Principles, Practices, Fragile Promises: Judicial
Review of Refugee Determination Decisions
Before the Federal Court of Canada
Mary C. Hurley*
This article examines judicial review practice at the Fed-
eral Court of Canada and the concerns it raises in the context of
Canada’s current refugee determination system. Under succes-
sive immigration statutes, the Federal Court of Canada has
played a supervisory role in refugee determination; however,
there is an absence of direct guidance from the Supreme Court of
Canada on the application of judicial review principles in the
refugee determination context. Focusing on the Trial Division’s
judicial review of refugee determination decisions since February
1993, the author concludes from a broad survey of the caselaw
that this practice has been characterized by inconsistency both in
approach and in result The author argues that this inconsistency
raises significant concerns particular to the refugee context be-
cause of the current statutory regime of refugee determination.
It is argued that the interests at stake in refugee determina-
tion set the process apart from the general sphere of administra-
tive adjudication. Canada’s statutory regime for refugee determi-
nation has evolved out of the government’s attempts to respond
to the 1985 Supreme Court of Canada decision in Singh which
determined that refugee claimants are entitled to oral hearings. It
is also a response to the rapid and significant increases in the
number of refugee claimants. The refugee determination system
has undergone a number of significant reforms which have
eroded procedural protections given to refugee claimants, abol-
ishing appeal of C.1RD.D. decisions and transferring judicial re-
view authority over all immigration decisions to singlejudges of
the Federal Court, Trial Division. In March 1995, the govern-
ment announced that it will reduce refugee determination panels
from two members to one.
The concerns raised by gaps in procedural and substantive
protection under the current Canadian refugee determination
system cannot be adequately addressed by the consistent appli-
cation of a severe judicial review standard. Given the inherent
limitations of judicial review, the author concludes that a sub-
stantive internal review mechanism is needed to redress inade-
quacies and to comply with Canada’s commitment to the princi-
ple of non-refoulement stated in the United Nations Convention
Relating to the Status of Refugees.
Cet article examine la pratique de Ia rdvisionjudiciaire s ]a
Cour f6drale du Canada et les questions qu’elle soulhve dans le
contexte du systhme de determination des rdfugi6s au Canada. La
Cour fitl rale ajou6 un rfle de supervision dans la determination
des rfugids en vertu des lois habilitantes successives en matire
d’immigration, malgrd l’absence de directives de la Cour su-
preme concemant l’application des principes de rivision judi-
claire dans co contexte. L’auteure concentre ses propos sur les
decisions de rvision judiciire de la division de premihre ins-
tance portant sur la d6termination des r6fugids depuis le mois de
f6vrier 1993. En s’appuyant sur un vaste 6chantillonnage de ju-
risprudence, elle conclut que cette pratique de revision judiciaire
est incohdrente tant dana son approche que dans son rsultat.
L’auteure estime que cette incohdrence cr6e des inquiltudes de
taille dans le contexte de la determination des r6fugids, dtant
donne le rgime statutaire actual.
L’auteure soutient que les intildts en jeu dans la determi-
nation des rfugila rent un processus h l’extrieur de Ia sphhe
globale de l’adjudication administrative. Le rgime statutaire
pour Ia determination des rfugi6s a dvolu6 suite aux tentatives
du gouverement de rpliquer i la d6cision de la Cour supreme
dans Singh, qui a 6eabli que les requ6rants avaient le droit d’etre
entendus oralement. Le syst~me de determination des rfugids a
connu un nombre important de rformes qui oat drode les pro-
tections procidurales accordiles aux requ6rants, notamment en
abolissant
Ia Commission de
l’immigration et statut des rfugids et en transf6rant l’autofit6 de
Ia rvision judiciairce de
les d6cisions en matire
d’immigration A la Cour flderale. En mars 1995, le gouveme-
ment a annonc6 qu’il riuirait les comits de d6termination des
rdfugids de deux hun membre.
l’appel des d6cisions de
toutes
Les inquiltudes soulevles par des lacunes dans les garan-
ties proc6durale et substantive du systbme actuel de determina-
tion des rfugi6s ne peuvent Eire &cartdes par l’application uni-
forme de normes strictes en matihue de rvision judiciaire. ttant
donne les limites inhdrentes ii ]a rvision judiciaire, l’auteure
conclut qu’un mcanisme substantif de r6vision interne est requis
afin de palier aux
l’engagemeant di Canada au principe de non-refoulement dinonc
dans la Convention relative au statut des rifugiis des Nations
uraes.
insuffisances et de se conformer
* B.A. (1966) (McGill); LL.B. (1987) (Montreal); LL.M. (1995) (Ottawa). The author gratefully ac-
knowledges Peter Showler’s patience and humour throughout his supervision of an earlier version of
this paper and thanks Professor David Mullan for the helpful comments that guided revisions to it.
The support of friends and family on both sides of the Atlantic has been a constant boon.
McGill Law Journal 1996
Revue de droit de McGill
To be cited as: (1996) 41 McGill L.J. 317
Mode de r6f6rence: (1996) 41 R.D. McGill 317
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
Synopsis
Introduction
I. The Principle: The “Reasonable Bear”
A. C.U.P.E.’s Wake
B. Dogma, Dissent and Still No Definition
C. The Supreme Court and Refrgee Determination
II. The Practice: The Reasonable Refugee
A. The Federal Court of Canada and Refugee Determination
B. Federal Court of Appeal: What’s a C.U.P.E.?
C. Federal Court Trial Division: Find the “Whopper”?
The February 1993 – November 1994 Period
The December 1994 – October 1995 Period
1.
2.
m. What’s Wrong With This Picture?
A. Causes
B. Cause for Concern
1.
2.
The Random Solution
The Very Severe Solution
IV. The Promise: Dusting Off a Modest Proposal
A. The Tinkering Approach
B. The Body-Work Approach
1.
2.
Judicial Appeal on the Merits
Internal Appeal
Conclusion
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
I
lOin prisente gingralement I’enjeu fondamental du droit admi-
nistratif moderne comme itant l’ quilibre entre les valeurs judi-
ciaires et les valeurs administratives plut6t que l’harmonisation
entre les valeurs de liberti individuelle et celles de justice so-
ciale.’
R. Dussault & M. Patenaude
Introduction
Judicial review principles attempt to prescribe the interaction of administrative
and judicial institutions. This paper is concerned with judicial review practice un-
der the present refugee determination system in Canada. The discussion proceeds
from the premise that the interests at issue in the refugee determination process set
it apart within the general sphere of administrative adjudication. Canada’s 1969
signing of the United Nations Convention Relating to the Status of Refugee? and
this country’s commitment to the principle of non-
its protocol
refoulement, or non-return, which represents the Convention’s main obligation.!
That principle entails determining whether individual refugee claimants are entitled
to Canada’s protection by reason of a well-founded ‘fear of persecution owing to
inadequate state protection in their countries of origin By its nature, refugee de-
termination routinely raises human rights issues involving severe suffering and
deprivation.
signified
‘ R. Dussault & M. Patenaude, “Le contrtle judiciaire de L’Administration : Vers une meilleure
synthse des valeurs de libert6 individuelle et dejustice sociale ?” (1983) 43 R. du B. 163 at 166-67.
2 28 July 1951, 189 U.N.T.S. 150, U.K.T.S. 39 (1954) (entered into force 22 April 1954)
[hereinafter Convention].
3 Protocol Relating to the Status of Refugees, 31 January 1967, 606 U.N.T.S. 267, 6 I.L.M. 78
(entered into force 4 October 1967).
4The Convention states:
No Contracting State shall expel or return [“refoule?”] a refugee in any manner what-
soever to the frontiers of territories where his life or freedom would be threatened on
account of his race, religion, nationality, membership in a particular social group or
political opinion (supra note 2 at art. 33, s. 1).
Canada’s Immigration Act, R.S.C. 1985, c. 1-2, as am. by R.S.C. 1985 (4th Supp.), c. 28, s. 1,
provides that:
“Convention refugee” means any person who
(a) by reason of a well-founded fear of persecution for reasons of race, religion, na-
tionality, membership in a particular social group or political opinion,
(i) is outside the country of the person’s nationality and is unable or, by
reason of such fear, is unwilling to avail himself of the protection of
that country, or
(ii) not having a country of nationality, is outside the country of the
person’s former habitual residence and is unable or, by reason of such
fear, is unwilling to return to that country, and
(b) has not ceased to be a Convention refugee … (Immigration Act, ibid. at s. 2(1)).
MCGILL LAw JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
In the 1985 decision of Singh v. Canada (M.E.L), three members of the Su-
preme Court of Canada ruled that the interests at stake in refugee determination
engage section 7 of the Canadian Charter of Rights and Freedoms.7 This finding
has since been widely accepted as stating a constitutional norm.’ The Singh deci-
sion was largely instrumental in triggering legislative reform of the administrative
framework for refugee determination.
Judicial review currently plays a pivotal role in the post-Singh system. In 1993,
amendments to the Immigration Acf vested judges of the Federal Court of Canada,
Trial Division, with original judicial review jurisdiction over decisions of the Con-
vention Refugee Determination Division (“C.R.D.D.” or “Board”) of the Immigra-
tion and Refugee Board (“I.R.B.”). The following discussion focuses primarily on
the exercise of this jurisdiction and implications arising from it.
Part I discusses the evolution of theoretically severe but ultimately indetermi-
nate judicial review principles in Supreme Court of Canada caselaw and the ab-
sence of direct guidance from the Court on their application in the refugee deter-
mination context. Part II outlines the supervisory role played by the Federal Court
of Canada in refugee matters under successive immigration statutes; it distin-
guishes the interventionism characteristic of Federal Court of Appeal decisions
from the deferential posture advocated by the Supreme Court of Canada. The dis-
cussion under this heading focuses on Trial Division judicial review practice in the
area of refugee determination since February 1993 and concludes from a broad
survey of the caselaw that inconsistency of approach and of result has been a
dominant feature of that practice.
Part III considers factors contributing to this outcome, and takes the position
that inconsistency in judicial decision-making in the refugee context raises particu-
lar concerns arising from the current statutory regime’s procedural and institutional
weaknesses. These concerns would not, it is contended, be resolved by more con-
sistent application of an ambiguous and inappropriately severe review standard in
light of ongoing issues of expertise at the C.R.D.D. Part IV discusses remedial op-
6 [1985] 1 S.C.R. 177, 17 D.L.R. (4th) 422 [hereinafter Singh].
7Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11
[hereinafter Charter]. The remaining three members of the Singh Court also found the absence of oral
hearings constituted a denial of fundamental justice under section 2(e) of the quasi-constitutional Ca-
nadian Bill of Rights, S.C. 1960, c. 44, reprinted in R.S.C. 1985, App. III.
‘See e.g.: Weerasinge v. Canada (M.E.L) (1993), [1994] 1 P.C. 330 at 334-35, 161 N.R. 200 (C.A.)
[hereinafter Weerasinge]; Huynh v. Canada (1994), [1995] 1 RC. 633 at 643 (T.D.), on appeal No, A-
658-94 [hereinafter Huynh].
‘An Act to atnend the Innigration Act and other Acts in consequence thereof S.C. 1992, c. 49,
amending R.S.C. 1985, c. 1-2, as amended [hereinafter Bill C-86].
“0 Under Bill C-86, access to judicial review as well as to subsequent appeal
is restricted
(lmmigration Act, ibid. at ss. 82.1, 83, as am. by Bill C-86, ibid. at s. 73). Both aspects are reviewed at
Parts HA and fIM.B. 1, below.
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
tions, concentrating primarily on internal review at the administrative level. Part V
concludes that under the current refugee determination system, judicial review in-
adequacies underscore the need for an administrative appeal mechanism as the best
review alternative consistent with Canada’s commitment under the Convention and
domestic law.
I. The Principle: The “Reasonable Bear”
A. C.U.P.E.’s Wake
Contemporary standards of judicial review are generally traced to 1979, when
the Supreme Court of Canada articulated a new deferential Common law standard
applicable to review of administrative decisions within an agency’s jurisdiction.
The central question in C.U.P.E., Local 963 v. New Brunswick Liquor Co.,” once
the Labour Board’s jurisdiction over the subject matter of its decision had been
confirmed, was whether the interpretation of its governing Statute had been so
“patently unreasonable” that its construction could not be rationally supported. That
is, if the Statute might reasonably bear the Board’s construction, the courts should
not intervene because the Board was entitled to err within its jurisdiction owing to
the protection of a privative clause. This has been coined the “reasonable bear” ap-
proach to judicial review.'” Entitlement to curial deference in C. U.RE. was based on
recognition that the Labour Board was a specialized tribunal administering a com-
prehensive statute with accumulated experience, considerable sensitivity and
unique expertise.
In the immediate post-C. U.RE. era, the Supreme Court extended the patently
unreasonable standard for decisions within jurisdiction to a variety of labour rela-
tions contexts, whether or not the agency in question was protected by a privative
clause.” This extension included agencies subject to the supervisory authority of
the Federal Court of Canada, Appeal Division,” under grounds of judicial review
“[1979] 2 S.C.R. 227,47 D.L.R. (3d) 417 [hereinafter C.U.RE.].
“The reference originates with: H.W. MacLauchlan, “Judicial Review of Administrative Interpre-
tations of Law: How Much Formalism Can We Reasonably Bear?” (1986) 36 U.T.L.J. 343
[hereinafter “Formalism”]; H.W. MacLauchlan, “Developments in Administrative Law: The 1989-90
Term” (1991) 2 Supreme Court L.R. (2d) 1 [hereinafter “Developments”].
” See e.g.: Douglas Aircraft Co. of Canada v. McConnell (1979), [1980] 1 S.C.R. 245, 99 D.L.R.
(3d) 385 (no privative clause); Teamsters Union, Local 938 v. Massicotte, [1982] 1 S.C.R. 710, 134
D.L.R. (3d) 385 (privative clause); A.U.RE., Branch 63 v. Olds College, [1982] 1 S.C.R. 923, 136
D.L.R. (3d) 1 (no privative clause); Bibeault v. McCaffrey, [1984] 1 S.C.R. 176, 7 D.L.R. (4th) 1
(privative clause).
‘” See: Canada (LR.B.) v. I.LA., Local 269, [1983] 1 S.C.R. 245, 144 D.L.R. (3d) 1 [hereinafter
Halifax Longshoremen] (partial privative clause); Fraser v. Canada (Public Service Staff Relations
Board), [1985] 2 S.C.R. 455, 23 D.L.R. (4th) 122 [hereinafter Fraser cited to S.C.R.] (no mention of
privative clause).
MCGILL LAw JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
codified in the Federal Court Act.'” However, the standard was applied inconsis-
tently.” Conceptually, the C.U.P.E. deference principle remained (and remains) in-
extricably linked to the elastic notion of jurisdiction as redefined from case to case;
as a result, the intra-jurisdictional, patently unreasonable standard had no relevance
once an error had been characterized as jurisdictional and thereby subject to review
for correctness.”
Even when adopting and advocating a deferential stance, the Court’s language
allowed for speculation as to the precise standard of review used and criteria rele-
vant to that standard.'” In one frequently-cited decision, the patently unreasonable
test was vaguely described as “a very severe test [that] signals a strict approach to
the question of judicial review”, one that was equally applicable to errors of law
and fact within jurisdiction and that recognized tribunals as having “the necessary
jurisdiction to make a mistake, and even a serious one, but not to be unreason-
able”. 9
B. Dogma, Dissent and Still No Definition
Over the past decade, the Supreme Court has enshrined doctrines of non-
intervention within jurisdiction, on the one hand, and review of excess of jurisdic-
tion, on the other, as the operative dogma of judicial review.”0 Deference to the ex-
‘” R.S.C. 1970 (2nd Supp.), c. 10, s. 28. See also infra note 79.
‘6 See: National Bank of Canada v. R.C.LU., [1984] 1 S.C.R. 269, 9 D.L.R. (4th) 10 [hereinafter
National Bank] (partial privative clause); Yellow Cab Ltd. v. Alberta (Board of Industrial Relations),
[1980] 2 S.C.R. 761, 114 D.L.R. (3d) 427 (incomplete privative clause); C.A.TC.A. v. Canada
(Treasury Board), [1982] 1 S.C.R. 696, 134 D.L.R. (3d) 577 (no mention of privative clause).
” See Syndicat des employds de production du Qudbec et de l’Acadie v. Canada (LR.B.), [1984] 2
S.C.R. 412, 14 D.L.R. (4th) 457 [hereinafter L’Acadie cited to S.C.R.] (partial privative clause; sub-
stantive interpretation upheld as within jurisdiction, remedy overturned as exceeding jurisdiction).
“The Court stated:
A reviewing court, whether under s. 28(l)(b) of the Federal Court Act, or under the
common law principles of judicial review, should not interfere with the decision of a
statutory decision maker in a case … unless the statutory decision maker makes a mis-
take of law, such as addressing his or her mind to the wrong question, applying the
wrong principle, falling to apply a principle he or she would [sic] have applied, or in-
correctly applying a legal principle (Fraser, supra note 14 at 464-65).
Compare the Court’s language in C. URE.’s precursor, SE.I. U., Local 333 v. Nipawin District Staff
Nurses Assoc. (1973), [1975] 1 S.C.R. 382 at 389, 41 D.L.R. (3d) 6, cited as authority for criteria
applicable to patently unreasonable errors in UE.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048 at
1084-85, 95 N.R. 161 [hereinafter Bibeault cited to S.C.R.]. On this point, see J.M. Evans,
“Developments in Administrative Law: The 1985-86 Term” (1987) 9 Supreme Court L.R. (1st) 1 at
61-63.
,9Blanchard v. Control Data Canada Ltd., [1984] 2 S.C.R. 476 at 493-94, 14 D.L.R. (4th) 289.
2″MacLauchlan observes: ‘The case for deference is familiar. decision-making expertise, political
empathy and efficient use of administrative, judicial and regulated parties’ resources” (H.W.
MacLauchlan, “Developments in Administrative Law: The 1990-91 Term” (1992) 3 Supreme Court
L.R. (2d) 29 at 32-33 [hereinafter “The 1990-91 Term”]).
19961
M. HURLEY- REFUGEE DETERMINATION DECISIONS
pertise of administrative decision-makers, as exercised through the patently unrea-
sonable test, has been reasserted as the governing principle in review of labour
board decisions,’ as well as those of other agencies,2 with the former continuing to
dominate the Court’s judicial review agenda. Accordingly, an administrative agency
protected by a privative clause and acting within its jurisdiction –
as determined by
pragmatic and functional analysis? –
is, in theory, entitled to deference provided it
does not err in a patently unreasonable manner.’
Moreover, the Court’s caselaw since C. U.RE. has made it clear that the priva-
tive clause “criterion” stated in that decision does not necessarily determine the
applicable standard of review. In the absence of such protection, curial deference
may be appropriate if the agency under review is highly specialized, and the ques-
tion of law at issue falls squarely within its jurisdiction.’ The concept of relative
21 See: Bibeault, supra note 18; TWU. v. British Columbia Telephone Co., [1988] 2 S.C.R. 564, 54
D.L.R. (4th) 385 [hereinafter TWU.J; C.A.I.M.A.W., Local 14 v. Paccar of Canada Ltd., [1989] 2
S.C.R. 983, 62 D.L.R. (4th) 437 [hereinafter C.A.LM.A.W. cited to S.C.R.]; W.W Lester (1978) Ltd. v.
U.A., Local 740, [1990] 3 S.C.R. 644, 76 D.L.R. (4th) 389 [hereinafter Lester cited to S.C.R.]; Can-
ada (A.G.) v. RS.A.C., [1993] 1 S.C.R. 941, 101 D.L.R. (4th) 673 [hereinafter P.S.A.C. II cited to
S.C.R.]; C.J.A., Local 579 v. Bradco Construction Ltd., [1993] 2 S.C.R. 316, 102 D.L.R. (4th) 402
[hereinafter Bradco cited to S.C.R.].
22 See: National Corn Growers Assoc. v. Canada (Import Tribunal), [1990] 2 S.C.R. 1324, 74
D.L.R. (4th) 449 [hereinafter National Corn Growers cited to S.C.R.]; Domtar Inc. v. Quebec
(Coimnission d’appel en matibre de lesions professionnelles), [1993] 2 S.C.R. 756, 105 D.L.R. (4th)
395 [hereinafter Domtar] (workmen’s compensation board).
‘ The pragmatic and functional approach for determining the limits of a tribunal’s jurisdiction and,
by extension, the scope of deference a tribunal is owed, requires that
the Court [examine] not only the wording of the enactment conferring juris-
diction … but the purpose of the statute creating the tribunal, the reason for its
existence, the area of expertise of its members and the nature of the problem
before the tribunal (Bibeault, supra note 18 at 1088).
This approach has been widely adopted as a basis for analysis in subsequent Supreme Court judicial
review decisions (see: C.A.LM.A.W, supra note 21; P.S.A.C. II, supra note 21; Bradco, supra note 21;
Domtar, ibid.; Canada (A.G.) v. Public Service Alliance of Canada, [1991] 1 S.C.R. 614, 80 D.L.R.
(4th) 520 [hereinafter P.S.A.C. 1]; Universitg du Quibec it Trois-Rivi~res v. Larocque, [1993] 1 S.C.R.
471, 101 D.L.R. (4th) 494; Dayco (Canada) Ltd. v. C.A.W.-Canada, [1993] 2 S.C.R. 230, 102 D.L.R.
(4th) 609 [hereinafter Dayco]; Canada (A.G.) v. Mossop, [1993] 1 S.C.R. 554, 100 D.L.R. (4th) 658
[hereinafter Mossop cited to S.C.R.]; Pezim v. British Columbia (Superintendent of Brokers), [1994] 2
S.C.R. 557, 114 D.L.R. (4th) 385 [hereinafter Pezim cited to S.C.R.]).
24 See: Bibeault, ibid.; TWU., supra note 21; C.A.I.M.A.W., ibid.; Lester, supra note 21; RS.A.C. II,
ibid.
, In addition to the cases cited supra note 13, see e.g. Bradco, supra note 21, in which deference
was shown on the basis of expertise despite absence of a full privative clause. By way of corollary, an
agency theoretically enjoying privative protection from review for correctness is not automatically
entitled to curial deference where its expertise to deal with a given issue is found wanting (see:
Dayco, supra note 23; Chambly (Commission scolaire rigionale) v. Bergevin, [1994] 2 S.C.R. 525,
115 D.L.R. (4th) 609 [hereinafter Bergevin cited to S.C.R.], in each of which a majority of the Court
applied a correctness standard to an arbitrator’s decision protected by a privative clause on the basis
of insufficient expertise). See also C.P Air Lines Ltd. v. C.A.L.RA., [1993] 3 S.C.R. 724, 108 D.L.R.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
expertise has thus become a dominant factor in the Court’s evaluations as to
whether deference is due.” Specialization of functions has also been recognized as
attracting deference in appeal proceedings from decisions of at least certain admin-
istrative agencies.”
Notwithstanding unanimity in the framing of applicable standards of review,
however, Supreme Court judges have disagreed as to the interpretation and appli-
cation of these standards in relation to a broad range of issues, including: jurisdic-
tional interpretation per se,” the appropriate standard to be applied,” whether or not
a given administrative interpretation is in fact patently unreasonable” or incorrect,”
the operation of deference per se” and the scope of review for patent unreason-
ableness” or in appeal proceedings.’ The Court’s division on related points of def-
erence and the scope of review has been particularly problematic for lower courts
in applying the patently unreasonable test. In C.A.I.MA.W, the majority concurred
in finding the labour relations agency’s interpretation reasonable but disagreed as
to how deference works. To La Forest J.,
[c]urial deference is more than just a fiction courts resort to when they are in
agreement with the decisions of the tribunal. Mere disagreement with the result
arrived at by the tribunal does not make that result “patently unreasonable”….
The emphasis should be not so much on what result the tribunal has arrived at,
but on how the tribunal arrived at that result.”
By this approach, it was unnecessary to determine whether the decision under re-
view was correct or to go beyond a finding that it was not patently unreasonable.
(4th) 1 [hereinafter C.R Air], in which a correctness standard was applied in the face of a partial pri-
vative clause because the matter was not “strictly” within the Board’s special fields of expertise. The
C.R Air approach may be contrasted with decisions in Halifax Longshoremen, supra note 14, Na-
tional Bank, supra note 16, and L’Acadie, supra note 17, in all of which the patently unreasonable
standard was used in the face of the same privative clause.
26 See quote accompanying note 49, below. See also infra notes 59-64 and accompanying text on
the question of whether the correctness standard of review applied by the Court to the decisions of
human rights tribunals on the basis of absence of greater expertise on questions of law is capable of
extrapolation to other “human rights” areas, such as refugee determination.
‘ See: Bell Canada v. Canada (C.R.TC.), [1989] 1 S.C.R. 1722, 60 D.L.R. (4th) 682 [hereinafter
Bell Canada]; Pezim, supra note 23. See also text accompanying notes 47-49, below.
See: TWU., supra note 21; PS.A.C. I, supra note 23.
See: RS.A.C. II, supra note 21; Bradco, supra note 21; RS.A.C. I, ibid.; Mossop, supra note 23;
Dayco, supra note 23; Bergevin, supra note 25.
See: C.A.I.M.A.W, supra note 21; Lester, supra note 21.
“See C.R Air, supra note 25.
See C.A.I.M.A.W, supra note 21.
“See: Lester, supra note 21; National Corn Growers, supra note 22.
See: Zurich Insurance Co. v. Ontario (H.R.C.), [1992] 2 S.C.R. 321, 93 D.L.R. (4th) 346
[hereinafter Zurich cited to S.C.R.]; Dickason v. University of Alberta, [1992] 2 S.C.R. 1103, 95
D.L.R. (4th) 439 [hereinafter Dickason].
” C.A.LM.A.W, supra note 21 at 1003-1004.
19961
M. HURLEY- REFUGEE DETERMINATION DECISIONS
Sopinka J., on the other hand, found:
Any adjudication upon the reasonableness of a decision must involve an
evaluation of the merits. Reasonableness is not a quality that exists in isolation.
When a court says that a decision under review is “reasonable” or “patently un-
reasonable” it is making a statement about the logical relationship between the
grounds of the decision and premises thought by the court to be true. Without
the reference point of an opinion (if not a conclusion) on the merits, such a
relative statement cannot be made… Curial deference does not enter the pic-
ture until the court finds itself in disagreement with the tribunal. Only then is it
necessary to consider whether the error (so found) is within or outside the
boundaries of reasonableness. The test is … a “severe test”. But even here an
appreciation of the merits is not irrelevant … So long as the court is satisfied
with the correctness of the tribunals’ [sic] decision, any reference to reason-
ableness is superfluous.”
This divergence in the majority opinion, together with the separate, dissenting rea-
sons of Wilson and L’Heureux-Dub6 JJ., prompted the observation that “[t]he Su-
preme Court is obviously having a great deal of difficulty applying the ‘reasonable
bear’ line of jurisprudence within its own walls.”3
The disagreement expressed in National Corn Growers as to the scope of re-
view permitted by the patently unreasonable test supports this view. There, a ma-
jority of the Court found it appropriate to scrutinize a number of aspects of the Im-
port Tribunal’s decision for patent unreasonableness.” The dissenting justices
found this “meticulous analysis” 9 inappropriate and antithetical to the C. U.PE.
deference principle. For the majority, Gonthier J. observed in response that a court
could not reach a conclusion as to the reasonableness of a tribunal’s interpretation
without considering the underlying reasoning, and that he “would be surprised if
that were the effect of this Court’s decision in C. U.RE.”.’
