
interventions.
[58]
No question will be certified in this file by the
Court.