The Court split along similar lines in Lester, in which the breadth of the ma-
jority’s analysis concluding in patent unreasonableness provoked a similar defence
of the C. U.RE. standard:
It is quite unrealistic in this age of increasingly complex and highly specialized
regulatory regimes to expect the courts to have the requisite knowledge and
skill to adjudicate properly on some of those regimes…. [r]here has been a
tendency in the post-C.URE. era to return to a less stringent test for judicial
review than the one established in C..RE…. That approach to curial review in
36Ibid. at 1018. Arguably, this analysis fails to consider that judicial agreement with an administra-
tive decision makes it neither correct nor reasonable and appears to have little to do with the notion of
comparative expertise, which is supposedly at the root of the deference principle.
37 “Developments”, supra note 12 at 50.
See National Corn Growers, supra note 22 at 1371-83.
3 Ibid. at 1349.
,’ Ibid. at 1383.
MCGILL LAW JOURNAL/REVUE DE DRO1TDE McGILL
[Vol. 41
the administrative context is, in my opinion, no longer appropriate given the
sophisticated role that administrative tribunals play in the modem Canadian
state. I think we need to return to C. U.RE. and the spirit which [it] embodies.4′
A review of post-C. U.P.E. cases reveals that notwithstanding frequent reitera-
tion of the fundamental importance of curial deference to administrative expertise
and despite ample opportunity, the Court has provided little guidance as to the es-
sence of the patently unreasonable standard that mandates deference.” According
to one view, the standard requires assessment
in terms of the reasonable Board. This must be so if we are to allow for the fact
that the Board is deemed to have special expertise. A patently unreasonable
decision is accordingly one which no reasonable Board applying its expertise
could possibly have arrived at.”
Another view, noting that “[w]hat is patently unreasonable to one judge may be
eminently reasonable to another,” considers that deference is called for unless the
tribunal exceeds its jurisdiction by arriving at a “clearly irrational” decision.” Still
another, acknowledging that “[a] patently unreasonable error is more easily defined
by what it is not than by what it is,”” holds that the standard should reflect some-
thing other than the court’s view of the correctness of an agency’s interpretation of
its governing legislation. One authority has observed that in the Lester decision,
“novelty and incorrectness” appeared sufficient to establish patent unreasonable-
ness.
46
In the 1994 Pezim case, the Court deferred to a decision of the B.C. Securities
Commission which was not protected by a privative clause and which was subject
to a statutory right of appeal, based on the “concept of the specialization of duties
requir[ing] that deference be shown … on matters which fall squarely within the tri-
bunal’s expertise”.” The unanimous Court used the narrow question before it as an
“‘ Lester, supra note 21 at 650-51.
, Holloway observes that in the years since C. U.PE., “not a year has gone by when the [Court] has
not been called upon to redefine the standard of administrative judicial review” (. Holloway, “The
Transformation of Canadian Administrative Law” (1993) 6 Can. J. Admin. L. & Prac. 295 at note
110, p. 329).
43 C.A.IMA.W, supra note 21 at 1021, Wilson J. (dissenting).
“RS.A.C. II, supra note 21 at 963, Cory J. See also: Bergevin, supra note 25 at 537, Cory J., at 554,
LHeureux-Dub J.
,5Bradco, supra note 21 at 340-41, Sopinka J.
J.M. Evans, “Jurisdictional Review in the Supreme Court: Realism, Romance and Recidivism”
(1991) 48 Admin. L. R. 255 at 261 [hereinafter “Jurisdictional Review”].
4 Pezim, supra note 23 at 591, relying on Bell Canada, supra note 27. Factors triggering deference
in Pezin included: the breadth of the Commission’s expertise, specialization and discretion, as re-
flected in its governing Statute; its role in policy development within a broad national framework for
the regulation of the securities industry; a tradition of curial deference towards the decisions of secu-
rities commissions; and the nature of the problems before the Commission which lay “at the heart of
[its] regulatory expertise” (Pezimn, ibid).
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
opportunity to consolidate some of its previous expressions of judicial review
principles:
There exist various standards of review with respect to the myriad of adminis-
trative agencies that exist … The central question in ascertaining the standard of
review is to determine the legislative intent in conferring jurisdiction on the
administrative tribunal. … Included in the analysis is an examination of the tri-
bunal’s role or function. Also crucial is whether or not the agency’s decisions
are protected by a privative clause. Finally, of fundamental importance, is
whether or not the questioh goes to the jurisdiction of the tribunal involved.!3
The Court also asserted that in light of the many factors relevant to determining the
appropriate standard of review,
the courts have developed a spectrum that ranges from the standard of reason-
ableness tread: “patent unreasonableness”] to that of correctness. Courts have
also enunciated a principle of deference that applies not just to the facts as
found by the tribunal, but also to the legal questions before the tribunal in the
light of its role and expertise. At the reasonableness end of the spectrum, where
deference is at its highest, are those cases where a tribunal protected by a true
privative clause, is deciding a matter within its jurisdiction and where there is
no statutory right of appeal. … At the correctness end of the spectrum, where
deference in terms of legal questions is at its lowest, are those cases where the
issues concern the interpretation of a provision limiting the tribunal’s jurisdic-
tion (jurisdictional error) or where there is a statutory right of appeal which al-
lows the reviewing court to substitute its opinion for that of the tribunal and
where the tribunal has no greater expertise than the court on the issue in ques-
tion, as for example in the area of human rights.”9
In theory, Pezim thus allows for a range of judicial control within the adminis-
trative law spectrum. However, the Pezim Court also reaffirmed the predominant
role of the C. U.P.E. deference principle – which it found to be applicable in the
narrow Pezim context –
as a general model for virtually all appellate and judicial
review proceedings where jurisdiction per se is not at issue, and where the legisla-
tor has placed a matter squarely within the jurisdiction of an expert administrative
decision-maker.”
Reassertion of that principle does not, however, define criteria for identifying
patently unreasonable errors within jurisdiction for purposes of situating a given
tribunal along the spectrum, and it remains to be seen how the Court will develop
‘ Ibid. at 589-90.
41 Ibid. at 590.
” In a recent decision, the Court was again unanimous in applying Pezim principles to the Canadian
Radio-Television and Telecommunications Commission which, like the B.C. Securities Commission,
has broad policy-making and regulatory authority and, hence, is “entitled to curial deference, even in
the absence of a privative clause and the presence of a statutory right of appeal” when acting within
its area of expertise and jurisdiction (Bitish Columbia Telephone Co. v. Shaw Cable Systems (B.C.)
Ltd., [1995] 2 S.C.R. 739 at 758-59, 125 D.L.R. (4th) 443).
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
that spectrum in future cases. According to one authority, it would be a mistake “to
conclude that the Pezim decision is capable of putting to rest the internal tensions
that have come to the surface in Canadian administrative law”,” in part because
the doctrine as enunciated contains too many pressure points for the re-
emergence of disputes that are papered over in the adoption of a consensus on
the framework of analysis of judicial review problems.
The most obvious of these pressure points are (1) what constitutes an issue that,
on a pragmatic and functional analysis, goes to the jurisdiction of a tribunal?
(2) what is meant by the degree of deference enunciated in the “patently unrea-
sonable” standard of review? and (3) which tribunals should be considered to
possess greater expertise than the courts on the issues being addressed in their
decisions and which do not?”
In fact, the unanimity in the Supreme Court’s theoretical approach articulated in
Pezim has indeed proved elusive in practice. Despite reiteration of the Pezimn
model, the Court’s next major judicial ruling, issued in January 1995,’ was marked
by the now-familiar division relating to the scope of the jurisdiction of a mature la-
bour relations tribunal protected by a broad privative clause, the standard of review
that should be applied to various aspects of its decision as well as to whether or not
those features were reasonable, patently unreasonable or incorrect.
C. The Supreme Court and Refugee Determination
In 1993, the Supreme Court of Canada decided an important “pure” refugee
determination case, that is, one involving the interpretation and application of the
statutory definition of Convention refugee per se.’ The unanimous decision in
Canada (A.G.) v. Ward’5 reversed a judicial review ruling of the Federal Court of
-” P.L. Bryden, “Developments in Administrative Law: The 1993-94 Term” (1995) 6 Supreme
Court L.R. (2d) I at 13 [hereinafter “Developments 1993-94”].
Ibid. at 14.
See Canadian Broadcasting Corp. v. Canada (LR.B.), [1995] 1 S.C.R. 157, 121 D.L.R. (4th)
385.
m For present purposes, “pure!’ refugee determination decisions are distinguished from those relat-
ing to refugee claimants but concerned more with “process”, such as Charter decisions in the Singh
case or, more recently, in Dehghani v. Canada (M.E.L), [1993] 1 S.C.R. 1053, 101 D.L.R. (4th) 654.
Leave to appeal has been denied in a number of other instances involving refugee determination per
se (see: Canada (M.E.L) v. Satiacun (1989), 99 N.R. 171 (F.C.A.), leave refused (21 December
1989), No. 21627 (S.C.C.); Canada (M.E.L) v. Villafranca (1992), 99 D.L.R. (4th) 334, 150 N.R. 232
(F.C.A.), leave refused (1993), 158 N.R. 400 (S.C.C.); Saini v. Canada (ME.L) (1993), 151 N.R. 239
(F.C.A.), leave refused (1993), 158 N.R. 300 (S.C.C.)). For a recent process case, see Nguyen v. Can-
ada (M.E.L), [1993] 1 EC. 696, 151 N.R. 69 (C.A.), leave refused (1993), 20 Imm. L.R. (2d) 245,
104 D.L.R. (4th) viii (S.C.C.).
5[1993] 2 S.C.R. 689, 103 D.L.R. (4th) 1 [hereinafter Ward cited to S.C.R.].
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
Appeal which had set aside an Immigration Appeal Board (“I.A.B.”) decision
granting the claimant, Ward, Convention refugee status.
Ward was clearly a case in which the I.A.B. decision might have been evaluated
against the C. U.PE. doctrine. The Court’s extensive reasons were, however – per-
haps understandably, given the precedential value of the decision –
concerned ex-
clusively with substantive interpretation of the Convention refugee definition. Al-
though the Court did observe that the Board’s approach had been correct in respect
of one aspect of its ruling, while erroneous in others, 7 the decision does not contain
a single item of C. U.PE. vocabulary relating to standard of review, expertise, rea-
sonableness or deference.”
The dearth of direct guidance in Ward in respect of review standards applicable
to refugee determination decisions seems to leave open the question of whether
general principles articulated by the Supreme Court in the post-C. U.P.E. line of ju-
dicial review cases are intended to guide Federal Court judges’ practice in that
area.” To what degree, if any, does the Court’s strong advocacy of deference and of
the high threshold patently unreasonable test apply to decisions concerned exclu-
sively with fundamental human rights questions such as those raised in Ward?
The Court remitted the matter to the C.R.D.D. for determination of an outstanding protection is-
sue. In October 1994, the C.R.D.D. panel reconsidering Ward’s claim, in light of the Court’s reasons,
denied him refugee status. As a fugitive from the Irish Republican Army, he had not demonstrated
clear and convincing evidence that the United Kingdom would be unable to protect him from his
former associates (see “Ex-terrorist Loses Bid to Stay in Canada” The [Montreal] Gazette (8 October
1994) Al 0). Ward’s application for leave to seek judicial review of that decision was subsequently
denied (see T. Falconer, “The Strange Saga of Paddy Ward” The [Toronto] Globe and Mail (26
August 1995) D2).
See Ward, supra note 55 at 723, 751,754.
s See Kiviatkowsky v. Canada (M.E.L), [1982] 2 S.C.R. 856, 142 D.L.R. (3d) 385, a semi-
procedural, semi-interpretive post-C.U.RE. decision, which involved a Federal Court of Appeal dis-
missal of an application for judicial review of an I.A.B. decision. In upholding the Federal Court of
Appeal’s ruling, the Court’s analysis focused, as it would in Ward, on substantive and interpretive is-
sues rather than on those of judicial review.
” Another recent process ruling touched upon the Federal Court’s supervisory role in immigration
matters. In Reza v. Canada, [1994] 2 S.C.R. 394, 116 D.L.R. (4th) 61, rev’g (1992), 98 D.L.R. (4th)
88 (Ont. C.A.) [hereinafter Reza cited to S.C.R.], the Court found that a provincial court’s refusal to
exercise concurrent jurisdiction to consider a refugee claimant’s Charter challenge was not review-
able on appeal where, as there, it had given sufficient weight to all relevant factors, notably the statu-
tory attribution of review jurisdiction to the Federal Court: “Parliament hats] created a comprehensive
scheme of review of immigration matters and the Federal Court [is] an effective and appropriate fo-
rum” (Reza, ibid. at 404-405).
Cheryl Mitchell has noted that, despite the Court’s acknowledgment of the divisional court
judge’s discretion, its view that he had exercised it correctly “essentially … amounts to a removal of
such discretion” for the future (C. Mitchell, “Update: Reza v. Canada” (1994) 4 N.J.C.L. 351 at 361).
Reza is significant to the degree that, for all practical purposes, it confirmed the exclusive nature of
the review jurisdiction of the Federal Court in immigration and refugee matters, including those rais-
ing constitutional issues.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
The Court’s own recent human rights jurisprudence may be relevant to this
question. As suggested in the above-noted excerpts from Pezim,w human rights tri-
bunals have been situated at the “correctness” end of the spectrum. The Court has
adopted a largely non-deferential stance in judicial and appellate review of human
rights decisions dealing with questions of law, in respect of which it considers that
human rights tribunals possess no particular expertise.’ Arguably, human rights and
refugee determination contexts are sufficiently similar to justify the application of a
common “correctness” standard, particularly given the Court’s endorsement, in
Ward, of human rights principles as appropriate guides in applying the Convention
refugee definition. ‘2 From this perspective, Ward might be interpreted as a case in
which the Court saw the I.A.B. as entitled to little or no deference in light of factors
such as the legal nature of the question at issue, the I.A.B.’s lack of particular ex-
pertise on that issue relative to that of the Federal Court’ and the apparently broad
grounds of review set out in the Federal Court Act.”
On the other hand, as Pezim also indicates, the principle of deference has been
repeatedly and explicitly extended in the Court’s caselaw to administrative agen-
cies’ findings of fact, whether in the human rights sphere or in general.’
In the
Court’s recent four-to-three decision in Chan, however, the majority ruling turned
,oSee quotes accompanying notes 48-49, above.
,Forjudicial review proceedings, see Mossop, supra note 23, in which a majority of the Court up-
held the Federal Court of Appeal’s ruling that the standard of review applicable to a human rights tri-
bunal in such matters was one of correctness, particularly in the absence of a privative clause. This
position was reaffirmed in University of British Cohnbia v. Berg, [1993] 2 S.C.R. 353, 102 D.L.R.
(4th) 665 [hereinafter Berg cited to S.C.R.], in which a majority again applied a correctness standard
with respect to the provincial Tribunal’s treatment of a general question of law.
For appellate proceedings, see Zurich, supra note 34, in which a majority of the Court declined
to defer to a tribunal’s findings of law in the presence of a broad right of appeal. Standard of review
and deference issues in light of tribunal expertise on questions of law prompted dissents in both Mos-
sop and Zurich. In Berg, L’Heureux-Dub6 J., who was in dissent in Mossop, adopted the majority’s
non-deferential posture.
62 See Ward, supra note 55 at 733-34. La Forest J. reiterated this approach in Chan v. Canada
(M.E.I.), [199513 S.C.R. 593 at 634-37, 128 D.L.R. (4th) 213 [hereinafter Chan cited to S.C.R.].
“Contrary to this position, it might be pointed out that the Court, in fact, rejected the Federal Court
of Appeal’s analysis of the persecution-protection nexus.
” Bryden seems to lean toward this view. The author comments that although the Court is not ex-
plicit that the correctness standard applies to refugee decisions, “the judgment makes no mention of
deference … and the legal issues raised by the case are addressed by the Court in the same way they
would be if they arose as ordinary appeals through the court system” (“Developments 1993-94″,
supra note 51 at 36-37).
6′ See e.g. Berg, supra note 61 at 370, in which a deferential standard was applied to a tribunal’s
findings of fact, despite the absence of a privative clause. See also: Mossop, supra note 23 at 577-78,
Lamer C.J., at 584, La Forest J., at 599, L’Heureux-Dub6 J. (dissenting); Zurich, supra note 34 at 338.
A notable exception to the general rule of deference to findings of fact may be found in Dickason,
supra note 34, in which the majority found (and the minority denied) that in light of the provincial
statute’s broad right of appeal on questions of fact, the Tribunal was not entitled to deference, while
the appellate court was entitled to review the evidence and make its own findings of fact. The hnini-
gration Act, of course, provides no such broad appeal rights.
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
on its own view of factual issues.” The case raises the question of whether the rul-
ing may be viewed as an attenuation of the long-standing rule of non-interference
in factual matters, in the context of review of refugee determination decisions.
In Chan, the majority upheld a Federal Court of Appeal majority ruling dis-
missing an appeal from the C.R.D.D.’s denial of refugee status. As in Ward, the
Court’s focus was on substantive issues. Standard of review considerations were
not addressed by the majority as grounds for upholding the Board’s decision, de-
spite Pezim’s extension of the deference principle to appellate review on the basis
of specialization of duties and despite restricted access to appeal under provisions
of the immigration Statute. Nor were standard of review grounds cited by the mi-
nority to justify interfering with the Board’s decision. What then, can be inferred
from the majority’s rare excursion into fact-finding?
Arguably, very little of broad significance can be inferred from the unusual
case. Chan presented the Court with a conflict between the decision under appeal
and the directly contradictory decision of the Federal Court of Appeal in Cheung v.
Canada (M.E.L),’ which had moreover been endorsed by a unanimous Court in
Ward. It is this context which seems to explain the majority’s readiness to enter into
detailed analysis of the merits of the appellant’s refugee claim in order to uphold a
deficient Board decision which had omitted making relevant findings of fact. The
majority’s approach appears, in short, to have been dictated by the circumstances of
the case and its desire to settle the matter expeditiously by distinguishing the appel-
lant’s case on narrow grounds, thereby avoiding the implications of conflict with
Cheung.” This close review may not be open to interpretation as indicative of a
general preparedness to undertake full appellate review of Board decisions in mat-
ters of fact.”
The minority would have remitted the matter to the C.R.D.D. because its mem-
bers “have the relevant experience and training” to make the omitted factual find-
ings, and “as triers of fact, it is incumbent on the Board” to make refugee determi-
nation decisions’ These terms suggest, in principle, a conventional approach to the
“See supra note 62.
‘7 [1993] 2 .C. 314, 102 D.L.R. (4th) 214 [hereinafter Cheung].
This view is supported by the majority’s conclusion:
In the absence of the appellant’s meeting the burden of establishing a proper fact foun-
dation … appellate courts are handicapped in attempting to determine legal issues not
grounded on the facts and should not attempt to do so. Therefore, the question of
whether Cheung should be followed in light of the decision of this Court in Ward
should await a case in which the necessary facts have been established in the refugee
determination hearing (Chan, supra note 62 at 673).
69 The Court’s deferential attitude in respect of tribunal findings of fact was, perhaps ironically in
light of the ambiguities of Chan, confirmed in a judgment issued concurrently on October 19, 1995
(see Large v. Stratford (City of), [1995] 3 S.C.R. 733 at 742-43, Sopinka J., at 754, L’Heureux-Dub6
J., 128 D.L.R. (4th) 193)).
7 Chan, supra note 62 at 618.
MCGILL LAW JOURNAL/REVUE DE DROITDE McGILL
[Vol. 41
relationship between appellate courts and first-level decision-makers. In light of the
context, it remains unclear whether, and to what degree, they and similar terms may
be interpreted as indicative of deference.7′
If Ward offers little direct guidance as to the standard of review applicable to
Board decisions on matters of law, the judicial review standard applicable to find-
ings of fact may be even more difficult to extrapolate from Chan. In any event, to
the extent that the distinction between questions of law and those of fact or those of
mixed law and fact72 remains important for purposes of judicial review, it is a far
from a negligible factor in the largely factual context of refugee determination,
where the subject matter of decisions is virtually always within the area of the tri-
bunal’s supposed expertise.
Under general principles set out in Pezim, legislated factors not directly ad-
dressed in Ward or Chan but relevant to situating the C.R.D.D. along the Pezim
spectrum would, in theory, entitle that agency to at least moderate deference. Such
factors include attribution of exclusive jurisdiction, partial insulation from review,
the abolition of a statutory right of appeal and the generalist nature of the reviewing
court. Nevertheless, as the ensuing discussion will argue, the weight to be attributed
to these items may be largely attenuated by other valid considerations.
II. The Practice: The Reasonable Refugee
The refugee determination context encompasses, of course, not only the inter-
ests of refugee claimants, but also the statutory framework governing the decision-
making process. That framework has come to be an increasingly dominant factor in
determining how the process – up to and including judicial review before the Fed-
eral Court –
affects the interests of those subject to it.
A. The Federal Court of Canada and Refugee Determination
From shortly after its creation in 1970 until 1993, the Federal Court of Appeal
has been the court of review in refugee-related matters, exercising various supervi-
sory functions under a number of statutes. From 1970 to 1978, it entertained ap-
peals, with leave, of I.A.B. decisions,” including, as of 1973, those relating to de-
71 In fact, La Forest J. appeared deliberate in declining to use terms commonly associated with def-
erence. The minority’s reluctant, but exhaustive, review of the factual and legal merits of the claim, in
response to the majority’s approach, was critical of the Board’s performance as trier of fact, suggest-
ing an absence of deference to it in practice (see Chan, ibid. at 619-20, 637, 647-48).
7 See Mossop, supra note 23 at 577, 599.
7′ See Immigration Appeal Board Act, R.S.C. 1970, c. 1-3, s. 23, as rep. and substituted by Federal
Court Act, supra note 15, Sch. II, Item 18.
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
portation orders against persons claiming to be refugees protected by the Conven-
tion.74
In 1978, the Immigration Act, 1976″ came into effect, introducing into domes-
tic Canadian law a definition of Convention refugee modelled on that of the 1951
Convention. At that time, Parliament also undertook to fulfil Canada’s international
obligations to refugees and to uphold its humanitarian tradition toward the dis-
placed and the persecuted, a policy objective which remains in effect.” Under the
Immigration Act’s cumbersome refugee determination process,” final determina-
tions of refugee status by the I.A.B!’ could not be appealed but were subject to ju-
dicial review as of right by the Federal Court of Appeal under section 28 of the
Federal Court Act.” This supervisory process remained in effect until 1989.” The
role of Trial Division judges was limited to review of immigration decisions of an
administrative rather than a quasi-judicial nature, reflecting the division of labour
dictated by the terms of sections 18 and 28 of the Federal Court Act.”
In January 1989, the reform package mounted to satisfy the 1985 Singh deci-
sion’s oral hearing requirement came into effect.” Bill C-55 created a new adminis-
” Immigration Appeal Board Act, ibid. at para. 11(1)(c), as rep. by S.C. 1973-74, c. 27, s. 5. This
marked the introduction of the Convention into Canadian law.
” S.C. 1976-77, c. 52 [hereinafter Immigration Act, 1976].
71 See ibid, at s. 3(g).
” Under the Immigration Act, 1976, a claim could only be made during an immigration inquiry (see
ibid. at s. 45). This involved an examination under oath, followed by a paper determination by the
Minister upon the advice of the Refugee Status Advisory Committee. Sections 70 and 71 provided
that applications for redetermination might be made to the I.A.B., which would hold a redetermina-
tion hearing if it considered, based on the file, that the claim would probably succeed.
” The hnmigration Act, 1976, repealed the Immigration Appeal Board Act (see ibid. at s. 128) but
maintained the agency itself within the new framework it established.
” This section, in effect from 1970 to 1992, provided that a judicial or quasi-judicial decision of a
federal board might be set aside on natural justice or jurisdictional grounds and for errors of law or for
errors of fact “made in a perverse or capricious manner or without regard to the material before [the
Board]” (Federal Court Act, supra note 15 at s. 28(1)).
” In March 1986, nearly a year after the Singh decision, Parliament amended the Immigration Act,
1976 to provide for an oral hearing by the I.A.B. on all redetermination applications by refugee
claimants and increased the number of Board members (see An Act to amend the Immigration Act,
1976, S.C. 1986, c. 13). On its own this modification was futile, because of the, by then, clear and
pressing need for overhaul of the entire system (see Part LV.B. 1, below).
” Section 18 referred to traditional prerogative remedies but was silent as to grounds on which they
might be granted (Federal Court Act, supra note 15 at s. 18).
2 See An Act to amend the Immigration Act, 1976 and to amend other Acts in consequence thereof,
S.C. 1988, c. 35. For purposes of clarity, it should be noted that in the 1985 consolidation of federal
statutes, which came into effect in December 1988, the Immigration Act, 1976, as amended by vari-
ous statutes since 1978, was consolidated at c. 1-2 as the Immigration Act. Thus, when S.C. 1988, c.
35 became R.S.C. 1985 (4th Supp.), c. 28, it was entitled An Act to amend the Immigration Act and to
amend other Acts in consequence thereof [hereinafter Bill C-55].
The number of refugee claimants awaiting processing was daunting. In 1980, Canada handled
1,600 inland claims to Convention refugee status; in 1985, the number had risen to 8,400, and the
MCGILL LAW JOURNAL/REvUE DE DROITDE MCGILL
[Vol. 41
trative agency, the I.R.B.,” to be composed of Governor in Council appointees
named to maximum five-year terms.”‘ Two-member panels of the C.R.D.D. of the
I.R.B. were given sole and exclusive jurisdiction over refugee status decisions.”‘ In
the event of a split decision, the claimant was entitled to the “benefit of the doubt”
and was recognized as a Convention refugee.” The Bill also created a two-person
screening or “credible basis” tribunal responsible for determining a refugee claim-
ant’s access and substantive entitlement to a full hearing before the C.R.D.D.” It did
not provide for an internal review of C.R.D.D. decisions. Under Bill C-55, the role
of the Federal Court of Appeal was varied once again; access to that court became
subject to a universal leave requirement” for appeals of C.R.D.D. decisions” and
for judicial review of credible basis decisions.” Leave applications were to be de-
cided without personal appearance;’ appeal from denial of leave was expressly
precluded. The grounds for appealing decisions of the C.R.D.D. duplicated judi-
cial review grounds set out at section 28 of the Federal Court Act.”
In February 1992 reforms to the Federal Court Act’ created a new, uniform
procedure for judicial review of all decisions of federal administrative tribunals,
over which the Federal Court, Trial Division, was vested with original jurisdiction.”
The Appeal Division retained judicial review authority over a prescribed list of
agencies, including the C.R.D.D.” The reforms. did not affect appellate jurisdiction
over that agency but did transfer judicial review jurisdiction over the significantly
fewer credible basis tribunal decisions to the Trial Division. The grounds of review
year following the Singh decision, that number had more than doubled (see Library of Parliament,
Canada’s Immigration Policy (Background Paper No. BP-190-E) by M. Young (Ottawa: Library of
Parliament, 1994) at App. 6). By the end of 1988, a backlog of 85,000 claims had accumulated (see
Office of the Auditor General, Report of the Auditor General to the House of Commons: Fiscal Year
Ended 31 March 1990 (Ottawa: Supply & Services Canada, 1990) at 347).
“See Immigration Act, supra note 5 at s. 57, as am. by Bill-C-55, ibid. at s. 18.
See Immigration Act, ibid. at s. 61, as am. by Bill C-55, ibid.
See Immigration Act, ibid. at s. 67, as am. by Bill C-55, ibid.
“See mmigration Act, ibid. at s. 69.1(10), as am. by Bill C-55, ibid. See also United Nations High
Commissioner for Refugees, Handbook on Procedures and Criteria for Determining Refugee Status
(Geneva: U.N.H.C.R., 1979) at paras. 203-204.
See Immigration Act, ibid. at s. 46.01, as am. by Bill C-55, ibid. at s. 14.
By virtue of section 82.1(2) of the Immigration Act, ibid., as am. by Bill C-55, ibid. at s. 19, only
visa officer rulings, subject as administrative decisions to the review jurisdiction of the Trial Division,
were exempted from the leave requirement.
See hmmigration Act, ibid. at s. 82.3, as am. by Bill C-55, ibid. at s. 19.
See Immnigration Act, ibid. at s. 82. 1 (1), as am. by Bill C-55, ibid.
“See Immigration Act, ibid. at s. 82.1(4), as am. by Bill C-55, ibid.
9′ See Immigration Act, ibid. at s. 82.2, as am. by Bill C-55, ibid.
” See Immigration Act, ibid. at s. 82.3, as am. by Bill C-55, ibid.
9′ Now R.S.C. 1985, c. F-7, as am. by An Act to amend the Federal Court Act, the Crown Liability
Act, the Supreme Court Act and other Acts in consequence thereof, S.C. 1990, c. 8 [hereinafter Fed-
eral Court Act 1992].
“See ibid. at ss. 18, 18.1.
“See ibid. at s. 28(1)(g).
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
were nominally expanded,97 and the grounds of appeal of C.R.D.D. decisions were
modified accordingly by a consequential amendment to the Immigration Act.”
By early 1992, cracks in Bill C-55’s refugee determination system were beyond
repair, and the backlog of refugee claimants awaiting determinations by the
C.R.D.D. and appellate review at the Federal Court of Appeal continued to grow.”
In February 1993, the second round of major legislative reforms in four years came
into force. Bill C-86 eliminated the credible basis panel and made the granting of
access to refugee determination an administrative decision within the exclusive ju-
risdiction of senior immigration officers.”‘ The practice of giving claimants the
The amended Act provided that relief was available if a tribunal:
(a) acted without jurisdiction, acted beyond its jurisdiction or refused to exercise its ju-
risdiction;
(b) failed to observe a principle of natural justice, procedural fairness or other proce-
dure that it was required by law to observe;
(c) erred in law in making a decision or an order, whether or not the error appears on
the face of the record;
(d) based its decision or order on an erroneous finding of fact that it made in a perverse
or capricious manner or without regard for the material before it;
(e) acted, or failed to act, by reason of fraud or perjured evidence; or
(f9 acted in any other way that was contrary to law (Federal Court Act 1992, ibid. at s.
18.1(4)).
“See hmnigration Act, supra note 5 at s. 82.3, as am. by S.C. 1990, c. 8, supra note 94 at s. 55.
I.R.B. figures show “intakes” of 20,267 refugee claimants in 1989, 36,200 in 1990, 30,550 in
1991 and 37,720 in 1992 (see Immigration and Refugee Board, “Statistics on Refugee Claims -Jan.
1, 1989-Mar. 31, 1994″ (Ottawa: I.R.B., 1994)). As a result of the Bill C-55 leave requirement, the
Court of Appeal issued 594 percent more judgments and orders without personal appearance in 1990
than it had in 1988 (see Federal Court of Canada Annual Report 1990 (Ottawa: FC.C., 1991) at 22).
From 1990 to 1992, the number of applications to the Court of Appeal for leave in immigration and
refugee matters increased from just over 2,000 to over 7,000, with refugee claimants filing the vast
majority of leave applications. In 1992, Court of Appeal judges disposed of over 8,000 leave applica-
tions (see Federal Court of Canada Report 1994 (Ottawa: F.C.C., 1995) at 10 [hereinafter EC.C. Re-
port 1994]). Total Court of Appeal proceedings, other than leave applications in immigration and
refugee matters, grew significantly from 1990 to 1992 owing to Bill C-55 appeals and judicial review,
with approximately 2,000 proceedings pending in 1990 and nearly 3,500 pending in 1992 (FC.C. Re-
port 1994, ibid at 9).
” See hnmigration Act, supra note 5 at s. 45, as am. by Bill C-86, supra note 9 at s. 35. Although
issues of access to the refugee determination process are beyond the scope of this paper, there is little
question that successive legislation has, in addition to gradually restricting access to appellate and ju-
dicial review, also progressively narrowed access to that process: in 1989, Bill C-55 and its compan-
ion statute, Bill C-84, known as the Deterrants and Detention Act, S.C. 1988, c. 36; in 1993, Bill C-
86; in 1995, An Act to amend the Immigration Act and the Citizenship Act and to make a consequen-
tial amendment to the Customs Act, S.C. 1995, c. 15 (or Bill C-44). The long-discussed “asylum-
sharing” agreement between Canada and the United States, when implemented, is expected to further
significantly reduce access to refugee adjudication in this country. Signing of the draft agreement, an-
ticipated for Spring 1996 (see House of Commons, Standing Committee on Citizenship, Minutes of
Proceedings and Evidence (19 March 1996) at 2:3), has been postponed pending the completion of
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
benefit of the doubt in split decisions was eliminated under certain circumstances.”‘
The leave requirement enacted by Bill C-55 was maintained. Bill C-86 also virtu-
ally nullified the role of the Federal Court of Appeal in refugee determination by
abolishing appeal of C.R.D.D. decisions and transferring judicial review authority
over all decisions taken under the Immigration Act, including those of the
C.R.D.D., to single judges of the Federal Court, Trial Division.'” One effect of the
move was to narrow the range of remedies available to refugee claimants.” Access
to appeal from a Trial Division judicial review decision was, in addition, subjected
to a requirement that the Trial Division judge certify, at the time of rendering judg-
ment, that the case involved a serious question of general importance.'” Under Bill
C-86 enabling provisions,” the transfer of nearly 2,000 pending cases from the Ap-
peal Division” meant that Trial Division judges began exercising their new judicial
review jurisdiction with a significant backlog.
In March 1995, then Minister of Citizenship and Immigration, Sergio Marchi,
announced cost-saving measures that will reduce refugee determination panels
from two members to one and cut the number of C.R.D.D. members.” Legislation
to this effect, expected in 1996, will mark the third major reform of the refugee de-
termination process in a seven-year period. An immediate result will be to abolish
altogether the legislated benefit of the doubt for refugee claimants.
B. Federal Court of Appeal: What’s a C.U.P.E.?
The Federal Court of Appeal’s fifteen-year history of reviewing refugee de-
termination decisions in judicial review and appellate proceedings roughly parallels
that of judicial review developments in the Supreme Court of Canada since
C.U.RE. One would be hard pressed, however, to find the latter’s influence re-
flected in the former’s refugee caselaw. Overall, Appeal Division refugee decisions
reveal frequent interventions in relation to errors of both law and fact, often on the
immigration-related debates in the United States Congress (see L. Sarick, “Refugee Pact Talks on
Hold” The [Toronto] Globe and Mail (16 April 1996) A6).
101 See Immigration Act, ibid. at s. 69.1(10.1), as am. by Bill C-86, ibid. at s. 60.
See Immigration Act, ibid. at s. 82.1, as am. by Bill C-86, ibid. at s. 73. Section 128 of Bill C-86
repealed paragraph 2 8(l)(g) of the Federal Court Act 1992.
‘”‘ Under the Federal Court Act, the Appeal Division sitting in appeal of a Board decision during
the Bill C-55 years might “give the decision that should have been given” and declare successful ap-
pellants to be Convention refugees (see: Federal Court Act, supra note 15; Federal Court Act 1992,
supra note 94 at s. 52). In fact, this authority was rarely exercised under grounds of appeal provided
by Bill C-55, and the claimant’s file was generally returned to the Board, often with specific direc-
tions (see e.g. Punniamnoorthy v. Canada (M.E.!.) (1994), 24 Imm. L.R. (2d) 1 (FC.A.)). Section
18.1(3) confers no authority for a Trial Division judge sitting in judicial review to give the decision
that should have been rendered (see Federal CourtAct 1992, ibid. at s. 18.1(3)).
” See Immigration Act, supra note 5 at s. 83, as am. by Bill C-86, supra note 9 at s. 73.
“‘4See Bill C-86, ibid. at s. 118.
” See EC.C. Report 1994, supra note 99 at 13.
,7See Citizenship and Immigration Canada, News Release 95/03, “Minister Marchi Announces
Single Member Refugee Panels” (2 March 1995) [hereinafter News Release 95/03].
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
basis of meticulous analysis of I.A.B. and C.R.D.D. records and without reference
to issues of deference, patent unreasonableness, tribunal expertise or to standards
per se under section 28 of the Federal Court Act.”‘ This interventionist practice has
the application of the
extended to all matters pertaining to refugee determination –
Convention refugee definition,” adverse credibility findings,”‘
the use of evi-
even though refugee determination decisions are overwhelmingly con-
dence'” –
cerned primarily with factual determinations and thereby, theoretically, most enti-
tled to the deferential posture advocated by the Supreme Court of Canada.
Several factors may be suggested for this apparent departure from judicial re-
view principles articulated by the Supreme Court. First, since the outset, the Federal
Court of Appeal has been responsible for establishing the legal tests governing
refugee determination and for supervising their implementation, a role that is ar-
guably inherently interventionist. Second, the Appeal Division may have consid-
ered that the I.A.B., whose primary jurisdiction related to immigration matters, but
which was also responsible for refugee determination from 1978 to 1989, did not fit
the bill of “expert” tribunal in that area. By the time the C.R.D.D. began operating
in January 1989, the practice of close supervision was long established. In addition,
” This is not to suggest references to tribunal expertise or patent unreasonableness are entirely ab-
sent from Court of Appeal caselaw (see e.g.: Gracielome v. Canada (ME..) (1989), 9 Imm. L.R. (2d)
237 (dissenting opinion); Aguebor v. Canada (M.E.I.) (1993), 160 N.R. 315 [hereinafter Aguebor];
Yuen v. Canada (M.E.!.) (7 July 1994), No. A-599-91).
0 See e.g.: Chowdluay v. Canada (A.G.) (12 May 1988), No. A-468-87; Lai v. Canada (ME.L)
(1989), 8 Imm. L.R. (2d) 245; Madelat v. Canada (M.E.!.) (28 January 1991), No. A-537-89; Kumar
v. Canada (M.E.!.) (12 February 1991), No. A-576-89; Butucariu v. Canada (M.E.I.) (5 February
1992), No. A-534-91; Veeravagu v. Canada (M.E.!.) (27 May 1992), No. A-630-89; Chichmanov v.
Canada (M.E.L) (3 September 1992), No. A-243-91; Diab v. Canada (M.E..) (24 August 1994), No.
A-688-91; Oyarzo v. Canada (M.E.!.), [1982] 2 P.C. 779; Rajudeen v. Canada (M.E.!.) (1984), 55
N.R. 129; Surujpal v. Canada (M.E.!.) (1985), 60 N.R. 73; Dhillon v. Canada (M.E.I.) (1990), 12
Imm. L.R. (2d) 118; Retnem v. Canada (M.E.I.) (1991), 13 Imm. L.R. (2d) 317; Wong v. Canada
(M.E..) (1992), 141 N.R. 236; Aluned v. Canada (ME.L) (1993), 156 N.R. 221; Cuadra v. Canada
(SoL Gen.) (1993), 157 N.R. 390.
“‘ See e.g.: Siddique v. Canada (ME.L) (23 November 1989), No. A-1137-88; Ye v. Canada
(M.E..) (1992), 17 Imm. L.R. (2d) 77; Vallejo v. Canada (M.E.I.) (26 March 1993), No. A-799-90;
Aden v. Canada (M.E.L) (28 April 1993), No. A-813-91; Fok v. Canada (M.E.L) (9 August 1993),
No. A-881-90; Chen v. Canada (M.E.I.) (4 October 1993), No. A-30-91; Maldonado v. Canada
(M.E.I.), [1980] 2 RC. 302, 31 N.R. 34; Owusu-Ansah v. Canada (M.E..) (1989), 8 Imm. L.R. (2d)
106, 98 N.R. 312; Frhnpong v. Canada (M.E.!.) (1989), 8 Imm. L.R. (2d) 183; Armson v. Canada
(M.E..) (1989), 9 Imm. L.R. (2d) 150; Hilo v. Canada (M.E.L) (1991), 15 Imm. L.R. (2d) 199, 130
N.R. 236; Giron v. Canada (ME.L) (1992), 143 N.R. 238.
“‘ See e.g.: Mensah v. Canada (ME..) (23 November 1989), No. A-I173-88; Djama v. Canada
(M.E..) (5 June 1992), No. A-738-90; Inzunza v. Canada (M.E.I.) (1979), 103 D.L.R. (3d) 105;
Galindo v. Canada (M.E.!.), [1981] 2 E.C. 781; Perez v. Canada (M.E..), [1981] 1 RC. 753; Irar-
razabal-Ohnedo v. Canada (ME.L), [1982] 1 E.C. 125; Ovakinoglu v. Canada (M.E.L) (1983), 52
N.R. 67; Chaudri v. Canada (M.E.I.) (1986), 69 N.R. 114; Tung v. Canada (M.E.L) (1991), 124 N.R.
388; Sathanandan v. Canada (M.E.!.) (1991), 15 Imm. L.R. (2d) 310; Padilla v. Canada (M.E.!.)
(1991), 13 Imm. L.R. (2d) 1, 160 N.R. 156; Okyere-Akosah v. Canada (M.E.I.) (1992), 157 N.R. 387;
Fajardo v. Canada (M.E.!.) (1993), 21 Imm. L.R. (2d) 113, 157 N.R. 392.
MCGILL LAw JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
the Court of Appeal’s role simultaneously became that of an appellate court enjoy-
ing a theoretically broader scope of intervention.”2 Third, early Court of Appeal
panels which set the tone and pattern for subsequent review of refugee determina-
tion decisions may simply have been influenced by the nature of the interests at
stake, in light of factors such as the perceived absence of I.A.B. expertise and, prior
to Singh, the real absence of procedural safeguards afforded refugee claimants. Fi-
nally, Court of Appeal judges may have viewed the patently unreasonable test –
a
product of jurisdictional analysis developed in the field of labour relations, which is
the example of expertise par excellence”‘ –
as less applicable in the context of
refugee determination, where jurisdiction, stricto sensu, has rarely been an issue.
Furthermore, although codification of judicial review grounds in the Federal Court
Act was not intended to supplant Common law grounds,”‘ Appeal Division judges
may have interpreted the grounds in section 28 relatively liberally in refugee law
cases.
Whatever the reasons for its activist approach when reviewing refugee determina-
tion decisions, it seems clear that the Federal Court of Appeal was routinely concerned
more with case-by-case scrutiny and evaluation of compliance with the legal tests it
had set than with application or articulation of general standards of review per se.
Closer analysis than is permitted here would be required to determine whether and to
what degree the Court of Appeal’s interventions in refugee cases also demonstrated
consistency of approach in that review role. It is generally acknowledged, however,
that during its tenure the Appeal Division did develop a relatively coherent set of prin-
ciples in a number of substantive areas of refugee law, and that the effect of Bill C-86’s
transfer of the supervisory role to Trial Division judges has been to silence a “major
judicial critic” of C.R.D.D. decision-making.”‘
..2 It is interesting to note, however, that it was in the 1989 Bell Canada decision, supra note 27,
that the Supreme Court of Canada extended the deference principle to appellate courts based on the
concept of “specialization of duties”.
“‘ See D.J. Mullan, “The Re-emergence of Jurisdictional Error” (1985) 14 Admin. L.R. 326 at 328
[hereinafter “Jurisdictional Error”].
“‘ See Law Reform Commission of Canada, The Federal Court Act: Administrative Law Jurisdic-
tion by D.J. Mullan (Ottawa: Department of Supply & Services, 1977) at 69 [hereinafter Adininistra-
tive Law].
” C. Rotenberg, “The Game of Jeopardy Justice: What is TWDNK and or TWDNC” (Paper pre-
sented to the Conference of Law Society of Upper Canada Department of Continuing Legal Educa-
tion on Understanding the New Immigration Act: How Bill C-86 Rewrites the Law, January 1993) at
F-i [unpublished].
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
C. Federal Court, Trial Division: Find the “Whopper””‘?
Judicial consistency has come into particularly sharp relief as a critical factor in
refugee determination in the post-Bill C-86 era. The pivotal review role previously
exercised by three-person panels subject to the levelling influence of each member
has devolved to single Trial Division judges. A survey of Trial Division refugee
decisions issued during the period from February 1993 through October 1995 indi-
cates a lack of uniform approach to judicial review among judges and an absence of
internal coherence in individual judges’ rulings. This absence of uniformity also
characterizes collegial and individual approaches to evidentiary and substantive is-
sues. An examination of examples drawn from the caselaw will serve to illustrate
the dimensions of the diversity.”‘
1.
The February 1993 – November 1994 Period
Over the course of this twenty-two month period, a strictly non-interventionist
approach is relatively rare among Trial Division judges. A consistent proponent is
Joyal J., whose rulings typically adopt language of deference to Tribunal expertise,
caution against substitution of judicial opinion- or microscopic analysis of Board
reasons and uphold the Tribunal’s right to be wrong.”‘ Although references to the
Supreme Court of Canada’s patently unreasonable test are infrequent and cursory,
Joyal J.’s method is explicitly premised on the view that the Supreme Court’s judi-
cial review guidelines impose “a primary duty of deference to any tribunal”.’
Court of Appeal refugee determination guidelines are apparently less authoritative:
[C]ertain doctrines propounded … by the Federal Court of Appeal … do provide
adequate safeguards to an unsuccessful refugee claimant who comes before
this court for review. … There has been … a plethora of Appeal decisions
quashing credibility panel or Refugee Board decisions.
“6 In De Calles v. Canada (M.E..) (1993), 19 Imm. L.R. (2d) 317, 67 F.T.R. 78 [cited to Imm.
L.R.], Muldoon J. expressed the view that where factual errors are alleged, an administrative decision
is not to be set aside “unless there is … a ‘whopper”‘ (ibid. at 317-18), and then proceeded to set aside
a Board decision on the basis of a perceived, arguably less than “whopping”, disregard for evidence.
“,7 This survey does not purport to contain an exhaustive review of the caselaw. The emphasis is on
the judicial review process rather than on substantive refugee issues per se.
“‘ See: Mahalingain v. Canada (Sol. Gen.) (2 November 1993), No. A-79-93, on appeal No. A-
649-93 [hereinafter Mahalingam] (internal flight alternative (“I.A.”)); Xu v. Canada (M.E.L) (21
December 1993), No. A-5-93 [hereinafter Xu] (persecution); Miranda v. Canada (M.E.I.) (1993), 63
F.T.R. 81 [hereinafter Miranda] (finding of fact); Owusu v. Canada (M.E.L) (1993), 64 F.T.R. 13
[hereinafter Oivusu] (credible basis); Estrella v. Canada (Sol. Gen.) (1993), 68 F.T.R. 65, aff’d
(1994), 176 N.R. 56 [hereinafter Estrella] (credible basis); Rajah v. Canada (ME.I.) (1993), 69 FT.R.
71 (I.F.A.).
“‘ Owusu, ibid. at 16. In both Owusu and Estrella, deference was shown to a credible-basis tribunal,
half-composed of an immigration adjudicator with greater experience in enforcement matters, on the
basis that it “must deal with hundreds of refugee claims and … could be said to have developed an ex-
perience or expertise in that arduous, onerous and difficult field” (Estrella, ibid. at 66).
MCGILL LAW JOURNAL/REVUE DE DROITDE McGILL
[Vol. 41
For purposes of judicial review, however, it is my view that a Refugee Board’s
decision must be interpreted as a whole. One might approach it with a pa-
thologist’s scalpel, subject it to a microscopic examination or perform a kind of
semantic autopsy on particular statements found in the decision. But mostly, in
my view, the decision must be analyzed in the context of the evidence itself. I
believe it is an effective way to decide if the conclusions reached were reason-
able or patently unreasonable.
The Trial Division of this court is just now entering into this field of judicial
review, and I believe that it will try to subscribe to and respect the guidelines
which have been expressed by the Court of Appeal. This is easily said but not
always easily applied when dealing with particular cases.”‘
Despite this repudiation of close scrutiny as a viable approach to judicial re-
view, some decisions suggest that microscopic analysis of Board reasons for judg-
ment may in fact be necessary to determine absence of patent unreasonableness.’ 2′
One would be hard pressed to differentiate this close-scrutiny method from review
for correctness or judicial agreement on the merits.
In addition, Joyal J. occasionally uses “palpable or evident error” language,
omitting any reference to patent unreasonableness or to Supreme Court of Canada
guidelines,'” or states no test at all in finding an absence of error of law.'” On at
least one occasion, Court of Appeal precedents are cited to support an uncharacter-
istic intervention “out of abundant caution”, where the basis for granting judicial
review appears indistinguishable from the bases for emphatic and principled non-
intervention in other cases.”4
By way of contrast, MacKay J. frequently adopts what amounts to a deferential
approach involving explicit reliance on the terms of paragraph 18.1 (4)(c) and/or (d)
of the Federal Court Act,’
rather than on Supreme Court of Canada authority.
,’ Miranda, supra note 118 at 81-82.
121 See Estrella, supra note 118 at 66.
” See: Siddiqui v. Canada (M.E.L), [1994] F.C.J. No. 108 (QL) (well-founded fear); Gomiez-Rejon
v. Canada (ME.IL) (1994), 93 ET.R. 96 [hereinafter Gomez-Rejon] (discrimination vs. persecution).
” See: Adan v. Canada (M.E.L) (6 January 1994), No. A-1246-92; Fuentes v. Canada (M.E.I.),
[1995] FC.J. No. 206 (QL) [hereinafter Fuentes] (change of circumstances (“change”)); Hersi v.
Canada (M.E.I.) (1993), 64 FT.R. 226 [hereinafter Herst] (civil-war context).
” In Abubaker v. Canada (Sol. Gen.) (1994), 70 FT.R. 74 [hereinafter Abubaker], a credibility de-
cision, Joyal J. noted that the Court of Appeal,
in a conscious effort to assure that the critical enquiry facing any refugee claimant be
conducted with an imperative emphasis on the rule of law, has repeatedly intervened
when legal fault was found and, without substituting its own judgment for that of the
Board, has simply instructed it to try again … (Abubaker, ibid. at 78).
‘ Supra note 94.
19961
M. HURLEY- REFUGEE DETERMINATION DECISIONS
These terms are, in turn, absent from Joyal J.’s vocabulary.’6 Under this approach,
MacKay J. scrutinizes Board decisions closely, concluding that intervention is not
warranted in the absence of errors of law, erroneous findings of fact made per-
versely or capriciously or unless, in weighing evidence, the Board’s “discretion can
be said to be unreasonably exercised”.’27 These reasons suggest a high threshold for
a finding of factual error, without, however, defining it other than by circular refer-
ence to the terms of the Act; some decisions make a point of noting the Board’s
“recognized” or “special” expertise.'”
Other decisions of MacKay J. upholding'” or setting aside’0 Board reasons refer
neither to the section 18.1 “test” nor to language of deference or expertise. Still
others confirm the impression of considerable fluctuation in language expressing
MacKay J.’s stringent review standard. In two decisions rendered the same week, in
which apparently significant Board lapses are acknowledged but found not to war-
rant granting judicial review, the operative standard in one case is that the Board’s
decision “cannot be said to be unreasonable on the evidence”,” and, in the other, a
combination of the notion of unreasonableness with the criteria of paragraph
18.1(4)(d).’ 2 A third decision issued during the same period suggests the Court’s
ability to intervene depends on whether the Board’s decision is “unreasonable … in
light of the evidence presented to it, or is otherwise contrary to law”.’ 3
Joyal and MacKay JJ.’s explicit adherence to what may broadly be described as
analytical frameworks for judicial review, however unevenly expressed or applied,
is exceptional. They are not, however, the only Trial Division judges to espouse a
high threshold. Richard J.’s rulings contain no explicit reference to either Supreme
Court of Canada guidelines or to the Federal Court Act but consistently refrain
from interfering with Board findings on the basis that intervention is not warranted
in the absence of patent unreasonableness,”‘ absence of unreasonableness ‘ or ab-
, See, for example, the following decisions by MacKay J.: Enuiet v. Canada (M.E.L), [1993] FC.J.
No. 855 (QL) [hereinafter Emnet]; Bhuiyan v. Canada (M.E.I.) (1993), 66 ET.R. 310 [hereinafter
Bhuiyan]; Tawfik v. Canada (M.E.L) (1993), 26 Imm. L.R. (2d) 148 [hereinafter Tawfik] (change).
“‘ Tawfik, ibid. at 152.
128 Ibid. See also Emnet, supra note 126.
’29 See e.g. Nkrunah v. Canada (M.E.L) (1993), 20 Imm. L.R. (2d) 246 [hereinafter Nkrumah]
(credibility, change).
,’ See e.g.: Dhaliwal v. Canada (M.E.L) (1993), 68 F.T.R. 129 [hereinafter Dhaliwal] (I.EA.);
Chowdhury v. Canada (SoL Gen.) (1994), 74 F.T.R. 115 [hereinafter Chowdhury] (change); Chen v.
Canada (M.E.I.) (1994), 76 FT.R. 235 (well-founded fear).
‘ , Fouchong v. Canada (Sec’y of State) (1994), 88 F.T.R. 37 at 38 (well-founded fear).
, See: Hazarat v. Canada (Sec’y of State) (1994), 88 ET.R. 128 (persecution, I.F.A.); MacKay J.’s
lengthy description of the applicable test in Tawfik, supra note 126 at 152.
. Danquah v. Canada (Sec’y of State), [1994] FC.J. No. 1704 (QL) (credibility).
‘3 SeeAguero v. Canada (M.C.L), [1994] RC.J. No. 1604 (QL) (well-founded fear, protection).
‘3’ See: Yanahida v. Canada (SoL Gen.), [1994] RC.J. No. 1527 (QL) [hereinafter Yanahida]
(discrimination vs. persecution, protection); Bukhari v. Canada (M.C.L) (1994), 87 F.T.R. 134
[hereinafter Bukhan] (prosecution vs. persecution, I.EA.).
MCGILL LAW JOURNAL/ REVUE DE DROITDE MCGILL
[V9ol. 41
sence of unreasonableness or other error of law.”‘
Similarly, although McKeown J.’s rulings do not refer to the Supreme Court of
Canada and seldom mention subsection 18.1(4) or notions of deference or tribunal
expertise, some explicitly endorse Joyal J’s approach to review” or decline to inter-
fere with Board findings that are not patently unreasonable,’3′ sometimes despite
admitted glitches in the decisions under review.’3′ These decisions may be con-
trasted with those which make no reference to the high threshold of review,’4′ which
appear to apply it less rigorously’4′ or which simply seem concerned with substitu-
tion of result in individual cases,’ 2 where findings of patent unreasonableness ap-
pear indistinguishable from disagreement on the merits in respect of conclusions
that have been theoretically open to the Board.”‘ In other decisions, McKeown J.’s
standard is expressed in broader, more generic terms, such as whether there is
“evidence to support” the Board decision, or whether it is “open to the Board” to
make a given finding, with or without mention of patent unreasonableness.'”
Leaving aside inferences arising from these cases, decisions of other Trial Di-
vision judges also suggest that omission of the term “patent unreasonableness” (or
of any reference to statutory review grounds in the Federal Court Act) cannot be
assumed to indicate that a strict standard has not been applied. 4 Similarly, mere
“‘ See Soma v. Canada (M.C.I), [1994] EC.J. No. 1696 (QL) [hereinafter Soma] (persecution,
protection).
’17 See e.g. Osoble v. Canada (M.E.L) (1993), 69 F.T.R. 184 (civil-war context).
,’ See: Lopez v. Canada (M.E..), [1994] EC.J. No. 753 (QL) (protection); Ahmed v. Canada (Sol.
Gen.), [1994] EC.J. No. 1270 (QL) (well-founded fear).
,’ See: Raheem v. Canada (ME.I.), [1993] EC.J. No. 626 (QL) [hereinafter Raheen] (I.F.A.);
Chaar v. Canada (ME..),. [1994] EC. No. 626 (QL) (well-founded fear); Murugesu v. Canada
(ME..), [1994] FC.J. No. 752 (QL) [hereinafter Murugesu] (I.F.A.); YusufV. Canada (Sec’y of State),
[1994] EC.J. No. 903 (prosecution vs. persecution) (QL); Soltan-Abadiramnhonnozi v. Canada (Sec’y
of State) (1993), 74 F.T.R. 51 (particular social group).
“41 See Fernando v. Canada (M.E.!.), [1993] F.C.J. No. 470 (QL) (L.EA.).
’41 See Ayimadu-Antwi v. Canada (M.E.L) (1993), 69 F.T.R. 157 (well-founded fear).
“‘ See: Adaros-Serrano v. Canada (M.E.!.) (1993), 66 FT.R. 300,22 Imm. L.R. (2d) 31 [hereinafter
Adaros-Serrano] (change); Dykon v. Canada (ME.L) (1994), 87 ET.R. 98, 25 Imm. L.R. (2d) 193
[hereinafter Dykon] (persecution).
“‘ See Arguello-Garcia v. Canada (M.E.!.) (1993), 70 FT.R. 1, 21 Imm. L.R. (2d) 285 [hereinafter
Arguello-Garcia) (change).
‘” Meedin v. Canada (Sec’y of State), [1993] RC.J. No. 1428 (QL) [hereinafter Meedin] (I.FA.);
Wong v. Canada (Sol. Gen.) (1993), 70 ET.R. 169 [hereinafter Wong] (well-founded fear).
“‘ Other rulings suggestive of a strict review standard and deferential approach but which make no
reference to patent unreasonableness include: Kumarasamy v. Canada (M.E..), [1994] FC.J. No. 671
(QL), Rothstein J. (I.F.A.); Siad v. Canada (M.E.!.) (1993), 64 ET.R. 271, 21 1mm. L.R. (2d) 6,
Rothstein J. [hereinafter Siad] (civil-war context); Oduro v. Canada (M.E.!.) (1993), 66 FT.R. 106,
Noel J. [hereinafter Oduro] (credibility); Antonio v. Canada (M.E.I.) (1994), 85 F.T.R. 241, Nadon J.
[hereinafter Antonio] (prosecution vs. persecution); Balasubramanian v. Canada (M.C.!.) (1994), 88
ET.R. 86, Muldoon J. [hereinafter Balasubramanian] (persecution).
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
reference to the term as the operative standard warranting the setting aside of Board
decisions need not denote a high review threshold.”‘
By way of further contrast, Reed J. proposes an apparently intermediate
threshold for review of decisions based on findings of fact.”‘ Under this approach,
paragraph 18.1(4)(d) of the Federal Court Act sets out “disjunctive conditions un-
der which a decision will be set aside, of which perversity is only one”.”‘ In this
light, the proper test “is to ask whether [findings of fact] are unreasonable on the
basis of all the evidence”.'” Nevertheless,
the court is unlikely to be quick to interfere with decisions regarding the exis-
tence or not of primary facts or with inferences drawn by the Board which are
particularly within its area of expertise and experience unless it is clear that
they are not supported by the evidence.’ 0
On the issue of deference to Board expertise, Reed J. also
takes an
“intermediate” line:
inhere has been a lot of criticism … but I think at the same time Boards do
deal with these kinds of cases on a regular basis. We have to give some weight
to that fact. … [The Board’s knowledge about documentary identification] is the
kind of experience (or expertise) to which I think a court would give some de-
gree of deference ……
Although Reed J. explicitly refers to this moderate standard in some subsequent
rulings,”‘ it is not immediately apparent from others in which no mention is made
of a “test” that it, in fact, represents the operative norm in all of her decisions.”‘
Furthermore, although Reed J.’s approach is occasionally cited by other Trial Divi-
sion judges as the appropriate review threshold,”‘ it generally appears that her “test”
“‘ See, for example, Cullen J.’s reasons in: Nina v. Canada (M.C.L) (1994), 87 ET.R. 169
(persecution); Li v. Canada (M.C.I.) (1994), 88 F.T.R. 46 (well-founded fear).
” See Singh v. Canada (M.E.L) (1993), 69 F.T.R. 142, Reed J. [hereinafter Singh and Narang
(Reed J.)] (credibility).
14 Ibid at 144.
41 Ibid. at 145-46.
‘ Ibid at 146.
“‘ Sambasivan v. Canada (Sec’y of State) (28 April 1994), No. IMM-3541-93 at 6 [hereinafter
Sanbasivan] (credibility).
,’ See e.g. Zhang v. Canada (M.E..), [1993] F.C.J. No. 1204 (QL) (causal link).
“. See e.g.: Osmani v. Canada (M.E.L), [1993] F.C.J. No. 1022 (QL) (change); Osei v. Canada
(M.E.L), [1993] FC.J. No. 1252 (QL) (credibility); Armagam v. Canada (M.E.L) (20 January 1994),
No. IMM-1406-93 [hereinafter Arumagan] (credibility); Appiah v. Canada (ME.L), [1994] FC.J.
No. 442 (QL) (change); Popov v. Canada (M.E.L) (1994), 75 F.T.R. 93, 24 Imm. L.R. (2d) 242
[hereinafter Popov] (well-founded fear).
” See rulings of Tremblay-Lamer J. in: Barabhuiyan v. Canada (M.E.L), [1993] EC.J. No. 1456
(QL), rev’d (11 September 1995), No. A-735-93 [hereinafter Barabhuiyan] (change, protection);
Ronero-Minero v. Canada (M.E.L) (1994), 75 F.T.R. 156 (change).
MCGILL LAw JOURNAL/REVUE DE DRO1TDE MCGILL
[Vol. 41
is not given broad endorsement or consideration, at least explicitly, in most of the
subsequent cases reviewed.”5
In a subsequent decision,’
‘ Reed J. describes the applicable test somewhat dif-
ferently, referring to the sufficiency of evidence as the appropriate standard:
I do not think there is serious doubt that a reviewing court will not set aside a
decision when the evidence supports the conclusion which the Board reached,
despite the fact that there are lacuna in the written text. If an absence of analy-
sis in a decision leads a judge to conclude that the Board did not consider a
crucial element of a legal test, or misunderstood or did not consider the evi-
dence, then, the decision will be set aside. … The administrative law test is
whether there is sufficient evidence to support the conclusion reached by the
Board.
5 7
The decision does not address what constitutes sufficient evidence or whether suf-
ficiency of evidence is to be equated with absence of unreasonableness.
Not surprisingly, the divergent approaches that Trial Division judges bring to
the task of reviewing refugee determination decisions may produce quirky results.
For example, in one of two Cullen J. decisions issued on the same day, Reed J.’s
“intermediate” approach is adopted to set aside a finding of fact where, paradoxi-
cally, no perverseness or capriciousness has been found.” In the second decision,
Cullen J. relies upon the stricter deferential approach advocated by Joyal and
MacKay JJ. to uphold an arguably questionable decision.”‘ In another instance,
Muldoon J. uses Joyal J.’s high standard to uphold a Board finding because the
Board has not gone “off the rails”, despite intellectual and emotional concern
caused by the sympathetic circumstances of an elderly, infirm Sri Lankan claim-
ant.” Muldoon J. then urges that the claimant be given humanitarian and compas-
sionate consideration on the basis of the same factors which arguably could have
justified setting aside the Board decision in the first place.
“‘ The Singh and Narang (Reed J.) ruling, supra note 147, was issued October 8, 1993. In Ajualip
v. Canada (Sec’y of State), [1994] F.C.J. No. 197 (QL) (T.D.) (credibility), for example, Rouleau J.
rules that he may not intervene unless the tribunal’s findings are “patently unreasonable and inconsis-
tent”. In Chowdhury, supra note 130, Mackay J. concludes that a tribunal finding was perverse “in
the sense that it was made without regard to the evidence” (ibid. at 119), suggesting a conjunctive in-
terpretation of paragraph 18.1(4)(d) criteria rather than the disjunctive interpretation advocated in
Singh and Narang (Reed J.). Although MacKay J. on occasion describes the applicable standard in
terms apparently reflective of Reed J.’s “intermediate” approach, the results suggest that, in practice,
for MacKay J. those terms reflect a high standard (see text accompanying notes 131-33, above).
“‘See Valencia v. Canada (M.E.L) (1994), 85 F.T.R. 218 (I.F.A.).
“s Ibid. at 225.
See Hristova v. Canada (M.E.L) (1994), 75 FT.R. 18, 23 Imm. L.R. (2d) 278 (well-founded
fear).
’59 See Hanna v. Canada (M.E.L) (1994), 74 ET.R. 202 (persecution).
‘6 See Naganathapillai v. Canada (ME.L) (1994), 24 Imm. L.R. (2d) 21 at 24 [hereinafter Nagana-
thapillai] (LEA.).
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
It should be re-emphasized that cases that articulate an analytical “approach” to
judicial review of refugee determination decisions are vastly outnumbered by Trial
Division decisions in which no such framework is discernible. Overall, factors in-
fluencing whether and when judicial intervention may be warranted appear to be
attributed differing significance in various circumstances, leading to a variety of
conclusions expressed in a number of different ways.
Judges differ widely, for example, in their readiness to intervene in the Board’s
treatment of evidence, generally, and of conflicting evidence” in particular. Appar-
ent disparities cut across the spectrum of substantive issues. Frequently, Trial Divi-
sion judges explicitly state a non-interference approach to the Board’s appreciation
of evidence.’
Interference with the Board’s role is, on the other hand, also com-
monplace.’ 2 The Board’s selective use,” or cursory or inadequate analysis,'” of
161 See: Fuentes, supra note 123 (change; Board competent to draw conclusions); Keerthaponrajah
v. Canada (M.E.L), [1993] EC.J. No. 627 (QL), Noel J. (L.FA.; “very questionable” conclusion open
to Board); Gabuzian v. Canada (ME.L), [1994] EC.J. No. 521 (QL), Nadon J. (well-founded fear;,
conclusion not unreasonable); Dolinski v. Canada (Sec’y of State) (31 March 1994), No. T-228-93,
McGillis J. (well-founded fear, findings reasonably open); Mortera v. Canada (M.E.I.) (1993), 71
F.T.R. 236, McKeown J., on appeal No. A-5-94 [hereinafter Mortera] (LEA.; findings open to the
Board); Megag v. Canada (M.E.L) (1993), 71 F.T.R. 64, Rothstein J., on appeal No. A-671-93
[hereinafter Megag] (I.F.A.; inappropriate to interfere with finding of fact); Badesha v. Canada (Sec’y
of State) (1994), 72 ET.R. 149, 23 Imm. L.R. (2d) 190, Wetston J., aff’d (1995), 180 N.R. 324
[hereinafter Badesha] (LEA.; Board considered circumstances); Yakinov v. Canada (Sec’y of State)
(1994), 76 F.T.R. 172, Rouleau J. (change; findings not unsubstantiated or frivolous); Kaler v. Can-
ada (M.E.L) (1994), 73 ET.R. 217, Cullen J. [hereinafter Kaler] (I.F.A.; evidence supports Board’s
finding).
162 See: Cortez v. Canada (M.E.L), [1993] FC.J. No. 882 (QL), Noel J. (well-founded fear;, unrea-
sonable conclusion); Soopramanien v. Canada (SoL Gen.) (5 October 1993), No. A-1572-92, Pinard
J. (LEA.; erroneous application of law); Choo v. Canada (Sec’y of State), [1994] FC.J. No. 303 (QL),
McGillis J. (discrimination vs. persecution; acted unreasonably); Carballo v. Canada (M.E.L), [1994]
F.C.J. No. 669 (QL), Cullen J. [hereinafter Carballo] (well-founded fear; error of law); Saez v. Can-
ada (M.E.L) (1993), 65 F.T.R. 317, 21 Imm. L.R. (2d) 15, Dub6 J. (LEA.; several glaring errors);
Aluned v. Canada (M.E.L) (1993), 68 F.T.R. 221, 22 Imm. L.R. (2d) 119, Nadon J. (change; findings
perverse); Cheng v. Canada (M.E.L) (1993), 70 F.T.R. 127, Denault J. (well-founded fear; ignoring
evidence); Mahmoud v. Canada (M.E.L) (1993), 69 F.T.R. 100, Nadon J. [hereinafter Mahmoud]
(change; error of law); Liyanagamage v. Canada (Sec’y of State) (1993), 71 ET.R. 67, McKeown J.,
rev’d upon cert. (1995), 176 N.R. 4 (sub nom Liyanagamage v. Canada (M.C..)) [hereinafter Liy-
anaganage cited to N.R.] (I.F.A.; evidence ignored); Kong v. Canada (M.E.L) (1994), 73 ET.R. 204,
23 Imm. L.R. (2d) 179, Reed J. (credibility; evidence misconstrued); Abdel-Khalik v. Canada (M.E.I.)
(1994), 73 ET.R. 211, 23 Imm. LR. (2d) 262, Reed J. (statelessness; not convinced Board understood
evidence); Moslim v. Canada (Sec’y of State) (1994), 75 F.T.R. 243, McGillis J. (law of general appli-
cation; patently unreasonable finding); Thannalingam v. Canada (M.E.I.) (1994), 76 F.T.R. 190,
Rouleau J. [hereinafter Tharmalingam] (LEA.; finding cannot stand); Jean v. Canada (M.E.L)
(1994), 82 F.T.R. 40, Noel J. (well-founded fear, no error per se identified); Ganeshalingam v. Can-
ada (M.C.L) (1994), 82 ET.R. 161, Rouleau J. (well-founded fear; satisfied that Board decision can-
not stand); Hotaki v. Canada (M.E.!.) (1994), 88 F.T.R. 43, Gibson J. [hereinafter Hotaki] (civil-war
context; error of law in failing to recognize, on evidence, that claimant targeted).
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
conflicting or inconclusive documentary evidence warrants intervention in some
instances but not in others.'” In some decisions, the absence of reference to cor-
roborative or other documentary evidence in Board reasons serves as a ground for
overturning the Board,'” while in others similar omissions do not produce similar
results.”7
” See: Dhaliwal, supra note 130 (LEA.; failure to apply test); Chowdhury, supra note 130
(change; serious error of law); Singh v. Canada (M.E..) (1993), 65 F.T.R. 110, Dub6J. (I.EA.; failure
to satisfy test); Hassanzadeh-Oskoi v. Canada (M.E.I.) (1993), 65 F.T.R. 113, Dub6 J. [hereinafter
Hassanzadeh-Oskoi] (change; decision cannot be allowed to stand); Kahlon v. Canada (Sol. Gen.)
(1993), 66 ET.R. 219,24 Imm. L.R. (2d) 219, Gibson J. (I.F.A.; failure to apply test); Rosales v. Can-
ada (M.E..) (1993), 72 ET.R. 1, 23 Imm. L.R. (2d) 100, Rothstein J. (well-founded fear, error of
law); Ventura v. Canada (M.C..) (1994), 86 FT.R. 9, Cullen I. (inability to protect, change; Board
erred by failing to consider relevant evidence).
‘” See: Elmanfoukh v. Canada (M.E..), [1993] FC.J. No. 1432 (QL), Gibson J. (change; error of
law); Boateng v. Canada (ME..) (1993), 64 FT.R. 197, Noel J. (change; Board erred); Benitez v.
Canada (Sol. Gen.) (1993), 66 F.T.R. 224, Gibson J. (change; misstatement of evidence, ignoring of
relevant considerations); Petrescu v. Canada (Sol. Gen.) (1993), 73 ET.R. 1, Tremblay-Lamer J.
(change; wrong test applied); Magana v. Canada (ME..) (1993), 70 F.T.R. 136, 22 Imm. L.R. (2d)
300, Rothstein J. (change; error of law); Mohamed v. Canada (M.E..) (1993), 74 F.T.R. 180, Denault
J. (change; findings failing to evaluate or not based on evidence); Costa v. Canada (ME..) (1994),
80 F.T.R. 12, Gibson J. (change; doubt as to whether totality of evidence taken into account); Risak v.
Canada (ME..) (1994), 86 ET.R. 67, 25 Imm. L.R. (2d) 267, Dub6 J. [hereinafter Risak] (well-
founded fear, Board erred in failing to analyze uncontradicted facts); Zetino v. Canada (M.C.!.)
(1994), 25 Imm. L.R. (2d) 300, Cullen J. (1.F.A.; not clear whether Board considered totality of evi-
dence; error of law).
“, See: Nkrumah, supra note 129 (change; decision not made without regard to the evidence);
Meedin, supra note 144 (1.EA.; finding open to the Board); Bonilla v. Canada (M.E.!.), [1993] F.C.J.
No. 625 (QL), Noel J. [hereinafter Bonilla] (change; conclusion open to the Board); Biko v. Canada
(Sec’y of State), [1994] EC.J. No. 1741 (QL), Gibson J. (well-founded fear; decision reasonably open
to Board); Boateng v. Canada (ME.L) (1993), 65 F.T.R. 81, McKeown J. [hereinafter Boateng]
(change; conclusions reasonably open to Board, no error of law); Hercules v. Canada (Sol. Gen.)
(1993), 67 ET.R. 131, Gibson J. [hereinafter Hercules] (change; conclusion reasonably open to
Board); Sekhon v. Canada (Sec’y of State) (1993), 73 F.T.R. 182, McKeown J., aff’d (1994), 176
N.R. 29 (I.EA.; finding reasonably open to Board); Bhinder v. Canada (M.EL) (1994), 76 ET.R. 74,
MacKay J. (I.EA.; inference open to Board, conclusion not requiring evidentiary support); Singh v.
Canada (ME..) (1994), 77 F.T.R. 234, Rothstein J. (I.F.A.; no conflicting evidence, Board not erring
in referring to evidence in summary fashion).
‘” See: Barabhuiyan, supra note 154 (change; conclusion not reasonable); Carballo, supra note
162 (well-founded fear, error of law); Haddoudi v. Canada (M.E.!.) (1 December 1993), No. 92-T-
1750, McGillis J. (well-founded fear; conclusion sufficiently unreasonable to warrant intervention);
Thind v. Canada (Sec’y of State), [1994] EC.J. No. 106 (QL), Rouleau J. (well-founded fear, error of
law); Garcia v. Canada (M.E.L) (11 March 1994), No. IMM-783-93, Nadon J. (change; error of law
and fact); loda v. Canada (M.E.L) (1993), 65 F.T.R. 166, 21 Imm. L.R. (2d) 294, Dub6 J. [hereinafter
loda] (discrimination vs. persecution; not satisfied Board applied proper principles to assessment of
evidence); Abubakar v. Canada (M.EL) (1993), 67 F.T.R. 313, Wetston J. (LEA.; test not met);
Mladenov v. Canada (M.EL) (1993), 74 F.T.R. 161, MacKay J. (discrimination vs. persecution; Court
uncertain as to whether evidence considered); Que v. Canada (M.E.L) (1994), 75 F.T.R. 154, Gibson
J. (well-founded fear; error of law); Raji v. Canada (M.E.L) (1994), 80 ET.R. 17, Wetston J. (change;
Board finding insufficient); Sivayoganathan v. Canada (M.C.L) (1994), 86 F.T.R. 152, Noel J.
[hereinafter Sivayoganathan] (credibility; not open to Board to disregard evidence without giving rea-
1996″I
M. HURLEY- REFUGEE DETERMINATION DECISIONS
Judicial evaluation of identical evidence also varies. In a number of I.F.A.
cases, the Board’s failure to adhere to, or to give sufficient weight to United Na-
tions High Commissioner for Refugees (“U.N.H.C.R.”) advisories, warrants judi-
cial intervention in light of a “reasonableness” test articulated by the Federal Court
of Appeal.'” Other judges decline to intervene in cases involving the very same
advisories, on the grounds that there is no requirement to follow them,’ 9 that they
represent non-exhaustive opinion evidence’ 0 or that they simply contain examples
of factors to be taken into account.”
A similar fluctuation of approaches and of readiness to intervene affects ap-
proaches to review of substantive issues. For instance, in cases involving persecu-
tion, Trial Division judges seem to apply significantly disparate thresholds, both
collectively and individually, in determining the sorts of factors and findings that
either do or do not justify the setting aside of negative Board decisions.'” Compa-
sons); Viebna v. Canada (M.C.I.) (1994), 87 F.T.R. 121, Rothstein J. (discrimination vs. persecution;
failure to make reference indicates evidence ignored); Velauthapillai v. Canada (SoL Gen.) (1994), 88
F.T.R. 315, Gibson J. (Board erred in law by concluding absence of well-founded fear without sup-
porting analysis of elements of claim); Campos v. Canada (M.C.I.) (1994), 89 F.T.R. 241, Muldoon J.
(change; several reviewable errors).
“67 See: Wong, supra note 144 (well-founded fear; evidence to support sufficient, open to the
Board); Ibrahim v. Canada (Sec’y of State), [1994] FC.J. No. 431 (QL) (well-founded fear; finding
reasonably open); Villacorta v. Canada (Sec’y of State) (1994), 77 F.T.R. 304, Mackay J. (change;
conclusion not perverse or capricious); Vafaei v. Canada (M.E.I.) (1994), 74 F.T.R. 60, Nadon J.
(change; conclusion neither perverse nor capricious).
‘” See: Thanbyayah v. Canada (M.E.!.), [1994] FC.J. No. 307 (QL), Gibson J. [hereinafter Than-
byayah] (failure to meet test); Sharbdeen v. Canada (M.E.!.) (1993), 66 F.T.R. 10, 22 Imm. L.R. (2d)
9, Dub6 J., aff’d upon cert. (1994), 23 Imm. L.R. (2d) 300, 167 N.R. 158 [hereinafter Sharbdeen] (the
finding cannot stand); Pathmakanthan v. Canada (M.E.I.) (1993), 71 F.T.R. 154, 23 Imm. L.R. (2d)
76, Denault J. (error); Kulanthavelu v. Canada (M.E.!.) (1993), 71 F.T.R. 129, Gibson J. [hereinafter
Kulanthavelu] (error of law).
“6 See Raheem, supra note 139 (result not patently unreasonable).
” See Eliyathamby v. Canada (M.E.L) (1994), 76 F.T.R. 156, Reed J. [hereinafter Eliyathamby
(test not improperly applied). In that case, Reed J. arrived at a result similar to that in Naganathapil-
lai, supra note 160.
..’ See Thanabalasingam v. Canada (Sec’y of State) (1994), 87 F.T.R. 188, Rothstein J.
… Contrast the following “severe” standard decisions in: Xu, supra note 118, Joyal J.; Mahalingam,
supra note 118, Joyal J.; Soina, supra note 136, Richard J.; Murugesu, supra note 139, McKeown J.;
Balasubramnanian, supra note 145, Muldoon J.; Brar v. Canada (M.E.L) (1993), 68 ET.R. 57,
Rouleau J.; Vlsagapernal v. Canada (M.C.I.) (1994), 87 F.T.R. 112, Mackay J. [hereinafter Visa-
gaperumal], with: Arguello-Garcia, supra note 143, McKeown J.; Adaros-Serrano, supra note 142,
McKeown J.; Dykon, supra note 142, McKeown J.; Bragagnini-Ore v. Canada (Sec’y of State),
[1994] EC.J. No. 143 (QL), Pinard J.; Castillo v. Canada (M.E.L.), [1994] FC.J. No. 1406 (QL),
Dubs J.; Shirwa v. Canada (M.E.I.), [1994] 2 F.C. 51, 71 ET.R. 136, Denault .; Xie v. Canada
(ME.L) (1994), 75 F.T.R. 125, Rothstein J.; He v. Canada (M.E.!.) (1994), 78 F.T.R. 313, 25 Imm.
L.R. (2d) 128, Simpson J. Contrast the high threshold decisions in: Manihani v. Canada (M.E..),
[1993] F.C.J. No. 878 (QL), Noel J.; Murugiah v. Canada (M.E.L) (1993), 63 FT.R. 230, on appeal
No. A-326-93, Noel J. [hereinafter Murugiah], with rulings in: Suthanthirarajah v. Canada (M.E.L)
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
rable disparity of approach is found in decisions concerning Board treatment of the
“cumulative acts” concept leading to findings of discrimination as distinct from
persecution.'” In cases with strikingly similar circumstances, Board decisions deny-
ing Convention refugee status to claimants from civil-war-tom countries elicit gen-
eral, but not uniform, reluctance to intervene.”‘
The voluminous credibility caselaw suggests that these Board findings, too, are
subjected to differing approaches. Trial Division judges show traditional judicial
reluctance to interfere with or to substitute their opinions for adverse credibility
findings by the “triers of fact,'”” on occasion despite explicit acknowledgment that
judicial discretion would have been exercised differently.’ On the other hand,
judges also interfere with credibility findings, despite the “rule” of reticence, with
remarkable frequency. In general terms, this group of decisions appears to subject
the Board’s entitlement to draw conclusions as to a claimant’s credibility to rela-
tively closer judicial scrutiny, indicating greater preparedness to intervene. As is the
case in decisions reviewed relating to evidentiary and substantive issues, the legal
bases justifying intervention'” or non-intervention ‘
in credibility cases are ex-
pressed in a variety of terms.
(1993), 20 Imm. L. R. (2d) 269, Noel J.; Castaneda v. Canada (M.E.L) (1993), 69 ET.R. 133, Noel J.;
Sinnathamby v. Canada (M.E.L.) (1993), 70 ET.R. 116, Noel J.
” Contrast decisions setting aside Board decisions in: Risak, supra note 164; loda, supra note 166;
Porto v. Canada (M.E.L), [1993] EC. No. 881 (QL), Noel J.; Freiberg v. Canada (Sec’y of State)
(1994), 78 FT.R. 283, Tremblay-Lamer l.; Khatib v. Canada (M.C.L) (1994), 83 ET.R. 310, McKe-
own J., on appeal No. A-592-94, with those in which Board decisions have been upheld: Gomez-
Rejon, supra note 122; Denes v. Canada (ME.I.), [1993] F.C.J. No. 1424 (QL), Dub6 J.; Larionov v.
Canada (ME.L), [1994] FC.J. No. 196 (QL), Denault J.; Bogdanov v. Canada (M.E.L), [1994] F.C.J.
No. 456 (QL), MacKay J.; Ponniah v. Canada (M.E.L) (1993), 68 ET.R. 149, Wetston J.
1
,”‘ Contrast non-interventionist approaches in: Her~si, supra note 123; Siad, supra note 145; Awale v.
Canada (M.E.L), [1994] FC.J. No. 679 (QL), Cullen J.; Abdulle v. Canada (M.E.L) (1993), 67 F.T.R.
229, Nadon J., with those in: Hotaki, supra note 162; Abdi v. Canada (M.E.) (1993), 68 F.T.R. 319,
Reed J.; Ahmed v. Canada (ME.L) (1994), 70 ET.R. 134, Noel J.; Abdi V. Canada (M.E.L) (1993), 70
ET.R. 209, Simpson J.
75 See: Ankrah v. Canada (ME.L), [1993] FC.J. No. 385 (QL), Noel J. (finding unassailable in ab-
sence of overriding error); Larue v. Canada (M.E.L), [1993] FC.L No. 484 (QL), Noel J. (evidence
allowing finding); Sun v. Canada (M.E.!), [1993] FC.J. No. 636 (QL), Noel J. [hereinafter Sun]
(finding unassailable in absence of perverseness); Bustillo v. Canada (M.E.!), [1993] FC.J. No. 1181
(QL), Denault J., on appeal No. A-647-93 (no interference as long as evidence to support); Wang v.
Canada (ME.L), [1993] FC.J. No. 1315 (QL), McKeown J. (conclusions well-founded); Muhzamned
v. Canada (M.E.L) (1993), 67 ET.R. 152, MacKay J. (findings not unreasonable); Asare v. Canada
(M.E.L) (1993), 72 F.T.R. 318, Gibson J. (credible basis tribunal; no substitution of decision based on
credibility).
171 See e.g. Oduro, supra note 145 (case giving Court “substantial difficulty” but no overriding er-
ror).
“n See: Sivayoganathan, supra note 166 (erroneous disregard of evidence leading to general finding
of lack of credibility); Pathmanathan v. Canada (ME.!), [1993] EC.J. No. 641 (QL), McKeown J
(evidence on which findings based unknown); Odunaike v. Canada (SoL Gen.), [1993] F.C.J. No.
1086 (QL), Dub6 J. (arbitrary conclusion not based on totality of evidence); Nizanul v. Canada
(M.E.L), [1993] EC.J. No. 1293 (QL), McGillis J. (failure to provide clear reasons reviewable error);
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
2.
The December 1994 – October 1995 Period
In the climate of diversity described, it is perhaps surprising that a November
30, 1994 Trial Division ruling apparently represents the first detailed survey of the
Supreme Court of Canada’s judicial review guidelines undertaken with a view to
determining their applicability to the C.R.D.D. under paragraphs 18.1(4)(c) and (d)
of the Federal Court Act. In Sivasamboo v. Canada (M.C.I.),” Richard J. infers
from those guidelines and, in particular, from criteria articulated in Pezim”‘ that
even though the terms of subsection 18.1(4) of [the Act] are quite broad in
scope, this Court should accord specialized tribunals “considerable” or
“significant” deference when they are acting squarely within their area of ex-
pertise.”‘
It is Richard J.’s conclusion that the C.R.D.D. is such a tribunal, based on the fol-
lowing “pragmatic and functional” analysis:
A review of the Immigration Act’s jurisdiction conferring provisions sug-
gests that despite the absence of a privative clause,
the effect of subsection 67(1) is similar in that the decisions of the Refugee
Division may be considered final and binding because of the exclusive grant of
Su v. Canada (M.E.I.), [1994] FC.J. No. 70 (QL), Reed J. (significant misunderstanding of evidence
laying foundation for negative finding); Uddin v. Canada (Sec’y of State), [1994] F.C.J. No. 140
(QL), Pinard J. (Board not on solid ground in finding inconsistency); Lozeva v. Canada (Sec’y of
State), [1994] FC.J. No. 215 (QL), McGillis J. (Board not entitled to make adverse credibility finding
without addressing cogent evidence; error of law); Ezi-Ashi v. Canada (Sec’y of State), [1994] RC.J.
No. 401 (QL), Wetston J. (inferences sufficiently perverse); Asiamah v. Canada (M.E.I.) (1993), 70
FT.R. 120, 23 Imm. L.R. (2d) 71, Rothstein J. (error in law’maldng findings unsupported by evi-
dence); Sadique v. Canada (M.E.L) (1993), 71 F.T.R. 37, Nadon J. (decision based on erroneous
finding of fact); Dumitru v. Canada (M.E.!.) (1994), 76 F.T.R. 116, Noel J. (conclusions based on
speculation); Callejas v. Canada (M.E.L) (1994), 73 F.T.R. 311, 23 Imm. L.R. (2d) 253, Gibson J. (in
absence of general finding against credibility; error of law).
“‘ See: Sambasivan, supra note 151 (conclusions reasonable); Arumagam, supra note 153 (errors
[1993] RC.J. No. 1037 (QL), Joyal J.
not substantial enough); Bah v. Canada (SoL Gem.),
(conclusion not spurious or without evidentiary base); Wang v. Canada (M.E.L), [1993] FC.J. No.
1180 (QL), Noel J., on appeal No. A-656-93 (finding not unreasonable); Campos v. Canada (M.E.I.),
[1993] FC.J. No. 1240 (QL), McGillis J. (Board entitled to weigh the evidence, conclusions not un-
reasonable); Dekereza v. Canada (M.E.L), [1994] FC.J. No. 62 (QL), Denault J. (within jurisdiction
of Board); Chaudhary v. Canada (M.E.L), [1994] FC.J. No. 497 (QL), Joyal J. (no interference ex-
cept on strong and obvious grounds); Atrned v. Canada (M.E.!.) (1993), 70 F.T.R. 124, McKeown J.
(open to the Board); Feng v. Canada (M.E.L) (1994), 77 FT.R. 231, Nadon J. (absence of error justi-
fying intervention).
‘” [1995] 1 F.C. 741, 87 F.T.R. 46, Richard J. (I.F.A.), appeal on cert. No. A-688-94, discontinued
January 19, 1995 [hereinafter Sivasanboo cited to F.C.].
“o See supra note 23.
ISI Sivasamboo, supra note 179 at 753.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
jurisdiction, the limited opportunities for judicial review, and the specialized
nature of the tribunal.”2
Both the Immigration Act and the Federal Court of Appeal have recog-
nized the Board’s specialized status. In addition to factors such as the complex
context of refugee determination and the enactment of specific provisions de-
signed to foster expertise,
[t]he Refugee Division has an important public interest mandate and a clear
policy development role with respect to the application of the Convention refu-
gee definition [insofar] as [the Immigration Act] authorizes the Chairperson …
to issue guidelines to assist the members in carrying out their duties under the
Act. … The importance of such a policy development role in determining the
extent of deference to be accorded to an administrative tribunal was recognized
in Pezim.”8
0 The Supreme Court of Canada’s determination that a correctness review
standard applies to human rights tribunal findings of law is distinguishable.
Refugee determination issues “are not broad questions involving general prin-
ciples of statutory interpretation and legal reasoning”, but rather, “the interpre-
tation of a statutory definition within a specific international law and regula-
tory framework”.’
0
In light of these factors, the C.R.D.D. is a specialized tribunal similar to
the B.C. Securities Commission considered in Pezim, and to other agencies
granted a high review threshold by the Supreme Court of Canada. It is thus
entitled to deference that applies ‘not only to the facts as found by the Refugee
Division, but also to the legal questions before it”.”‘
O
Since the principle of deference extends to appeal courts, and because the
standard of manifest or palpable error applies for purposes of appellate review
of findings of fact, it follows that,
with respect to the findings and conclusions of fact of a specialized tribunal, a
supervisory court will intervene only when it has been shown that there is a
manifest or palpable error, that is that the findings and conclusions of fact are
patently unreasonable.”‘
O
It can be inferred from assorted Federal Court rulings that “deference also
extends to the legal questions before the tribunal in the light of its role and ex-
pertise.'”97
“Ibid. at 755-56.
Ibid at 760.
“UIbid. at 761.
Ibid.
‘ Ibid. at 762.
197Ibid.
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
0 Hence, on the authority of recent rulings by the Court of Appeal and by
McKeown J., the standard of review to be applied to Board findings on ques-
tions of law and of fact is that of patent unreasonableness, a “very strict” test
under Supreme Court guidelines. ‘”
It is noteworthy that at no point of this lengthy and problematic judicial review
discussion”‘ does Richard J. attempt to rationalize his conclusions in light of the
Supreme Court’s Ward decision, which related specifically to the refugee determi-
nation context.”‘ Ward is simply not mentioned.
A review of Richard J.’s decisions since Sivasamboo shows that the majority do
not explicitly reflect his conclusions therein. Few make any reference to patent un-
reasonableness. ‘”‘ On occasion, an abbreviated non-refugee Appeal Division deci-
sion, observing that no practical difference exists between the standard of review
for alleged errors of fact set out in paragraph 18.1(4)(d) of the Federal Court Act
and the standard of patent unreasonableness,'” is cited by Richard J. in support of
his commitment to that standard.9 3
“‘The question certified by Richard J. for appeal purposes (see ibid at 765-66), apparently on his
own motion, as to whether the appropriate standard under paragraphs 18.1(4)(c) and (d) of the Fed-
eral Court Act is patent unreasonableness, has not gone to the Federal Court of Appeal. As indicated,
the claimants’ appeal was discontinued in January 1995 (see supra note 179).
‘” To identify just a few troublesome aspects of the analysis: Richard J.’s equation of the “manifest
error” standard used in appellate review for errors of fact to the patently unreasonable judicial review
standard for errors of fact and law may extrapolate somewhat broadly from the few rulings cited to
support that equation. Three of the five decisions cited with reference to the manifest error standard
(see Sivasanboo, ibid. at 762), for example, were concerned with credibility findings, not questions
of law. In addition, the three Appeal Division rulings cited (see ibid. at 764) as authority for Richard
J.’s conclusion with respect to patent unreasonableness as the appropriate standard are cursory, issued
by the same appellate panel over the same three-day period and cannot be said to represent a cross-
section of Appeal Division rulings. They are, arguably, not as categorical as Richard J. seems to sug-
gest. For the same reason, reliance on one McKeown J. ruling for authority as to the appropriate stan-
dard may not be fully warranted when, as described above, McKeown J.’s decisions are just as likely
to make no reference to patent unreasonableness. Overall, it is most remarkable that Richard J.’s
analysis simply fails to acknowledge the overwhelming majority of Court of Appeal and Trial Divi-
sion rulings that are silent as to standard and/or not deferential to the C.R.D.D.
“‘ See supra note 55.
“‘In Drougov v. Canada (M.C.L), [1995] FC.J. No. 709 (QL) (credibility, discrimination vs. perse-
cution), Richard J. finds that questions of credibility are entirely within the Board’s expertise, and no
interference is warranted unless findings are perverse, capricious or patently unreasonable. In both
Flores v. Canada (M.C..), [1995] FCJ. No. 710 (QL) and Gnanasundaram v. Canada (M.C.L),
[1995] RC.J. No. 707 (QL) (I.F.A.), Richard J.’s citing of Sivasamboo is accompanied by cursory and
virtually identical statements upholding Board decisions. By way of contrast, in Ratnam v. Canada
(M.C.I.), [1995] EC.J. No. 516 (QL), a Board decision based on a glaring error in its LA. finding is
set aside because the Board erred in its conclusion, not on the basis that it was patently unreasonable.
, 2 See Stelco Inc. v. Canada (Canadian International Trade Tribunal), [1995] FC.J. No. 832 (QL).
“‘ See: Farnaghi v. Canada (M.C.L), [1995] F.C.J. No. 987 (QL) (well-founded fear; factual find-
ings not meeting paragraph 18.1(4)(d) criteria and not patently unreasonable, no other reviewable er-
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
If explicit references to patent unreasonableness remain infrequent, their omis-
sion does not indicate a lowered review threshold. Richard J. continues to apply the
generally strict standard that was evident in his pre-Sivasamboo decisions.”‘ Ex-
ceptional departures from that standard, in apparently no more compelling circum-
stances than those subjected to the “very strict” test, seem difficult to explain other
than as substitution of opinion or disagreement on the merits.”‘
Contrasting results in Richard J.’s decisions support earlier impressions that the
presence or absence of explicit references to patent unreasonableness may not accu-
rately reflect the rigour of the standard applied. In one ruling,”‘ for example, a rare
conclusion of patent unreasonableness appears at odds with previous decisions in
which Richard J. finds an identical Board finding to be a determination entirely
within the Board’s expertise ‘” or otherwise not unreasonable.”‘ This conclusion, to-
gether with his further finding of a “number of [other] errors which warrant the in-
tervention of this Court”,'” also seems difficult to reconcile with an earlier ruling in
ror); Singh v. Canada (MC.L) (1995), 97 FTR. 139, Richard J. [hereinafter Singh and Narang
(Richard J.)] (persecution, LA., change; Board finding patently unreasonable; failure to consider is-
sue reviewable error).
“‘ See e.g.: Balendra v. Canada (M.C.1), [1995] FC.J. No. 191 (QL) (persecution, LEA.; finding
within Board’s expertise to make and supported by evidence); Chen v. Canada (M.C.L), [1995] FCJ.
No. 189 (QL), Richard J. [hereinafter Chen (Richard J.)] (well-founded fear; determination within
Board’s expertise, Board inferences not so unreasonable as to warrant intervention); Ivanov v. Canada
(M.C.L), [1995] FC.J. No. 184 (QL) [hereinafter Ivanov] (change, discrimination vs. persecution;
Board properly concluded no well-founded fear, within Board’s expertise to prefer documentary evi-
dence, conclusions reasonably open); Valdivia v. Canada (M.C.L.), [1995] EC.J. No. 730 (QL)
(protection, change; no reviewable error); Zanbrano v. Canada (M.C.I.), FC.J. No. 592 (QL)
[hereinafter Zanbrano] (well-founded fear, protection; not persuaded Board committed reviewable
error). Other post-Sivasanboo rulings by Richard J. upholding Board decisions without reference to
patent unreasonableness include: Poopalasingam v. Canada (M.C.!.), [1995] EC.J. No. 190 (QL)
(LA.); Hussain v. Canada (M.C.!.), [1995] F.C.J. No. 495 (QL) (change); Morales v. Canada
(M.C..), [1995] F.C.J. No. 510 (QL) (credibility); Oduro v. Canada (M.C..), [1995] FC.J. No. 515
(QL) (credibility); Falberg v. Canada (M.C.I.) (19 April 1995), No. IMM-328-94 (well-founded fear,
discrimination vs. persecution); Mohamud v. Canada (M.C.L), FC.J. No. 595 (QL) (change, I.F.A.);
Velickovic v. Canada (MC..) (1995), 96 FT.R. I (well-founded fear, prosecution vs. persecution).
Post-Stelco decisions by Richard J. not referring to patent unreasonableness include: Brown v. Can-
ada (M.C.L) (1995), 99 F.T.R. 271 [hereinafter Brown] (change); Paraneshvaran v. Canada (M.C.L),
[1995] FC.J. No. 996 (QL) (credibility); Botnarenko v. Canada (M.C.L.), [1995] FC.J. No. 1147 (QL)
(discrimination vs. persecution, protection); Oppong v. Canada (M.C.L), [1995] F.C.J. No, 1903 (QL)
(credibility); Petrisoaea v. Canada (M.C.L), [1995] EC.J. No. 1184 (QL) (well-founded fear, prose-
cution vs. persecution); Sakovich v. Canada (M.C.L), [1995] EC.J. No. 1406 (QL) (well-founded
fear).
‘” See Paul v. Canada (M.C.L.), [1995] EC.J. No. 187 (QL), Richard J. (well-founded fear; Board’s
conclusion made without regard to the material before it).
“‘ See Singh and Narang (Richard J.), supra note 193.
’97See Chen (Richard J.), supra note 194.
“‘See: Bukhari, supra note 135; Yanahida, supra note 135; Soma, supra note 136.
‘Singh and Narang (Richard J.), supra note 193 at 142.
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
arguably, equally compel-
grounds for review, Richard J. finds no reviewable error without any refer-
which, despite remarkably similar issues and similar –
ling –
ence to absence of patent unreasonableness.'”
A further survey of judicial review decisions issued in the months following
the Sivasamboo ruling indicates that, to date, Trial Division judges neither follow
nor refer to it.”‘ The ruling does, however, occasionally appear to feature implicitly
in some of Reed J.’s subsequent decisions. In one, the standard issue is addressed in
response to direct arguments that patent unreasonableness alone warrants quashing
a Board decision:
There is no doubt that deference should be paid to the decisions of specialized
tribunals, particularly when those tribunals deal with complex technical or so-
cial policy matters…. mhe closer the decision being made by a tribunal is to
the ordinary work of a court, the less deferential a reviewing court will likely
be.
If the use of a test of “patently unreasonable” means no more than that a re-
viewing court should be alert to the rules of deference… then, it goes without
saying that that verbal formulation is appropriate…. [I]f it signifies something
more, I have difficulty knowing what that is. A test of unreasonableness re-
quires … a more substantial type of error than mere incorrectness.
… The meaning of “patently” … is “plain” or “obvious”. I cannot envisage
overturning a decision of a tribunal as unreasonable where the unreasonable-
ness is hidden or where it is not plain or obvious. Also, I cannot envisage refus-
ing to overturn a decision because, while I might find it to be unreasonable, I
could not find it to be patently unreasonable…. I simply do not know what the
word “patently” adds to the applicable test.’
Reed J. further states that while this “verbal formulation” may differ from that of
some colleagues, “I am not persuaded that I differ in result.”‘
2 See Brown, supra note 194.
201 Muldoon J.’s ruling in Acosta v. Canada (M.C.L), [1995] EC.J. No. 1291 (QL) [hereinafter
Acosta] (well-founded fear), finding the Board’s selective use of documentary evidence “neither un-
reasonable or capricious or perverse”, is the only decision reviewed in which a Trial Division judge
other than Richard J. relies on Sivasamboo as authority for the patently unreasonable standard and a
posture of deference toward the C.R.D.D.
20 Efremov v. Canada (M.C.L) (1995), 90 ET.R. 259 at 260-61 [hereinafter Efremov] (discrimination
vs. persecution, credibility). It is worth noting that the explicit argument raised by counsel in this
case is not chronicled in Trial Division rulings prior to Sivasamboo and is rarely mentioned in subse-
quent cases.
203 Ibid at 261. Reed J.’s concluding reference to her 1993 Singh and Narang decision, supra note
147, described as containing a review of authorities “which had led me to conclude as I have”
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
A later rulinge expresses disagreement in principle –
albeit not explicitly –
with Richard J.’s view of the scope of deference owed the “expert” Board. Uphold-
ing a negative Board decision based on findings of implausibility, Reed J. adds
somewhat ambiguously:
I do not accord a high degree of deference to Board decisions. I do not consider
the Board to be an expert tribunal in the same way that a Securities Commis-
sion is expert in a technical area [citing Pezim]. In addition, the statutory pro-
visions of the Immigration Act indicates [sic] that privative-clause type defer-
ence is not intended. The Board does, however, have extensive expertise as a
result of dealing with a large number of cases-although sometimes this has a
deadening and fatiguing effect, rather than sharpening an expertise 2
Reed J.’s interventions are, however, exceptional. No direct impact of the unre-
solved theoretical debate over the appropriate scale of deference in the refugee de-
termination context is observable in other Trial Division decisions reviewed since
Sivasamboo. Richard J.’s conclusions in that ruling do not seem to be influencing
the various approaches that Trial Division judges continue to bring, collectively and
individually, to their review of C.R.D.D. rulings.
References to patent unreasonableness,” deference” or to Federal Court Act
provisions’
remain sporadic, including within the rulings of those Trial Division
judges observed to favour a stringent standard.” The cases do suggest increased
use by some Trial Division judges, at least some of the time, of indeterminate
“reversible” or “reviewable” error language to express the operative norm.”‘ In
(Efremov, ibid. at 261) suggests that the applicable standard remains, in her view, unchanged. Her
denial of certification of a question relating to the standard of review of Board decisions once again
delays direct confrontation of the issue by the Appeal Division.
20 See Connorv. Canada (M.C.I.) (1995), 95 F.T.R. 66, Reed J. (credibility, well-founded fear).
Ibid. at 68.
See e.g.: Zelentsov v. Canada (MC..), [1995] FC.J. No. 154 (QL), McGillis J. (protection); Tor-
res v. Canada (Sol. Gen.), [1995] FC.J. No. 628 (QL), Simpson J. [hereinafter Torres] (well-founded
fear, persecution); Maya-Gomez v. Canada (M.C..), [1995] FC.J. No. 861 (QL), McKeown J.
[hereinafter Maya-Gomez] (well-founded fear, change); Kroon v. Canada (ME..) (1995), 89 FT.R.
236, 28 Imm. L.R. (2d) 164, MacKay J. (well-founded fear, discrimination vs. persecution).
‘” See e.g.: Acosta, supra note 201; Hristov v. Canada (MC..), [1995] F.C.L No. 32 (QL), Cullen
J. [hereinafter Hristov] (credibility, well-founded fear); Vega-Solis v. Canada (M.C.I.) (1995), 93
FT.R. 211, Joyal J. (well-founded fear); Rokni v. Canada (M.C.L), [1995] EC.J. No. 182, Muldoon J.
(credibility).
‘” See e.g.: Okore v. Canada (M.C.L), [1995] FC.J. No. 428 (QL), MacKay J. [hereinafter Okore]
(well-founded fear, credibility); Canada (M.C.L) v. Lin, [1995] F.C.J. No. 1317 (QL), Gibson J.
(exclusion).
See Part I.C. 1, above.
2See e.g.: Zambrano, note 194; Dhillon v. Canada (M.C.L), [1995] EC.J. No. 198 (QL), Muldoon
J. [hereinafter Dhillon] (I.EA.); Sethi v. Canada (M.E.L), EC.J. No. 1828 (QL), Joyal J. (well-
founded fear); Singh v. Canada (M.C.L), [1995] EC.J. No. 183 (QL), Denault J. (well-founded fear);
Angeleveska v. Canada (M.C.L), [1995] EC.J. No. 591 (QL), Wetston J. (I.F.A.); Mohamad v. Can-
1996]
M. HURLEY- REFUGEEDETERMINATIONDECISIONS
broader terms, the impression persists that diversity of approach and of result sug-
gesting a variety of review thresholds continues to characterize Trial Division de-
cisions, whether in review of the Board’s treatment of evidence,”1 ‘ credibility”2 or of
substantive issues, generally.”3
ada (M.C.!.), [1995] EC.J. No. 657 (QL), Cullen J. (well-founded fear, LEA.); Sultana v. Canada
(Sec’y of State), [1995] FC.J. No. 663 (QL), Simpson J. [hereinafter Sultana] (credibility, well-
founded fear); Hossain v. Canada (M.C.!.) (1995), 99 F.T.R. 241, Nadon J. (credibility); Slavkovic v.
Canada (M.C.L), [1995] EC.J. No. 975 (QL), Heald D.J. (discrimination vs. persecution, protection);
Jiang v. Canada (M.E.!.), [1995] FC.J. No. 980 (QL), Gibson J. [hereinafter Jiang] (credibility); Ma-
noharan v. Canada (M.C.L), [1995] EC.J. No. 1111 (QL), McKeown J. [hereinafter Manoharan]
(well-founded fear, discrimination vs. persecution); Perera v. Canada (M.C.I.), [1995] EC.J. No.
1215 (QL), Jerome A.C.L (well-founded fear, credibility).
211 The voluminous caselaw under this general heading continues to show contrast in Trial Division
to evidentiary issues. Decisions taking a non-
judges’ general and individual approaches
interventionist approach include: Okore, supra note 208; Radosevic v. Canada (M.C..), [1995] RC.J.
No. 74 (QL), Rouleau J.; Oscal v. Canada (Sec’y of State), [1995] EC.J. No. 211 (QL), Gibson J.; De
Busto v. Canada (M.C.L), [1995] FC.J. No. 262 (QL), Rothstein J.; Kioreskou v. Canada (M.C..),
[1995] EC.!. No. 457 (QL), Muldoon J.; Thavamani v. Canada (M.C.L), [1995] EC.J. No. 518 (QL),
Wetston J.; Soma v. Canada (M.C.I.) (1995), 94 F.T.R. 203, Cullen J.; Liu v. Canada (M.C.L) (1995),
98 F.T.R. 88, Reed J.; Yaguna v. Canada (M.C.!.), [1995] EC.J. No. 813 (QL), Simpson J.;
Tchernilevski v. Canada (M.C.!.) (1995), 30 Imm. L.R. (2d) 67, Noel J. [hereinafter Tchernilevski];
Singh v. Canada (M.C.L), [1995] EC.J. No. 989 (QL), McKeown J. [hereinafter Singh (H.)]; Hamid
v. Canada (ME.L), [1995] FC.J. No. 1293 (QL), Nadon J. [hereinafter Hamid].
Decisions taking a more interventionist approach include: Bacanu v. Canada (M.E.!.), [1995]
F.C.J. No. 83 (QL), Jerome A.C.J. [hereinafter Bacanu]; Ivanouna v. Canada (M.C.L), [1995] FC.J.
No. 197 (QL), Reed J.; Oblitas v. Canada (M.C.L), [1995] FC.J. No. 280 (QL), Muldoon J.
[hereinafter Oblitas]; Kachine v. Canada (M.C.!.), [1995] FC.J. No. 214 (QL), Noel J.; Balasingam v.
Canada (Sec’y of State) (17 February 1995), No. IMM-2469-94, Rothstein J.; Jeyachandran v. Can-
ada (SoL Gen.), [1995] FC.J. No. 487 (QL), McKeown J. [hereinafter Jeyachandran]; Rivas v. Can-
ada (M.C.!.), [1995] F.C.J. No. 624 (QL), Wetston J.; Thavanayagapathy v. Canada (M.C.!.), [1995]
EC.J. No. 629 (QL), Simpson J. [hereinafter Thavanayagapathy]; Ameyaw v. Canada (M.C.L),
[1995] EC.J. No. 660 (QL), Cullen J.; Bougai v. Canada (M.C.!.) (1995), 98 ET.R. 152, Gibson J.
[hereinafter Bougai]; Contreras v. Canada (M.C.1.), [1995] EC.!. No. 798 (QL), MacKay J.; Zrig v.
Canada (M.C.L), [1995] ECJ. No. 1037 (QL), Pinard J.; Zidarevic v. Canada (M.C.L) (1995), 90
F.T.R. 205,27 Imm. L.R. (2d) 190, Dub6 J.
2′ Credibility decisions also continue to indicate marked variety in approach. Rulings taking a strict
non-interventionist approach apparently on the basis that credibility questions do not allow judicial
interference, such as Islam v. Canada (ME..), [1995] EC.J. No. 704 (QL), Tremblay-Lamer J., may
be contrasted with decisions undertaking close review of credibility findings (see, in particular: Sul-
tana, supra note 210; Ali v. Canada (ME.L), [1995] EC.J. No. 301 (QL), Gibson J.; Castro-Andino v.
Canada (M.C.L), [1995] F.CJ. No. 461 (QL), Pinard J.). Strict non-interventionism is also suggested
in, for example: Jiang, supra note 210; Amoakohene v. Canada (M.E.L), [1995] EC.J. No. 267 (QL),
Rothstein J.; Dhillon v. Canada (M.E.!.), [1995] EC.J. No. 390 (QL), Denault J.
Other rulings declining to interfere with credibility findings include: Zeng v. Canada (M.E.I.),
FC.J. No. 368 (QL), Cullen J.; Ponnampalam v. Canada (M.C.L), [1995] F.C.J. No. 511 (QL),
McKeown J.; Boateng v. Canada (M.C.L), [1995] EC.J. No. 517 (QL), Wetston J.; Singh v. Canada
(M.C.L), [1995] EC.J. No. 606 QL, Pinard J.; Akinremi v. Canada (M.C.!.), [1995] EC.J. No. 808
(QL), Simpson J.; Chen v. Canada (M.C..), [1995] FC.J. No. 1318 (QL), Gibson J.
Interference with credibility findings remains relatively common (see: Ganesalingam v. Canada
(M.C.L), [1995] EC.J. No. 26 (QL), Wetston J.; Navaramnam v. Canada (M.C.L), [1995] FC.J. No.
MCGILL LAw JOURNAL! REVUE DE DROITDE MCGILL
[Vol. 41
Finally, the subject of tests should be mentioned. Earlier research suggested that
“tests” articulated by the Appeal Division in relation to substantive issues might be
viewed by Trial Division judges as “specific standard[s] that … [preclude] the need
for recourse to generally applicable … principles” or for deference.’ This view is
tempered, however, by the further observation that the operative test, where one
exists, may not be referred to at all as a basis for evaluating Board decisions.”‘ It is
142 (QL), Rouleau J.; Opondo v. Canada (M.C..), [1995] F.C.!. No. 324 (QL), McKeown J.; Garcia-
Henao v. Canada (Sec’y of State) (1995), 93 FT.R. 197, Gibson J.; Tutu v. Canada (M.C.L.), [1995]
EC.J. No. 1035 (QL), Cullen J. [hereinafter Tutu]).
23 On the issue of well-founded fear and domestic violence, contrast approaches in rulings quashing
Board decisions in: Diluna v. Canada (M.E.L) (1995), 92 F.T.R. 67, 29 Imm. L.R. (2d) 156, Gibson
J.; Narvaez v. Canada (MC.L), [1995] 2 P.C. 55, 28 Imm. L.R. (2d) 170, McKeown J., on appeal No.
A-111-95 [hereinafter Narvaez], with: Manorath v. Canada (M.C.L.), [1995] F.C.J. No. 134 (QL),
Cullen J.; Sandy v. Canada (M.C.!.), [1995] FC.J. No. 1012 (QL), Reed J. [hereinafter Sandy]. Con-
trast Sandy with Reed J.’s decision of the same day in Williams v. Canada (Sec’y of State), [1995]
EC.J. 1025 (QL).
On the issue of discrimination vs. persecution, compare approaches in: Manoharan, supra note
210; Bougai, supra note 211; Kadenko v. Canada (Sol. Gen.), [1995] EC.J. No. 889 (QL), Tremblay-
Lamer J., on appeal No. A-388-95 [hereinafter Kadenko], with those in: Talman v. Canada (Sol.
Gen.) (1995), 93 P.T.R. 266, Joyal J.; Gidoiu v. Canada (Sec’y of State) [1995] F.C.J. No. 537 (QL),
Wetston J.; Victorov v. Canada (M.C..), [1995] EC.J. No. 900 (QL), Noel J.; Ramirez v. Canada (SoL
Gen.) (1994), 88 F.T.R. 208, Rouleau J.
On the issue of well-founded fear, contrast interventions in, for example: Howard-Dejo v. Canada
.C.J. No. 176 (QL), Noel J.; Cabello v. Canada (MC.L) (1995), 93 E.T.R. 156, Simp-
(M.C.L), [1995]
son J.; Annan v. Canada (MC..), [1995] 3 .C. 25, Dub6 J., with rulings declining to intervene in: Tor-
res, supra note 206; Tchernilevski, supra note 211; Marvin v. Canada (M.C.L), [1995] EC.J. No. 38
(QL), Joyal J. [hereinafter Marvin]; Cheng v. Canada (M.C.!.), [1995] EC.J. No. 95 (QL), Wetston J.
[hereinafter Cheng]; Toah v. Canada (Sec’y of State), [1995]
.C.J. No. 857 (QL), Gibson J.
On the issue of persecution, compare approaches of judges quashing Board rulings in, for ex-
ample: Vthani v. Canada (M.C.!.) (1995), 96 F.T.R. 313, McKeown J.; Rajathurai v. Canada
(M.C.L) (1995), 97 F.T.R. 134, Gibson J. [hereinafter Rajathurail, with those taken in: Singh v. Can-
ada (M.C.!.), [1995] EC.J. No. 232 (QL), Denault J.; Naguleswaran v. Canada (M.C.!.), [1995] F.C.J.
No. 587 (QL), Muldoon J.; Kandiah v. Canada (M.C.!.) (1994), 87
.T.R. 72, Cullen J. [hereinafter
Kandiah].
“‘ M.C. Hurley, Developments in Refugee Law before the Federal Court of Canada: Internal Flight
Alternative, Change of Circumstances (Directed Research paper, Graduate Studies in Law, University
of Ottawa, December 1993) at 41-42.
215 See, for example, LEA. decisions issued over the first period of reference in: Megag, supra note
161; Tharmalingam, supra note 162; Gill v. Canada (Sec’y of State), [1994] EC.J. No. 1458 (QL),
Rouleau J; Chaudhary v. Canada (ME.L) (1993), 71 E.T.R. 150, Joyal J. More recent decisions illus-
trate the same point. Although the Court of Appeal’s Aguebor “test”, supra note 108, is frequently
relied upon as authority for declining to intervene with adverse plausibility findings (see e.g.: Hristov,
supra note 207; Okore, supra note 208; Hamid, supra note 211; Sinha v. Canada (M.E.!.), [1995]
P.C.J. No. 422 (QL), Rouleau J.; Appau v. Canada (M.C.I.) (1995), 91 ET.R. 225, on appeal No. A-
178-95, Gibson J.) explicit reliance on it is by no means uniform in such cases. Furthermore, rulings
quashing credibility findings do not do so as a rule on the basis that intervention is justified despite
Aguebor, but simply do not refer to it (see generally, credibility decisions cited supra note 212). The
same observations appear applicable in respect of the “test” stated in Sagharichi v. Canada (ME.!L),
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
also true that references to appellate authority may not necessarily reflect the “test”
as originally stated.”‘ Both caveats suggest that a more focused evaluation would be
required before concluding on how such references, or their absence, affect the
standard of review applied under any substantive heading.
In addition, the Board’s application of substantive Court of Appeal tests which
require it to evaluate inherently relative factors like “reasonableness” when deter-
mining refugee status –
for instance, assessment of the objective foundation of a
claimant’s fear or of the availability of an I.EA. –
attracts a broad range of re-
sponses from Trial Division judges.”7 Furthermore, one of the effects of a deferen-
tial stance could be the sanctioning of potentially suspect Board interpretations of
higher authority, as in Ward,”‘ at a time when judicial analysis and interpretation of
the complex issues discussed in that decision and which are increasingly prominent
in refugee determination seem prescribed.
Finally, the theoretical levelling influence of the “test” is not necessarily avail-
able in respect of a number of substantive areas. At the time of the transfer of re-
view jurisdiction to the Trial Division, the Federal Court of Appeal had not formu-
lated “tests” per se for several legal issues”9 or had not yet fully developed princi-
ples relevant to their application.’
[1993] EC.J. No. 796 (QL) (C.A.), leave refused (1994), 170 N.R. 159 (S.C.C.), which is often cited
in decisions addressing Board findings of discrimination vs. persecution (see decisions cited supra
note 213).
216 See e.g.: Molina v. Canada (M.C.L.), [1995] F.C.J. No. 615 (QL), Nadon J.; Kazi v. Canada
(M.C.L), [1995] F.C.J. No. 605 (QL), Pinard J. [hereinafter Kazil; Ravasinghe v. Canada (M.E.L), (11
April 1995), No. IMM-1631-94, Nadon J. in which the Appeal Division’s ruling in Aguebor relating
to plausibility is cited as authority for non-interference in cases in which credibility does not appear to
be at issue.
” Contrast e.g.: Mortera, supra note 161, McKeown J.; Liyanaganage, supra note 162, McKeown
J., with: Badesha, supra note 161, Wetston J.; Nalliah v. Canada (M.E.I.), [1994] F.C.J. No. 403
(QL), Wetston J.; Kaler, supra note 161, Cullen J.; Rajaratnam v. Canada (M.E.I.), [1994] EC.J. No.
301 (QL), Cullen J.; Manoharan v. Canada (M.E.L) (1993), 70 F.T.R. 304, Rouleau J.; Thambi-
rathain v. Canada (M.E.I.), [1993] FC.J. No. 462 (QL), Rouleau J.; Sahi v. Canada (ME.L), [1993]
EC.J. No. 1398 (QL), Denault J.; Jbeili v. Canada (ME.L), (23 December 1993), IMM-917-93,
Denault J. Contrast decisions of Muldoon J. in Naganathapillai, supra note 160 and Reed J. in Eli-
yathamby, supra note 170, with: Thambyayah, supra note 168, Gibson J.; Kulanthavelu, supra note
168, Gibson J. Contrast interpretations of the Court of Appeal’s ruling in: Thirunavukkarasu v. Can-
ada (M.E.I.), [1994] 1 EC. 589; Singh and Narang (Richard J.), supra note 193; Jeyachandran, supra
note 211; Rajathurai, supra note 213, with those in Kandiah, supra note 213; Ravimohan v. Canda
(M.C.L.), [1995] EC.J. No. 1252 (QL), Muldoon J.
2′ Contrast approaches to Board interpretation of Ward in the matter of state protection in, for in-
stance: Marvin, supra note 213; Narvaez, supra note 213.
219 This seems to be the case, for example, in matters relating to application of the “exclusion”
clause of the Convention refugee definition, the prosecution vs. persecution analysis, the “availment”
of protection issue and family unity, to name just a few.
‘ “Change” decisions over the initial period following the transfer are characterized by disagree-
ment over the nature of the analysis required of the Board as well as the interpretation of factors set
out in earlier appellate decisions. Contrast: W1llalta v. Canada (Sol. Gen.) (1993), 68 F.T.R. 304, Reed
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
IH. What’s Wrong With This Picture?
If any generalization may be advanced based on the foregoing selected survey,
it is that no uniform standard has been applied consistently by Trial Division judges
when reviewing refugee determination decisions.’ No unifying features governing
the decision-making process appear discernible by judge, by issue, by the tradi-
tional classification of administrative error or by the terms invoked in reasons for
decision. The language is confusing, and the factors responsible for the shifts in ap-
proach among cases decided by the same judge are unknown.
This lack of uniformity both defies and demands further analysis. Why is a
single administrative decision-maker subjected to such differing degrees of scru-
tiny? Does an apparently lesser degree of scrutiny signify deference to the Board or
agreement with its findings? Conversely, does closer scrutiny signify mere disa-
greement on the merits? Which circumstances determine when the Board is entitled
to deference? What makes similar Board findings patently unreasonable in some
cases, not sufficiently unreasonable in others and reasonable in still others? Why is
it open to the Board to weigh and select evidence in some cases but not in others?
Why are some adverse credibility findings unassailable by their very nature while
others are subjected to close analysis? What are the identifying features of Board
errors that justify judicial intervention? What is reversible error? To date, searching
for analytically satisfactory answers has been a frustrating endeavour. Nevertheless,
likely factors contributing to the current unsettling state of refugee caselaw can be
found.
A. Causes
Related historical, theoretical and pragmatic considerations may explain both
Trial Division judges’ varied responses to supposedly “strict” judicial review stan-
J., on appeal No. A-677-93; Penate v. Canada (M.E.L), [1994] 2 F.C. 79,71 FT.R. 171, Reed J., with
Mahmoud, supra note 162. In Osei v. Canada (Sec’y of State), [1994] F.C.J. No. 1083 (QL), Reed J.
refused to sign an order, and in Manu v. Canada (M.E.L), [1994] FC.J. No. 1279 (QL), Nadon J. ad-
journed a hearing until the Appeal Division had provided further guidance as to the applicable legal
test in change cases. The “test”, subsequently clarified in Yusuf v. Canada (ME.L) (1995), 179 N.R.
11 (F.C.A.) [hereinafter Yusuj], is now frequently cited by some Trial Division judges as mandating
non-interference with Board findings of fact relating to change. However, contrast decisions taking
this approach in: Maya-Gomez, supra note 206; Singh (H.), supra note 211; Kazi, supra note 216;
Ivanov, supra note 194; Echeverria v. Canada (M.C.L), FC.J. No. 1214 (QL), McGillis J, with those
interpreting Yusuf more broadly when setting aside Board decisions in: Tutu, supra note 212; Jor-
danov v. Canada (M.C..), [1995] EC.J. No. 483 (QL), McKeown J. In a third group of change deci-
sions, Yusuf is not cited at all as authority for quashing Board decisions (see e.g.: Bacanu, supra note
211; Antonio v. Canada (M.C.L), [1995] F.C.J. No. 162 (QL), Noel J.).
” This does not suggest that Appeal Division decisions necessarily did, either. As suggested earlier,
however, the potential for inconsistency seems greater when decision-making is left to single deci-
sion-makers as opposed to three-member appellate panels.
19961
M. HURLEY- REFUGEEDETERMINATION DECISIONS
dards articulated by the Supreme Court of Canada and their generally fluctuating
approach in reviewing Board decisions. First, Trial Division judges have relatively
limited experience with judicial review supervision of quasi-judicial bodies, in
contrast to their much longer history as appellate judges under various statutes.22
Whether or not some judges have been relying on this appellate experience some of
the time, their exercise of the relatively new judicial review role, at least in refugee
cases, seems to vary significantly.’
Second, mixed messages emanating from the higher courts have probably also
contributed to the ambiguous caselaw. As previously discussed, a marked contrast
exists between the non-interventionist Common law judicial review norm advo-
cated by the Supreme Court and the interventionist practice of the Federal Court of
Appeal in the context of refugee determination. The fact that Trial Division judges
are now called upon to apply the same statutory grounds of review in the same
context, albeit in a different capacity, may well have contributed to confusion about
their role and about the operative standard that finds expression in the disposition
of individual cases.
A third factor has been the absence of unanimity at the Supreme Court level re-
garding the legitimate scope of judicial review and the operation of the deference
principle. This consideration, together with the lack of definition of the patently un-
reasonable test, may also have affected Trial Division judges’ willingness to en-
dorse a “very severe”, but essentially indeterminate standard that, moreover, the
Supreme Court has never applied in the refugee context.
Finally, judicial interference with substantive aspects of administrative deci-
sions is generally acknowledged to be a commonplace feature of judicial review, ir-
respective of context or nominally applicable standard:
[W]hile we have chosen that administrative tribunals will play a very important
role in modem society, we also hold the common law model of individual
consideration very dear, and that for the same reasons that we believe it would
be unacceptable not to have a right of appeal in, say, a criminal case, it is
equally repugnant to our notion of justice to preclude any sort of substantive
reconsideration where other individually significant legal issues are being de-
cided.24
22 See e.g.: Unemployment Insurance Act, R.S.C. 1985, c. U-i, under which they function as um-
pires; Income TaxAct, R.S.C. 1985 (5th Supp.), c. 1; PatentAct, R.S.C. 1985, c. R-4.
2 See A. Roman, “The Pendulum Swings Back” (1991) 48 Admin. L.R. 274 at 278-79, who ob-
serves that judges invest the term “judicial review” itself with different meanings, and that these un-
articulated differences can be crucial to their decision-maling processes.
‘ Holloway, supra note 42 at 328. See also: A. Hutchison, ‘The Rise and Ruse of Administrative
Law and Scholarship” (1985) 48 Mod. L. Rev. 293 at 320; J.M. Evans et aL, Administrative Law, 3d
ed. (Toronto: Emond Montgomery, 1989) at 381.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
It seems probable that, at least a good deal of the time, Trial Division judges have
not confined review of refugee determination rulings to decision-making processes
at the Board level but have been influenced by and responsive to merits-based, re-
sults-oriented arguments. This would account for apparently hybrid Trial Division
decisions in which various judicial review terms seem to be adapted to conclusions
concerning the merits of individual refugee claims.
Appeals to judicial sensibilities have long been recognized as significant to the
outcome of judicial review:
mhe courts intervene when counsel manage to push them across their
“threshold of shock.” … [EIxperienced counsel in judicial review matters don’t
start by trying to persuade the court as to the existence of an error of law on the
face of the record or of an error going to jurisdiction, but seek rather to call at-
tention to a wrong done their client. Thereafter, once they have the court “on
side,” the technical aspects of the law of judicial review are put forward as the
legally acceptable pegs on which a court may hang its visceral initial conclu-
sion as to the fairness of the matter.”‘
A significant feature of a threshold of shock, or merits-based, approach is that it in-
evitably depends on triggering a subjective judicial response to substantive cir-
cumstantial factors rather than on the objective judicial application of abstract legal
tests. Given the rules of judicial review, it is hardly surprising that where the merits
of a case are driven home or the threshold of shock is breached, error will be found
and the language fitted to an approximate legal standard allowed by the lexicon,
including patent unreasonableness. Where the appeal to judicial sensibilities fails,
the strict legal test, including absence of patent unreasonableness, will be called
upon to justify the opposite conclusion. Such a decision-making process seems to
have little to do with standards per se or with deference to administrative expertise.
It has been suggested, however, that “[t]o reject this realist dynamic outright might
amount to a reversion to … ‘bloodless abstractions of process’.”n
The fine line dividing process and substance may be particularly difficult to
maintain in the context of review of refugee determination decisions. Arguably, the
current system accentuates this difficulty. Trial Division judges are no doubt famil-
iar with the legislative evolution of the refugee determination process which has re-
sulted in their uniquely defined role within it. The enactment of Bills C-55 and
C-86 was dogged by controversy, much of which directly involved that role. The
controversy centred largely on process issues. It was argued that, in seeking to ex-
pedite the process, the reforms produced a system in which mistakes would inevitably
22 H.N. Janisch, “Towards a More General Theory of Judicial Review in Administrative Law”
(1989) 53 Sask. L. Rev. 327 at 331-32 [hereinafter “General Theory”]. See also H.N. Janisch, Book
Review of Administrative Law: Rethinking Judicial Control of Bureaucracy by C. Edley (1992) 37
McGill L.J. 339 at 343-44 [hereinafter “Book Review”].
“Book Review”, ibid at 344.
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
result from a failure to provide internal procedural safeguards.’ Refugee advocates
also raised concerns about altered rules of access to the courts ‘ to meet the dominant
concern for speedy disposition of large numbers of cases. 29 These features were said to
undermine and to be irreconcilable with Canada’s protection obligations under both the
Convention and domestic law.”
The legislated narrowing of options available to refugee claimants, combined
with the nature of the interests at stake in every refugee claim, have arguably pro-
duced the ideal setting for appeals to the judicial threshold of shock and judicial
interference with administrative decision-making.’ Trial Division judges know
they represent the denied claimant’s first, last and only kick at the can for relief.
One might expect this circumstance alone to evoke judicial reluctance to apply a
strict review standard or to defer to a Board which, notwithstanding Richard J.’s as-
sessment, differs significantly from those agencies toward which the Supreme
227 See: Minutes of Proceedings and Evidence of the House of Commons Legislative Committee on
Bill C-55 (Ottawa: Queen’s Printer, 1987) at 2:56-58 (J.B. Stem, Chair, former Refugee Status Advi-
sory Committee), 6:32 (P. Duquette), 6:134-36 (Amnesty International), 8:66-68 (Canadian Council
for Refugees), 8:178-79
(Union of Employment and Immigration Commission employees)
[hereinafter C-55 Proceedings]. See also: Canadian Bar Association (Ontario), Submission to House
of Comnons re: Bill C-55 Refigee Determination Process (August 1987) at 58-9 [hereinafter
“C.B.A.O. Submission”]; Minutes of Proceedings and Evidence of the House of Commons Legislative
Committee on Bill C-86 (Ottawa: Queen’s Printer, 1992) at 4:16, 4A:10-11 (Amnesty International),
4:70 (Inter-Church Committee for Refugees), 5A:19 (Canadian Council for Refugees), 5A:36-37
(Refugee Lawyers Association), 7:9-10 (U.N.H.C.R.), 8:46 (Table de concertation des organismes de
Montreal au service des rfugirs Inc.), 8:64-65 (Association qub~coise des avocats et avocates en
droit de l’immigration), 8:65 (P. Duquette), 8:75 (Refugee Project, Lawyers Committee for Human
Rights, New York), 10:10,17 (Canadian Jewish Congress), 10:93 (Quaker Committee for Refugees),
10A:32 (Jesuit Centre for Social Faith and Justice), 11:113, 11A:22-24 (J. Frecker, former Law Re-
form Commission of Canada (“L.R.C.C.”)), 12A:45-46 (V.I.G.I.L.) [hereinafter C-86 Proceedings].
2 See: C-55 Proceedings, ibid. at 3:5-6, 11, 30 (Canadian Bar Association (“C.B.A.”)), 6:9 (Rabbi
G. Plaut), 7:6, 11 (Professor W. Angus, Osgoode Hall Law School), 8:69 (Canadian Council for
Refugees); “C.B.A.O. Submission”, ibid at 60. See also C-86 Proceedings, ibiai at 4A:68 (Inter-
Church Committee for Refugees), 5:25 (Canadian Council of Churches), 5A:36-37 (Refugee Law-
yers Association), 7:65 (C.B.A., National Immigration Law Section), 7:117 (Professor J. Hathaway,
Osgoode Hall Law School), 8:64 (Association qu6brcoise des avocats et avocates en droit de
l’immigration), 11:112 (J. Frecker), 17A:114-15 (C.B.A., National Immigration Law Section).
See: C-55 Proceedings, ibid at 9:21ff (Minister G. Weiner); C-86 Proceedings, ibid. at 2:7,
2A:13 (Minister B. Valcourt), 14:13 (P. Harder, Assistant Deputy Minister, Immigration).
” See e.g. C-55 Proceedings, ibid at 4:6 (Inter-Church Committee for Refugees), 7:18 (Professor
J. Hathaway).
“‘ See e.g.: Oblitas, supra note 211; Nadelasingam v. Canada (M.C.L) (13 December 1994), No.
IMM-5711-93 [hereinafter Nadelasingam]. These were “borderline” cases in which Muldoon J. al-
lowed judicial review, in the first instance, because “to hold that the application fails could well put
[the claimants] on an aeroplane back to Peru” (Oblitas, ibid. at 6), and in the second, “for better or for
worse, in the complete knowledge that there is no appeal to this decision” (Nadelasingam, ibid. at 7).
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
Court of Canada has advocated deference.2 However, as has been noted, judges are
certain to vary considerably in their capacity for shock.”3
B. Cause for Concern
The availability of plausible theoretical explanations for variation in Trial Division
judges’ decision-making practices does little to alleviate a number of urgent practical
concerns and consequences for refugee claimants arising from that variation. The con-
cerns are both structural and normative.
1.
The Random Solution
Even under the most favourable circumstances, it is clear that reliance on judi-
cial impulse as a routine or intermittent basis for decision-making may produce
disorder and promote an absence of standards. As the confused caselaw described
above attests, to accept this acknowledged judicial tendency “to correct error and to re-
as a defacto norm of contempo-
spond to claims of unfairness in process and result’
rary judicial review would be particularly risky under the current refugee determination
system, with prior review of Board decisions absent and further review virtually
closed.
Judicial incoherence is not restricted to the refugee context. It has generally
been attributed to judges’ desire
to do the right thing, to shield citizens against perceived injustices, to vindicate
legal values. These tendencies are so strong that they lead judges to reach re-
sults by whatever means come to hand: … most importantly, selection of a re-
strained or interventionist approach towards the judges’ own role. What hap-
pens on the surface of the judgment is, in the end, determined not so much by
text-book maxims as the judges’ conviction that the interest of justice will or
will not be served by a particular result23
The refugee context, however, raises several causes for concern peculiar to it.
These relate to procedural aspects previously touched upon but which now require
closer attention. First, changes effected by Bills C-55 and C-86 have, as suggested,
placed refugee claimants at a distinct disadvantage relative to other applicants or
claimants subject to federal administrative procedures in other contexts, in terms of
22 See text accompanying notes 276-93, below.
See: “Jurisdictional Error”, supra note 113 at 334; P. Bryden, “Canadian Administrative Law in
Transition: 1963-88″ (1988) 23 U.B.C. L. Rev. 147 at 163 [hereinafter ‘Transition”].
2 D. Brown, “Privative Clauses” in N. Finkelstein & B. Rogers, eds., Recent Developments in
Administrative Law (Toronto: Carswell, 1987) 53 at 55.
25 H.W. Arthurs, “Protection against Judicial Review” (1983) 43 R. du B. 277 at 284 [hereinafter
“Protection”].
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
the hierarchy of statutory remedies available to them.” Not only have they not been
granted an appellate mechanism for substantive review at the administrative level,
but their only access to review by the courts has been subjected to an unprece-
dented leave requirement.”7
The leave requirement is a genuine refugee’s black hole. Leave applications are
decided, unless otherwise specified by the judge, in the absence of claimant and
counsel and, unless specifically requested, without the complete Board record.’
According to a study of the period prior to the transfer of review jurisdiction to the
Trial Division, single judges of the Federal Court of Appeal were deciding this
“make-or-break” access proceeding with statistically significant variations. These
results suggested to researchers that claimants did not have an equal chance of
convincing the judge that leave ought to be granted.” Although no equivalent study
has been conducted in respect of individual Trial Division judges’ disposition of
leave applications, there seems little reason to doubt that it would show similar di-
versity. Bill C-55 provided no criteria for the determination of leave applications,
and the “fairly arguable case” criterion that has been applied in theory ‘
is suffi-
ciently ambiguous to support different approaches in practice. Furthermore, in the
absence of an appeal from denial of leave or reasons for that denial “there are few
objective standards around which a caselaw could develop which could mitigate in
favour of a greater uniformity of the application of the law.”‘”
‘ Under other federal statutes, multi-tiered administrative mechanisms are not uncommon in addi-
tion to judicial review as of right or de novo appeals as of right to the Federal Court of Canada
(see e.g.: Canadian Human Rights Act, R.S.C. 1985, c. H-6; Unemployment Insurance Act,
supra note 222; Public Service Employment Act, R.S.C. 1985, c. P-32; Canada Labour Code,
R.S.C. 1985, c. L-2; Income Tax Act, supra note 222; Patent Act, supra note 222; Trade-Marks Act,
R.S.C. 1985, c. T-13).
” The leave requirement, while unusual, is not unique in federal law in respect of appellate review
of administrative decisions. It does represent a “first’ in relation to judicial review. Under section 84
of the pre-Bill C-55 Immigration Act, 1976, decisions of the second-level I.A.B. on appeal of first-
level administrative decisions regarding sponsorship or deportation had been subject to a leave re-
quirement governing access to further appeal to the Federal Court of Appeal. Bill C-55 did not affect
this process but added a leave requirement applicable in respect of appeals offirst-level C.R.D.D. de-
cisions, which also applies to judicial review of those first-level decisions under Bill C-86.
“a See Federal Court Rules, r. 10, 14. Refugee claimants face obstacles to gaining access to the
leave process itself, owing to growing restrictions in the availability of legal aid for that proceeding in
many provinces and the inability of impoverished claimants to retain counsel independently.
‘” See I. Greene & R Shaffer, “Leave to Appeal and Leave to Commence Judicial Review in Can-
ada’s Refugee-Determination System: Is the Process Fair?” (1992) 4 Int’l J. Refugee L. 70 at 82. The
authors’ study showed positive decisions on applications for leave to appeal C.R.D.D. rulings ranging
from three percent to 57 percent, with an acceptance rate on combined categories of leave applica-
tions varying from nine percent to 49 percent for an overall acceptance of 25 percent. This figure is
higher than that inferred from the available data for the 1993-1995 period.
2,” In Bains v. Canada (M.E.I.) (1990), 109 N.R. 239, 47 Admin. L.R. 317, the Court of Appeal
ruled the leave requirement did not violate section 7 of the Charter.
“‘ Greene & Shaffer, supra note 239 at 81-82. In Adjei v. Canada (M.E.I.) (1994), 74 F.T.R. 57,
Reed J., in dismissing a motion to reconsider a decision denying leave, acknowledged that the claim-
ant “had no way of knowing why leave had not been granted” (ibid. at 59) owing to the absence of
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
Federal Court annual reports do not show percentages of leave applications
granted at the Trial Division level. According to figures obtained from the Federal
Court Immigration Registry, 13,228 applications for leave were instituted by refu-
gee claimants from February 1, 1993 to October 31, 1995. The total number of ju-
dicial review proceedings commenced over the same period is 1,920. It cannot be
assumed that these figures overlap precisely to provide an exact percentage of
leaves granted overall. Leave applications filed one year may not be decided until
the following year; leave applications allowed may not be reviewed until the fol-
lowing year. Some judicial review proceedings tallied represent files in which leave
was granted under the pre-Bill C-86 regime. Nevertheless, assuming substantial
overlap, the figures indicate a low acceptance rate at the leave stage. A government
statistical summary covering the period up to November 1, 1993 suggested that ap-
proximately one leave application in five had been successful since 1989, with the
percentage down slightly in 1993.2 If this approximation holds to the present, only
twenty percent of refugee claimants denied at the Board level are granted access to
judicial review, as compared to one-hundred percent access for applicants in any
other context. Thus, refugee claimants to whom leave is granted may be said to
have already “survived” one lottery-like process. ” Because judicial review is not
finally determinative of refugee claims, some are subjected to the process more
than once. ”
The leave screening mechanism may be an unacknowledged factor favouring
the shock approach in the ensuing judicial review, influencing at least some Trial
Division judges to be susceptible to merits-based arguments at least some of the
reasons and accepted arguments that the Board’s credibility decision was substantially flawed. Never-
theless, in Reed J.’s view, the “fairly arguable test” requires the claimant to show “not only … that
there may be substantial errors in the decision under review but also that on the evidence as a whole
the applicants’ case is an arguable one” (ibid.). Because her decision denying leave “was not based on
a conclusion that the Board’s decision was valid but on the lack of merit to the claim on its face, the
procedure followed in issuing that decision becomes irrelevant” (ibid. at 60). Because reasons are
very seldom issued on leave applications, the result of this approach is that, in addition to leaving the
claimant in the dark, the Board is left uninstructed and unimpeded with respect to significant flaws in
its procedure and decision-making.
242 See Citizenship and Immigration Canada, Refugee Determination and Resettlement (Ottawa:
C.I.C., 1994). From inquiries to Citizenship and Immigration Canada, it would appear that no more
recent version of this document has been prepared.
243 In 1992, the former L.R.C.C. suggested it was unlikely that inconsistency in Board decision-
making could be remedied by Federal Court judges because “the basis on which questions of interpre-
tation reach the Federal Court is itself unpredictable” (L.R.C.C., The Determination of Refugee Status
in Canada: A Review of the Procedure, Draft Final Report (Ottawa: L.R.C.C., 1992) at 82 [hereinafter
Determination]). An example of the unpredictability of the leave process is found in Shau v. Canada
(M.C.I.) (1996) 97 F.T.R. 313, Nadon J. In that case, an application for leave was denied by Joyal J.
without reasons in January 1994 and the applicant’s motion to reopen the Board hearing was also de-
nied on the basis of Joyal J.’s ruling. In January 1995, MacGuigan J. then granted the applicant leave
to challenge the Board’s denial on grounds similar to or the same as those rejected by Joyal J. Nadon
J. applied the doctrine of issue estoppel.
2″ See infra note 356.
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
time. This, however, offers little assurance that the genuine refugee can weather a
second process characterized by similar unpredictability.
A second significant procedural element involves the certification process in-
troduced by Bill C-86, under which rulings rendered on immigration and refugee
judicial review applications filed after February 1, 1993 became the only final Trial
Division decisions not appealable as of right.2′ The certification requirement ap-
pears to be both unprecedented in Canadian law and more draconian in effect than
rare analogous measures adopted in other Common law jurisdictions. ”
According to government spokespersons, the changes wrought by Bill C-86
were essentially plumbing;”7 the certification process would not be left to Trial Di-
vision judges’ discretion without any guidelines.”‘ In practice, by its very nature the
process has been, as intended, entirely restrictive. ‘ The limited nature of “certified”
appeals was confirmed in an Appeal Division ruling, noting that the effect of certi-
fication in a pre-Bill C-86 file –
in respect of which an appeal remained available
as of right – was that “the Appellant’s rights of appeal have been unnecessarily re-
stricted, in the sense that his ground of appeal is limited by the question stated.”‘”
Furthermore, the notion that once a question was certified in a post-Bill C-86 file,
the Court of Appeal might, within its jurisdiction, consider all aspects of the ap-
peal” has also been put to rest. The Appeal Division has ruled that not only must a
certified question “be one which, in the opinion of the motions judge, transcends
the interests of the immediate parties to the litigation and contemplates issues of
2.See: text accompanying note 104; Federal CourtAct 1992, supra note 94 at s. 27.
” In hearings before the House of Commons Legislative Committee on Bill C-86, the process was
defended by pointing to a similar provision in the British Criminal Appeal Act 1968 (U.K.), 1968, c.
19, s. 33(2), under which no appeal lies to the House of Lords unless the Court of Appeal has certified
that a point of law of general importance is involved (see C-86 Proceedings, supra note 227 at 14:64
(E. Bowie, Assistant Deputy Minister, Litigation Section, Department of Justice). The British model
differs significantly from the Bill C-86 process, in that it occurs at a “higher rung” in the judicial hier-
archy, taling place only after an appellate court has already rendered a decision. The same distinction
applies to the process referred to in Huynh, supra note 8, in which the Constitution of the State of
Florida subjects appeals from district courts of appeal to the State Supreme Court to certification by
the former (see Fla. Const. art. V, s. 3(b)(4)).
2″, See C-86 Proceedings, ibitl at 2:28 (Minister B. Valcourt).
24 Ibid. at 14:59-62 (E. Bowie).
249 In other words, its drafters intended it to be used only if a matter “really raised a question of law,”
or if this was a case “to which the time of three appellate court judges should be devoted” (ibid. at
14:62 (E. Bowie)).
‘ Villalta v. Canada (SoL Gen.) (3 February 1994), No. A-677-93 (interlocutory ruling). Inadver-
tent application of the certification process occurred because in the immediate post-Bill C-86 period,
Trial Division judges were dealing with three different categories of files, not all of which were sub-
ject to certification.
.. This had been the view expressed by Rothstein J. in Ramoutar v. Canada (M.E.I.), [1993] 3 FC.
370 at 379-80,65 F.T.R. 32, on appeal No. A-403-93.
MCGILL LAW JOURNAL/REVUE DE DROITDE McGILL
[Vol. 41
broad significance or general application”, but “the jurisdiction of the Court where
a question is certified … is limited to an appeal on the question certified.”2 ‘
A subsequent Trial Division decision that found the certification process in com-
pliance with the Charter’s fundamental justice norm was blunt about the implications
of the limitation, stating that certification is not dependent on the facts of the case and
is divorced from its disposition. ‘3 For the genuine refugee, whose claim may hinge
largely on factual issues specific to his or her situation, divorcing the facts from the
process aggravates the problem of reduced access to appeal. Given the numbers of
judicial review proceedings involved, the low numbers of questions certified appear
to support this conclusion. ‘
The low number of questions certified and a review of decisions in which cer-
tification has been requested also attest to the intrinsically discretionary nature of
the process.’ This is reflected in reasons given for granting or denying certification in
cases raising various issues of equal significance’
as well as in contrasting views
among Trial Division judges as to what constitutes a “serious question of general im-
p Liyanagamage, supra note 162 at 5. D6caiy J. added that the Court should not “be unduly re-
strictive in interpreting the scope of [that] question” (ibid.).
..3 See Huynh, supra note 8 at 646-47, 651.
‘ According to information obtained from the Federal Court, from February 1 to December 31
1993, ten questions were certified in post-Bill C-86 files, of which two duplicated previously certified
questions. In 1994, 17 appeals were filed in post-Bill C-86 files. In 1995, 44 appeals had been lodged
in the Appeal Division to October 31, 1995 (telephone interview with Federal Court, Judicial Infor-
mation Services (November 1995)).
“- In llanko v. Canada (Sol. Gen.) (1995), 93 F.T.R. 284, 27 Imm. L.R. (2d) 106 [hereinafter
Mlanko], Simpson J. explicitly affirms the discretionary nature of certification.
.. For cases in which certification was granted, see e.g.: Sharbdeen, supra note 168; Murugiah, su-
pra note 172; Narvaez, supra note 213; Kadenko, supra note 213; Seifelmlioukova v. Canada (M.E.I.)
(1994), 83 FT.R. 229, Denault J.; Casetallanos v. Canada (Sec’y of State), [1995] 2 F.C. 211, Nadon
J., on appeal No. A-73-95 [hereinafter Casetallanos]. For cases in which certification was denied, see
e.g.: Antonio, supra note 145; Hercules, supra note 165; Visagaperunal, supra note 172; Sun, supra
note 175; Bouianova v. Canada (M.E.I.) (1993), 67 F.T.R. 74, Rothstein J.; Jeyakumaran v. Canada
(M.EL) (1994), 74 ET.R. 64, Wetston J.
19961
M. HURLEY- REFUGEE DETERMINATION DECISIONS
portance” in cases raising the same issue.”7 This, in turn, suggests that the certifica-
tion process has compounded the potential for ongoing, overall inconsistency.’
Consistency in administrative decision-making is valued.” ‘ Stephen Legom-
sky’s comprehensive article’ on forum selection in the immigration context notes
that, as a component of the adjudication process, a review system shares the goals
generally associated with administrative procedure, including accuracy, efficiency,
acceptability and, overlapping these three, consistency:
One benefit of consistency is enhanced stability. Conflicts among equally
authoritative bodies have ways of becoming reconciled eventually, either by
gradual evolution or by pronouncements from above. The mere presence of a
momentary conflict, therefore, can create at least the perception of imminent
change, leaving affected sectors of the population uncertain how to plan for the
future. Consistency reduces this uncertainty….
[Another] reason for valuing consistency is one that does not fit neatly within
any of the three traditional goals of administrative procedure. Consistency as-
sures equal treatment of similarly situated litigants. 6′
It is clear that Bill C-86 has virtually eliminated access to “pronouncements from
above” that would remedy inconsistency at the Trial Division level.
2″ Questions denied or granted certification by different judges may relate to issues that have been
controversial among Trial Division judges, such as the “test” for “change” prior to the Yusuf ruling of
the Court of Appeal, supra note 220. Contrast: Bhuiyan, supra note 126; Boateng, supra note 165;
Bonilla, supra note 165, with Hassanzadeh-Oskoi, supra note 163.
In declining to reconsider his denial of certification in Bhuiyan in the interests of consistency in
the certification process, MacKay J. commented that such consistency was only significant if the un-
derlying facts of different cases were clearly similar. Otherwise, the role of the Trial Division in
“finally disposing” of cases would be hampered. This, in his view, “would ignore the evident inten-
tion of Parliament … to provide an expeditious process for final determination of refugee claims by
the Trial Division except where the underlying facts of the claim and its disposition … give rise to a
serious question of general importance” (Bhuiyan v. Canada (M.E.L) (1994), 77 F.T.R. 286 at 289).
.’ In addition to differing views on substantive aspects, Trial Division judges have also taken dif-
ferent approaches to the question of the timing of certification itself. Contrast approaches in: lllanko,
supra note 255, Simpson J.; Cruz v. Canada (M.E.!.), [1995] F.C.J. No. 185 (QL), Simpson J., with
those in: Popov, supra note 153, Reed J.; Isa v. Canada (Sec’y of State) (1995), 91 FT.R. 71, 28 Imm.
L.R. (2d) 68, Reed J.; Paukovac v. Canada (M.C..) (1994), 89 F.T.R. 227, Gibson J.; Sheikh v. Can-
ada (Sec’y of State) (6 February 1995), No. IMM 2814-94, Gibson J.; Casetallanos, supra note 256,
Nadon J.
It is also worth noting, however, that on occasion the certification process has been resorted to
for the specific purpose of resolving Trial Division inconsistencies on substantive questions (see e.g.
Kadenko, supra note 213).
.. See: Dointar, supra note 22 at 784ff; H.W. MacLauchlan, “Some Problems with Judicial Review
of Administrative Inconsistency” (1984) 8 Dalhousie L.J. 435 at 446; Y-M. Morissette, “Le contr6le
de la competence d’attribution: thse, antithse et synthese” (1986) 16 R.D.U.S. 591 at 632ff.
26″ See S.H. Legomsky, “Forum Choices for the Review of Agency Adjudication: A Study of the
Immigration Process” (1986) 71 Iowa L. Rev. 1297.
262 Ibid. at 1313-14.
MCGILL LAw JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
Absence of consistency is potentially more harmful, in Legomsky’s view, when the
outcome of a case is particularly critical for the litigant, as it is in such “high stakes”
cases that “[t]wo of the interests that uniformity serves -predictability
and equality –
assume greater importance”..2 The refugee caselaw reviewed suggests that at present
neither a sensitivity to the high stakes issues nor an understanding of the need to re-
sort to substantive precepts to protect individual interests2″ has been able to deliver
uniform practice. In fact, process features of the present system militate against uni-
formity. The system seems unsuited to, or incapable of, remedying the potential for
actual harm arising from unpredictable and unequal results in the disposition of in-
dividual applications.
The most obvious harm is to genuine refugees who are mistakenly denied
status at the Board level, denied relief from that mistake because of an elusive, in-
dividually variable judicial norm and for whom no other form of relief is available.
The coherent development and application of legal principles relevant to refugee
determination also suffer. Uncertainty in the law, for all concerned –
including
Board members denied adequate guidance in their decision-making processes
becomes an altogether foreseeable consequence.
–
2.
The Very Severe Solution
That Trial Division judges reviewing Board decisions currently invoke a pat-
ently unreasonable test at all, or indeed any deferential, high threshold standard,
however erratically, is a legitimate area of concern. Under the present refugee de-
termination process, negative repercussions for genuine refugees seem equally in-
evitable with the application of an unduly severe standard. A second concern is that
a high threshold standard risks becoming increasingly entrenched as the operative
standard, not because it is appropriate, but because of considerations antithetical to
the purpose and promise of refugee determination under the Convention and do-
mestic law.
One such consideration involves the workload of Trial Division judges, which
has undoubtedly increased significantly as a direct result of Bill C-86.1″ Although
six “C-86 judges” were named to the Trial Division specifically to deal with the
backlog of cases transferred from the Appeal Division,”‘ the suggestion that this
2 Ibid. at 1321.
2″3 See Hutchison, supra note 224 at 320.
2
M The total number of proceedings commenced in the Trial Division jumped from just over 5,000
in 1992 to around 11,500 in 1993, with a decrease to approximately 8,500 in 1994:
Immigration has proved to be the most significant source of new proceedings in both
1993 and 1994, with 8,454 new proceedings commenced in 1993 (including [those]
transferred from the Court of Appeal) and 5,224 commenced in 1994 (FC.C. Report
1994, supra note 99 at 13).
:” See C-86 Proceedings, supra note 227 at 14:12-13 (P. Harder).
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
measure could effectively reduce the impact of such a sudden jump in the Trial Di-
vision’s inventory was questioned from the outset.” Parliament’s failure to transfer
any of the existing Trial Division caseload, such as the high volume of appeals un-
der the Unemployment Insurance Act,”
to other judicial fora has meant that the
thousands of leave and hundreds of judicial review applications in refugee matters
have placed enormous additional demands on stretched judicial resources.'” The re-
sulting time pressures affecting all trial Division judges on a continuous basis
might, over time and however inadvertently, influence the degree of scrutiny they
are prepared to bring to bear on C.R.D.D. decisions.” Judicial workload appears to
be considered an increasingly relevant factor in establishing the threshold of judi-
cial review generally” and may assume a particularly prominent, if unspoken, role
in high-volume areas, such as refugee determination.’ However unpalatable, given
the interests at stake in the refugee context and the procedural weaknesses imple-
mented by Bill C-86, the prospect of such an eventuality should not be underesti-
mated.
A second factor that might propel Trial Division judges into a more uniform
high threshold approach involves the generally negative public attitude toward
immigrants and queue-jumping “bogus” refugees. As Legomsky has noted, sharp
increases in the volume of immigrants to the United Kingdom and the United States
have been accompanied historically by corresponding increases in anti-immigrant
sentiment.2 The Canadian reaction to higher refugee claimant numbers since the
late 1980’s, fanned by anti-immigration predictions of growing unemployment and
recession, has been no different. In Legomsky’s view, judicial decisions may be in-
fluenced by contemporary political forces in society, either because judges are sus-
ceptible to or share prevailing public opinion, or because they are hesitant to defy
that opinion.
In theory at least, this factor, conflicting as it does with the pre-
sumptive judicial impulse to protect individual interests, may also influence the in-
consistency of approaches discussed earlier.
26 See ibid. at 7:64,7:67 (L. Waldman), 11:113, I1A:24 (J. Frecker).
267 Supra note 222. According to the Office of the Umpire, Trial Division judges sitting as umpires
heard 3,352 unemployment insurance appeals in 1992-1993. In 1993-1994, the figure was 3,811, with
a backlog of 7,435 pending appeals. The former L.R.C.C. had recommended since 1980 that this
workload be transferred to some other jurisdiction (see: L.R.C.C., Judicial Review and the Federal
Court, Report 14 (Ottawa: Department of Supply & Services, 1980) at 21 [hereinafter Judicial Re-
view]; L.R.C.C., Independent Administrative Agencies (Working Paper No. 25) (Ottawa: Department
of Supply & Services, 1980) at 156).
269 In their capacity as ex officio judges of the Trial Division, Appeal Division judges did decide
over a thousand leave applications in both 1993 and 1994 (see EC.C. Report 1994, supra note 99 at
10).29 See C-86 Proceedings, supra note 227 at 4A:68 (Inter-Church Committee for Refugees).
270 See “The 1990-91 Term”, supra note 20 at 31.
. The high-volume factor may also, of course, be relevant at the leave application stage.
272 See S.H. Legomsky, hnmigration and the Judiciary: Law and Politics in Britain and America
(Oxford: Clarendon Press, 1987).
211 See ibid. at 241-42.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
Leaving aside these considerations –
and even assuming that a consistently
applied severe standard might be preferable to none at all – Supreme Court of
Canada rulings and administrative law scholarship suggest persuasively that the in-
herently indeterminate and subjective patently unreasonable standard is incapable
of consistent implementation. Like Reed J., critics have noted that the distinction
drawn between “patently unreasonable” and “reasonable” or “clearly wrong” is
largely semantic. The lack of practical criteria of rationality against which to meas-
ure when a threshold level of acceptability has been crossed renders the test un-
workable as a basis for judicial assessment;… it also reduces the test to a threshold
of shock exercise in which “patently unreasonable” serves as new jargon for post
hoc rationalization once a court has decided, for whatever reason, to intervene or to
defer. Without the type of structure that might make patent unreasonableness a
helpful concept, courts are left with considerable discretion to embark on expansive
review. The other side of the coin, of course, is that egregious errors may elude ju-
dicial control on the basis that they are not sufficiently unreasonable.277 Much of the
refugee caselaw testifies to the validity of these criticisms and to the validity of
concerns related to the absence of prior or subsequent procedural safeguards in the
present refugee determination system.
Limited potential for consistent application is not unique to the patently unrea-
sonable test, and it is not the only objection to its application in review of C.R.D.D.
decisions. A fundamental question concerns whether that Board qualifies as an
“expert” agency entitled to curial deference irrespective of errors committed within
its jurisdiction. A review of the literature highlights a number of relevant points.
Rules of deference developed largely in a labour relations context may be less
compelling where, as in most areas of administrative law (including refugee law),
the relationship between the parties is not ongoing. Judicial review issues do not
lend themselves to monolithic standards; generalizations based on labour relations
experience fail to account for the distinctiveness of administrative regimes. Most
judicial review decisions from the Supreme Court of Canada have been concerned
with mature administrative tribunals having a claim to judicial respect; not all tri-
” See Roman, supra note 223, who suggests that judges appear to have at least three different
views as to what it is that is reviewed for patent unreasonableness: “Debate is likely to be protracted,
if not endless, if we have three significantly different ways of describing the same thing” (ibid. at
280).
277 For further discussion of these and related points, see: “Formalism”, supra note 12 at 372-73;
“Jurisdictional Review”, supra note 46 at 261; “Jurisdictional Error”, supra note 113 at 327; Roman,
ibid at 282; “General Theory”, supra note 225 at 331; “Transition”, supra note 233 at 153; Moris-
sette, supra note 259 at 613-14, 641; D.J. Mullan, “Judicial Deference to Administrative Decision-
Making in the Age of the Charter” (1986) 50 Sask. L. Rev. 203 at 208 [hereinafter “Age of the Char-
ter”]; B.A. Langille, “Judicial Review, Judicial Revisionism and Judicial Responsibility” (1986) 17
R.G.D. 169 at 207; R. Yalden, “Deference and Coherence In Administrative Law: Rethinking Statu-
tory Interpretation” (1988) 46 U.T. Fac. L. Rev. 136 at 153-54; B. Etherington, “Arbitration, Labour
Boards and the Courts in the 1980s: Romance Meets Realism” (1989) 68 Can. Bar Rev. 405 at 408;
D. Dyzenhaus, “Dicey’s Shadow” (1993) 43 U.T.L.J. 127 at 140.
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
bunals can make this claim. Furthermore, notwithstanding legislative choice of the
statutory body as decision-maker, deference should be earned, and “an undifferen-
tiating or universal application of the patently unreasonable standard is simply not
appropriate.”27′ A need exists for “differentiated deference or a sliding scale of enti-
tlement to court respect”.2.’
Generally acknowledged indicia of expertise include qualifications required for
appointment, length of term of office and security of tenure.” Although agency ex-
pertise is an important factor in determining when curial deference is appropriate,
the term “expert” is not synonymous with “infallibility”‘” nor necessarily indicative
of equivalent competence for all boards or panels within boards.” Unless adminis-
trative decision-makers engage in a process “which is consistent with [a] purposive,
field-sensitive approach””‘ and have the capacity and the expertise to do so, “hand-
waving references to judicial restraint will only ensure that the entire legal system
co-operates in a collective denial of responsibility.”‘
Turning this critical perspective on the Board, there is no doubt that the semi-
privative leave requirement in place since Bill C-55’s enactment reflects Parlia-
ment’s intention to limit high-volume review of-Board decisions as a means of en-
suring expedited finality of decision-making. On the other hand, refugee claimants
who satisfy that statutory screening requirement are also statutorily entitled –
since
the Bill C-86 “plumbing” exercise –
to judicial review. In this ambiguous circum-
stance, the Board’s mandate, structure and history are most relevant to determining
whether it is owed deference as an expert agency.
With respect to Richard J.’s opposing view, at a theoretical level, the Board’s
mandate does not entail the administration of a complex statutory scheme or broad
policy-making or regulatory functions of the sort considered by the Supreme Court
in Pezim to warrant a deferential posture. Some might argue the C.R.D.D.’s more
confined role in interpreting and applying the Immigration Act’s definition of Conven-
tion refugee confers “expert” status in practice. However, this factor, while relevant,
276 D.J. Mullan, “The Supreme Court of Canada and Jurisdictional Error: Compromising New
Brunswick Liquor?” (1988) 1 Can. J. Admin. L. & Pract. 71 at 97 [hereinafter “Compromising”]. See
also D.J. Mullan, “The Future of Canadian Administrative Law” (1991) 16 Queen’s L.J. 77 at 78
[hereinafter “Future”].
27 “Compromising”, ibid. at 100-101. See also: “Formalism”, supra note 12 at 372-73;
“Jurisdictional Review”, supra note 46 at 261; “Jurisdictional Error”, supra note 113 at 323; Roman,
supra note 223 at 289; “Book Review”, supra note 225 at 342ff; “Age of the Charter”, supra note
275 at 208; A. Harvison Young & R.A. Macdonald, “Canadian Administrative Law on the Threshold
of the 1990’s” (1991) 16 Queen’s L.J. 31 at 42-43.
276 See: “Compromising”, ibid. at 99-100; “Future”, supra note 276 at 78ff.
27 See T.D. Schiller, “Paccar of Canada Ltd.: Curial Deference for the 1990s?” (1990) 40 Admin.
L.R. 290 at 299.
2″ See: Roman, supra note 223 at 289; “Formalism”, supra note 12 at 378-79.
21 “Formalism”, ibid. at 376-77.
21 Ibid at 373.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
should be assessed in context. For the Board, the context includes its structure as a
regional, first-level decision-maker in matters of fundamental human rights that is
not subject to substantive review by a superior administrative body prior to judicial
review proceedings.
Equally importantly, the Board is a relatively immature agency whose credibil-
ity as an expert board has been continuously called into question since its 1989
creation. Criticism has focused largely on the appointment process and its conse-
quences for decision-making. Well before the introduction of Bill C-55, reports to
the federal government on reforming the refugee determination system stressed that
the “care with which individual decision-makers and panellists of the [proposed
Refugee Board] are selected will reflect the commitment of government to the true
nature of the determination process”.’ Prior to the Bill’s enactment, the need for
quality appointments to the C.R.D.D. was again emphasized as essential to the
agency’s ability to function credibly and with consistency. ‘ Nevertheless, the ap-
pointment process under Bills C-55 and C-86 remained non merit-based and open
to charges of patronage. The I.R.B. is not exceptional in this regard. The implica-
tions flowing from a flawed appointment process related to refugee determination,
however, are exceptional.
From 1989 to 1992, repeated highly publicized and embarrassing incidents of
non-expert behaviour by ill-equipped or ill-disposed appointees were followed by
the introduction, via Bill C-86, of a mechanism for judicial inquiry into the conduct
of C.R.D.D. members.’ Both the 1993 I.R.B.-commissioned report on deficiencies
in the conduct of refugee hearings” and the 1992 Draft Final Report of the former
L.R.C.C. ‘ on refugee determination procedures underscored the appointment proc-
ess’ compromising effects on the integrity of the refugee determination system and
the urgent need for introducing merit-based recruitment to the Board.’ More
“field-sensitive” (but shorter term) appointments by Minister of Citizenship and
Immigration Sergio Marchi from 1993 to 1995 did not alter the inherently political
Report to the Minister of Employment and Immigration, Refugee Determination in Canada by
G.W. Plaut (Ottawa: M.E.I., 1985) at 133-34 [hereinafter Plaut Report]. See also Report to the Minis-
ter of Employment and Immigration, A new refugee status determination process for Canada by E.
Ratushny (Ottawa: M.E.I., 1984) [hereinafter Ratushny Report], observing that “the fairness of the
process will depend not only upon procedural requirements, but also upon the quality of the decision-
maker” (Ratushny Report, ibid. at xii).
‘ See e.g. C-55 Proceedings, supra note 227 at 2:51-52 (M. Falardeau-Ramsay, former Chair,
I.A.B.), 5:26 (L. Waldman).
‘ See Immigration Act, supra note 83 at ss. 63.1, 63.2, as am. by Bill C-86, supra note 9 at s. 53.
The inquiry process, initiated by the Minister on the I.R.B. Chair’s recommendation, provides for a
member’s removal from office for misconduct or incompatibility with the execution of office.
” See Report to the Chairperson of the Immigration and Refugee Board, Rebuilding Trust: Report
of the Review of Fundamental Justice in Information Gathering and Dissemination at the hmnigra-
tion and Refugee Board of Canada by J.C. Hathaway (Ottawa, 1993) [hereinafter Hathaway Report].
See Determination, supra note 243.
See: Hathaway Report, supra note 286 at 19-26, 37-42; Determination, ibid. at 141-46.
1996]
M. HURLEY- REFUGEEDETERMINATIONDECISIONS
and discretionary nature of the process. In fact, the Board’s vulnerability to politi-
cally-charged dissent among members became particularly apparent in the wake of
these appointments, paralleled as they were by the non-renewal of many former ap-
pointees.
In the fall of 1994, the Liberal-appointed I.R.B. Vice-Chair resigned following
his suspension by the Conservative-appointed Chair on the basis of C.R.D.D.
members’ allegations that they were “bullied” for higher refugee acceptance rates.
This event accentuated the agency’s political volatility and deepened divisions
within it.’ Related charges and counter-charges by polarized present and former
members, refugee advocates and refugee policy critics over the C.R.D.D.’s role and
practice multiplied over this period in a highly open forum,’ further compromising
the agency’s ability to function effectively.91 Since this low-water mark, charges of
incompetence have continued to be levelled against the agency.”
In short, the Board’s claim to expertise over its brief but controversial history
has not been that of a stable, mature agency. These factors, combined with weak-
nesses in procedural protection afforded refugee claimants, seem to argue strongly
that a high threshold, deferential standard is simply unsuitable under present cir-
cumstances.’
In March 1995, the Minister of Citizenship and Immigration and the I.R.B.
Chair simultaneously announced initiatives to revise the refugee determination
hearing to a special board of inquiry model,2″ introduce single-member Board pan-
2 “The resignation pre-empted the first Federal Court inquiry initiated under the Bill C-86 mecha-
nism (see supra note 285).
See: “Choosing the Refugee Board” The [Toronto] Globe and Mail (15 November 1994) A22; S.
Delacourt, “Marchi Accused of Impropriety in Refugee Board Squabble” The [Toronto] Globe and
Mail (15 November 1994) A4; L. Sarick, “Power Struggle at IRB Could Be Fatal, Observers Say”
The [Toronto] Globe and Mail (18 November 1994) A5; R. Platiel & E. Greenspon, “Schelew Re-
signs; Inquiry Cancelled” The [Toronto] Globe and Mail (10 December 1994) Al; S. Fine, “IRB Fu-
ror Reflects on Lawyers” The [Toronto] Globe and Mail (10 December 1994) A10-11; “But What
About the IRB?” The [Toronto] Globe and Mail (13 December 1994) A22.
291 In January 1995, the I.R.B. was reported to have instituted a formal mechanism for dealing with
public complaints about Board members in an attempt to restore public confidence (see A.
Thompson, “Refugee Panel Gets System for Complaints” The Toronto Star (4 January 1995) A9).
“‘ See: “Refugee Board Called Incompetent” The [Toronto] Globe and Mail (15 March 1995) A4;
D. Stoffman, “Canada’s Farcical Refugee System” The [Saint John] Telegraph Journal (6 September
1995) A9; A. Norris, “Refugee Board Not Rife With Incompetence- Official” The [Montreal] Ga-
zette (2 October 1995) A4.
29 It has been suggested that “the case for deferring to an agency’s interpretation is less compelling
when the decision under review has apparently jeopardized the very interests that the agency was
created to protect” (“Jurisdictional Review”, supra note 46 at 259). See also text accompanying notes
344ff, below.
294 See Immigration and Refugee Board, News Release, “Refugee Status Determination Process to
be Strengthened” (3 March 1995) [hereinafter I.R.B. Release].
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
els ‘ and reform the appointment process. The first two reforms were criticized by
many on the basis that they would aggravate rather than remedy existing Board
problems, with serious implications for the safety of refugee claimants.” The pro-
jected “highly unusual” move to single-member Boards ‘ was seen as one which,
owing to its potentially negative impact, ought at least to have been accompanied
by an internal review mechanism.”‘ Furthermore, although the proposed board of
inquiry model purported to be along the lines of recommendations in the Hathaway
Report, the author was quoted as stating that the model would not solve existing
problems and would, in fact, change little.’
Ironically, I.R.B. documents outlining the “improved” refugee hearing process
cited Richard J.’s conclusions in Sivasamboo as to the Board’s entitlement to sig-
nificant curial deference in support of a claim of increasing acknowledgment of
Board expertise in Federal Court decisions.” In fact, no such broad conclusion can
be drawn from the cases reviewed. As discussed previously, Trial Division judges
continue to show highly variable attitudes toward the Board generally and, on oc-
casion, more or less openly question Board members’ grasps of procedural and
substantive issues.?”
The third March 1995 reform involved the creation of a five-member Advisory
Committee, mandated to subject all applications for appointment to the I.R.B. to a
selection process based on objective standards provided by the Minister and to form
“5 See News Release 95/03, supra note 107.
“‘See: “Marchi Moves ‘Weakening’ Board” The [Victoria] Times-Colonist (3 March 1995) A14;
D. Rinehart, “Refugee-Board Changes Will Make System Worse, Critics Say” The [Montreal] Ga-
zette (7 March 1995) A8; M. Farrow, “Tougher Refugee Hearings Feared” The [Vancouver] Sun (4
March 1995) A22.
29 D. Rinehart, “Marchi to Revamp Refugee Boards, Appeal Process” The [Montreal] Gazette (2
March 1995) A10 [hereinafter “Revamp Refugee Boards”].
” See: M. Comellier, “Les aspirants r6fugi6s seront entendus par un seul commissaire” La Presse
(3 March 1995) B5; L. Sarick, “Refugee Board to Assign Single Person Panels” The [Toronto] Globe
andMail (3 March 1995) A4 [hereinafter “Single Person Panels”].
See: L. Sarick, “Refugee Hearing Process Reformed” The [Toronto] Globe and Mail (4 March
1995) A5. See also M. Valpy, “The Suspicion of a Gelded Refugee Process” The [Toronto] Globe and
Mail (7 March 1995) A2.
” See “Backgrounder” in I.R.B. Release, supra note 294. The I.R.B. Chair reiterated this assertion
in testimony before the House of Commons Standing Committee on Citizenship and Immigration
(see Minutes of Proceedings and Evidence of the House of Commons Standing Committee on Citizen-
ship and Immigration (Ottawa: Queen’s Printer, 1995) at 41:7).
“‘ See e.g.: Thavanayagapathy, supra note 211; Njoko v. Canada (M.E.L), [1995] FC.J. No. 119
(QL), Jerome A.C.J.; Canada (M.C.L) v. Salazar, [1995] F.C.J. No. 562 (QL), Nadon J.; Diallo v.
Canada (M.C.I.), [1995] EC.J. No. 893 (QL), Noel J., as well as Appeal Division rulings such as Ni-
zamov v. Canada (M.E.I.), [1995] EC.J. No. 805 (QL). It is worth noting that the Hathaway Report
excerpt cited in Sivasamboo, supra note 179 at 757-58, in support of Richard J.’s view of the Board
as “expert” is an introductory description of the purported purpose for which the C.R.D.D. was estab-
lished, not a conclusion. As noted earlier, the report was in fact critical of the Board.
1996]
M. HURLEY- REFUGEE DETERMINATION DECISIONS
an inventory of suitable candidates.”‘ In announcing this important policy shift,
Minister Marchi stated his commitment “to recommend to the Governor in Council,
only those candidates for appointment or re-appointment who have been recom-
mended by the Advisory Committee”.’ At the same time, negative reaction to pro-
posed single-member Board panels was anticipated by the suggestion that the pro-
posed Advisory Committee could or would resolve the thorny competence issue.”
Although some critics greeted this initiative more positively,3 ” others expressed
reservations on grounds that the selection process would not affect present ap-
pointments and that the Advisory Committee might not be sufficiently isolated
from political pressures.’ While the Advisory Committee model is less than ideal
in that it reflects ultimately discretionary ministerial policy rather than a legislated
norm,3′ it seems to offer the best hope to date for developing the Board’s expert
potential. However, even assuming the Advisory Committee can, as is hoped, posi-
tively influence the overall competence of decision-making at the Board level, it is
not at all apparent that this factor alone could resolve other enduring concerns
about the present system. These continue to centre around the question of whether,
in light of consequences of mistakes in refugee determination, first-level panels
composed of single decision-makers named to short terms and whose decisions are
not subject to internal review can, or should, be considered expert bodies entitled to
broad immunity from the sole avenue of review available to refugee clairiiants. In-
stitutional factors such as these arguably outweigh the benefits of greater compe-
tence from the outset.
… See Citizenship and Immigration Canada, News Release, “Minister Marchi Announces Advisory
Committee for Selecting Board Members” (2 March 1995) [hereinafter C.I.C. Release].
3 “Backgrounder”, reproduced in C.I.C. Release, ibid
3o Another Minister’s statement discussing the proposed committee declared:
We want the best people for the job [of Board member]. It is mandatory that the best
people are named to the board because the reduction in panel size requires unques-
tioned and unflinching competence from every member (Speaking notes of Minister S.
Marchi for the House of Commons (2 March 1995), reproduced in C.I.C. Release,
ibid).
Hathaway described the move as “philosophically a really important change” that would “[work]
very well, and … [avoid] a whole lot of political flak” (J. Miller, “Merit System Promised for Refugee
Board” The Ottawa Citizen (3 March 1995) A3).
6 See: R. Boulakia, “Board ‘Reform’: Plus 9a change …” Refugee Update (Spring 1995) at 9;
“Single Person Panels”, supra note 298. As has been observed, only two members “will clearly and
necessarily be ‘advocates for refugees’ in the sense used by both Board supporters and detractors.
These are the representatives from the bar and from non-governmental organizations involved in
refugee matters” (Library of Parliament, The Immigration and Refugee Board: Proposed Changes
(Background Paper No. BP-399-E) by M. Young (Ottawa: Library of Parliament, 1995) at 15). Of the
remaining members, one is chosen by the government from the general public; both the Committee
Chair and the I.R.B. Chair are order-in-council appointees, as are the ex-offlcio I.R.B. Vice-Chairs.
” This feature is highlighted by the February 1996 federal Cabinet shuffle, in which former Minis-
ter of Citizenship and Immigration Marchi, who had given his personal undertaking to respect the
Committee’s recommendations, was moved to another portfolio.
MCGILL LAW JOURNAL/REVUE DE DRO1TDE MCGILL
[Vol. 41
IV. The Promise: Dusting Off a Modest Proposal
A consideration of the problems associated with judicial review principles and
practice in the context of refugee determination leads to the issue of remedial op-
tions. If, as contended, judicial review of refugee determination decisions has not
yet provided an effective route for addressing the current system’s potential for
harm to genuine refugees, the question becomes how best to reduce that potential.
A. The Tinkering Approach
The tinkering approach contemplates defining a standard of review in light of
the refugee context and the Board’s characteristics that would prove capable of
consistent implementation by Trial Division judges. A number of factors militate
against such an approach. First, it probably amounts to a largely academic exercise
in circumstances calling for more immediate remedial potential. Numerous reforms
of considerable theoretical interest have been proposed over the years but have had
little noticeable effect on judicial practice.”‘
Second, although the notion of tailoring the scope 6f permissible judicial inter-
vention to specific agencies has been thought to have considerable merit,” such
scaling is also viewed as an “excruciatingly difficult task”.”‘ Even if an appropriate
“scale” could be articulated, the “standard” approach presumes that uniformity and
consistency of application are attainable objectives in judicial review'” –
a pre-
sumption which has yet to prove justified. Standards are by their nature conceived
in value-laden terms such as “rational”, “fair”, “reasonable”, “substantial” and
“relevant”, all susceptible to myriad subjective interpretations or usages and
“threshold of shock” inconsistency: “Law tends to be particularly uncertain when
judges apply semantic formulae which convert subtle or highly subjective differ-
ences in degree to the language of qualitative distinction.””‘2 Although the present
refugee determination system certainly calls for a low-threshold, non-deferential
standard to redress the absence of procedural safeguards available to refugee claim-
” See e.g.: “General Theory”, supra note 225; N. Lyon, “Continuing Search for a General Theory
of Judicial Review of Administrative Action for Legality” (1980) 58 Can. Bar Rev. 646; N. Lyon,
“Reforming Administrative Law” (1989) 2 Can. J. Admin. L. & Pract. 315; D.J. Mullan, “Natural
Justice and Fairness -Substantive
as well as Procedural Standards for the Review of Administrative
Decision-Making?” (1982) 27 McGill L.J. 250; J.M. Evans, “Developments in Administrative Law:
The 1988-89 Term” (1990) 1 Supreme Court L. R. (2d) I [hereinafter ‘The 1988-89 Term”].
3″‘ See: “Book Review”, supra note 225 at 341, 349-50; D.J. Mullan, “The Federal Court Act: A
Misguided Attempt at Administrative Law Reform” (1973) 23 U.T.L.J. 14 at 43, 51.
31’ “Book Review”, ibid. at 343.
311 See “The 1990-91 Term”, supra note 20: “There must be standards” (ibid. at 37).
312 Roman, supra note 223 at 280. See also: “General Theory”, supra note 225 at 336; “Protection”,
supra note 235 at 284.
19961
M. HURLEY- REFUGEEDETERMINATIONDECISIONS
ants, it seems doubtful that such a standard, if capable of articulation,”‘ would effect
that redress or reduce apparent –
arbitrariness in judicial deci-
sion-making.
albeit good faith –
The problem of defining just standards for judicial review is not exclusive to
the refugee context. The refugee context, however, is exceptional in the nature of
consequences flowing from the problem and in the scope of statutory procedural
restrictions imposed upon refugee claimants, making the problem particularly
acute. The harmful results of inconsistency at the judicial level do not, however,
appear to have their roots at that level, nor do they seem capable of meaningful re-
dress through tinkering with standards. Administrative law scholars have argued
that, in general, judicial review is unlikely to compensate for inadequacies of ad-
ministrative decision-making.”
B. The Body-Work Approach
Substantive reconsideration of negative decisions has been accepted as an inte-
gral feature of a fair refugee determination system since long before the Singh de-
cision. At the international level, the U.N.H.C.R. Executive Committee, of which
Canada is a member, recommended in 1977:
If the applicant is not recognized, he should be given a reasonable time to ap-
peal for a formal reconsideration of the decision, either to the same or a differ-
ent authority, whether administrative or judicial, according to the prevailing
system.’
Although not binding, this recommendation has been viewed as “a minimum stan-
dard to which the Canadian system should conform,'”” and one which is not satis-
fied by providing only for recourse to the Federal Court by way of leave. At the
domestic level, refugee advocates have sought without success, since before the
hnmigration Act, 1976 was proclaimed in 1978, to warn successive governments
313 See e.g. “The 1988-89 Term”, supra note 308 at 57-58, where a “rights-based approach” is dis-
cussed.
“‘ See H.W. Arthurs, “Rethinking Administrative Law: A Slightly Dicey Business” (1979) 17 Os-
goode Hall L.J. 1:
Judicial review, seen as general deterrence, is a less promising method of securing
proper administrative decisions at first instance than a host of other methods …
Constructive measures to enhance the original quality of decisions will not result in
perfection, but they will, in the aggregate, ensure greater justice for more people than
could possibly benefit from any system of judicial review. Such measures would in-
clude … more careful training of administrative decision-makers, systems of internal
appeal … (Arthurs, ibid. at 41-44).
.. Report of the Executive Committee of the Program of the U.N. High Commissioner for Refugees,
UN GAOR Doe. AIAC.96/549 (1977) para. 53.6.
“‘6 Determination, supra note 243 at 101-102.
MCGILL LAW JOURNAL/REvuE DE DROITDE MCGILL
[Vol. 41
that the absence of adequate mechanisms for review on the merits of negative deci-
sions in successive refugee determination systems would result in predictable inef-
ficiencies and inevitable injustices to genuine refugees.”‘
The following discussion concentrates on internal review. It opens with a brief
review of the feasibility and theoretical advantages of judicial appeal on the merits
before the Federal Court. The discussion assumes the Board’s continued jurisdic-
tion over first-instance decisions.”‘
1.
Judicial Appeal on the Merits
Refugee claimants have never had an appeal on the merits to the Federal Court;
appeal under Bill C-55 was a narrower recourse, given both the leave requirement and
the merging of judicial review and appeal grounds in immigration and Federal Court
statutes.?9 Although a number of factors might seem to favour this form of appeal, it
has not been widely advocated as a mechanism of formal reconsideration.”‘ These in-
clude judicial impartiality, judicial independence from administrative and political in-
stitutions and influence, broad authority to reverse erroneous first instance decisions
and, in theory, avoidance of ambiguities of process, result and posture associated with
judicial review.
These benefits, while undoubtedly significant, seem on balance largely outweighed
by other considerations. In the current political climate, a minimally viable review
mechanism would have to satisfy not only Convention obligations to protect genuine
refugees, but also political obsession with fiscal constraints, expeditious process and
real or perceived abuse of the refugee system. One might certainly argue these latter
factors ought not to exert a determining influence in light of the interests at stake; it
would be difficult, however, to maintain that they do not. Realistically, then, not only
would cost and delay factors associated with judicial appellate review2 ‘ prove politi-
cally unacceptable, but equally importantly, the option would require legislative back-
pedalling to a more liberal version of an already discarded appellate option. The politi-
cal choice of narrow judicial review, with its surrounding procedural impediments, as
the sole judicial recourse available under the Immigration Act, does not appear likely to
be countermanded.
3” See e.g.: C-55 Proceedings, supra note 227 at 2:58 (J.B. Stem), 5:6-7, 13, 34 (L. Waldman), 6:41
(K. Zaifman), 6:159 (Nanaimo Immigrant Settlement Society and Refugee Committee). See also
“C.B.A.O. Submission”, supra note 227 at 1.
‘” Recent upheavals at the Board prompted some observers to call for the replacement of lay Board
members as first instance decision-makers by immigration judges or the transfer of Board jurisdiction
to a refugee division of the Federal Court (see: EC. Leclerc, “Faulty Justice: Immigration Tribunals
Have Been a Disaster” The [Montreal] Gazette (14 November 1994) B3; Stoffman, supra note 292).
319 See text accompanying notes 88-98, above. See also: C-55 Proceedings, supra note 227 at 8:69
(Canadian Council of Churches); “C.B.A.O. Submission”, supra note 227 at 57-59.
320 In 1992, the Refugee Lawyers Association did recommend that Federal Court review of Board
decisions be on the merits (see C-86 Proceedings, supra note 227 at 5:64).
32, See T.G. Ison, “Appeals on the Merits” (1992) 30 Osgoode Hall L.J. 139 at 146.
19961
M. HURLEY- REFUGEEDETERMINATION DECISIONS
On the assumption that appeals on the merits to the Federal Court were either po-
litically palatable or feasible, it is open to question whether they would, in theory, pro-
vide the most appropriate avenue for formal reconsideration of negative Board deci-
sions. The 1985 Plaut Report suggested that a fair refugee determination model might
consist of an initial hearing by a three-member panel followed by an appeal as of right
to the Federal Court of Appeal for purposes of compliance with the 1977 recommen-
dation of the U.N.H.C.R. Executive Committee?2 In this structure, Rabbi Plaut consid-
ered an appeal on the merits of “questionable value” owing to the relative inexpertise
of the Federal Court in areas affecting refugee determination?’
Ten years later, the Federal Court remains a general court of law, not a specialist
court. As such, arguably, it is still not the appropriate body to conduct an appeal on the
merits in the absence of the expert first instance Board panel envisaged by Rabbi
PlautY Federal Court judges have undoubtedly acquired experience in dealing with is-
sues relevant to refugee determination. On the other hand, the breadth of the Federal
Court’s adjudication jurisdiction and the ensuing diversity and volume of workload
militate against its ability to develop special expertise in refugee law issues, such as
applicable international norms and country conditions.”
In theory, the review and remedial scope of judicial appeal on the merits before the
Federal Court would present advantages over the present judicial review process. It has
been argued, however, that an “overriding difficulty” with statutory provision for ap-
peals on the merits to courts of general jurisdiction is that it might not work. Ison sug-
gests: “[W]hen legislatures have provided for an unrestricted appeal from a tribunal to
a court, the judges have tended to shrink the scope of that appellate jurisdiction to a
point at which it is indistinguishable from judicial review. 3.. The explicit merging of
appellate and judicial review values and roles in the post-Pezim era seem to argue
against introducing a potentially “shrunken” judicial appeal on the merits as the sole
3 See Plaut Report, supra note 283 at 108-10.
3 Ibid. at 170-71.
3 See: “C.B.A.Q. Submission”, supra note 227 at 59; C-86 Proceedings, supra note 227 at 10:17
(Rabbi Plaut). But see also C-86 Proceedings, ibid. at 8:75 (Refugee Project, Lawyers Committee for
Human Rights, New York), describing the “evidentiary merits element” to refugee appeals to federal
courts in the United States.
.. Nor does the creation of a specialized Immigration Division equipped with the human, research
or documentation resources, arguably necessary to provide a meaningful appeal on the merits, seem a
realistic prospect. The idea of a special administrative law division within the Federal Court was
raised as long ago as 1977, on the basis of workload considerations that would seem equally germane
in 1995 (see Administrative Law, supra note 114 at 66. See also Judicial Review, supra note 267 at
24). Although Ison has noted that appellate courts of specialized jurisdiction have advantages over
generalist courts, he also points to a strong tradition against creating such courts (see Ison, supra note
321 at 152-53). Presumably, a specialized appellate Immigration Division within the Federal Court is
just as unlikely.
Ison, ibid. at 151.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
remedial avenue for formal reconsideration open to refugee claimants.’ Such a
mechanism might differ little in practice from the present process and might fail to ad-
dress concerns outlined under previous headings, including judicial preoccupation with
rationing of resources in a high-volume area like refugee determination.
Finally, creation of an appeal structure within the Federal Court might hinder nec-
essary reforms at the Board. Arguably, judicial appeal would not address problems of
primary adjudication or enhance Board efficiency and might divert the focus away
from improvements in the Board’s expertise and structures.” Under present circum-
stances, the I.R.B. would appear better suited institutionally to build on existing re-
sources in order to deliver the expertise required for quality decision-making at first
instance, and to develop expert structures for formal reconsideration that would satisfy
concerns for cost effectiveness and expeditiousness.
2.
Internal Appeal
Internal appeal is the review option consistently favoured by refugee advocates
and frustrated at the political level since the beginning of contemporary refugee
determination in Canada.
Three expert reports prepared for the Liberal government during the early
1980’s addressed already existing systemic problems of the process introduced in
1978. The authors called for comprehensive change to that inaugural system to re-
duce delays that promoted abuse and imposed hardship for those with legitimate
claims””9 and warned against reliance on the courts’ “blunt solution” to address in-
adequacies in the refugee determination process.”‘ A full hearing for initial deter-
mination as well as a clearly-defined administrative appeal to assess the factual and
legal correctness of that initial decision were judged essential.” It was seen as im-
perative, in light of the consequences of mistakes, that refugee claimants benefit
from standards of procedural protection equivalent to those afforded through the
criminal system.’.’2 The government was urged to take action and cautioned against
complacency that would erode Canada’s “fortunate position” of being able to pre-
vent potential problems from materializing.”
… An appeal on the merits in refugee cases would have to extend to examination of factual issues, a
task judges have traditionally been unwilling to undertake (see Ison, ibid at 152).
3 See ibid at 143.
See Report to the Minister of Employment and Immigration, Illegal Migrants in Canada by
W.G. Robinson (Ottawa: M.E.I., 1983) at 107-108.
“0 Task Force on Immigration Practices and Procedures, The Refugee Status Determination Process
(Ottawa: M.E.I., 1981) at xix, 76. Previously, the L.R.C.C. had recommended that “immigration ap-
peals” should be transferred from the Federal Court to a specialized administrative tribunal (see Judi-
cialRev iew, supra note 267 at 21, 24).
331 See Ratushny Report, supra note 283 at 45, 53.
332 See ibid at 33.
3 Ibid. at 21.
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
In April 1985, the Plaut Report commissioned by the Progressive Conservative
government expanded upon these concerns. It cautioned that tension between gov-
ernment strategies relating to immigration enforcement and refugee determination
“must not … shade the ensuing process to the detriment of the [refugee] claimant” 33′
Of three possible models proposed for the refugee determination process, 35 two in-
corporated internal appeal procedures, with subsequent appeal by way of leave to
the Federal Court of Appeal seen as adequate, provided the grounds of appeal were
sufficiently broad. The third provided, as described in the previous section, for ini-
tial determination by a three-person panel of experts, with appeal as of right to the
Federal Court of Appeal in the absence of an internal mechanism for formal recon-
sideration.
Government rejection of an internal appeal mechanism for substantive review
of the merits of Board decisions prior to limited-access court proceedings prompted
heated repudiation of the system introduced by Bills C-55 and C-86. Apologists de-
fended the chosen system on the basis that “front-loading” it with quality Board
hearings ensured adequate procedural protection.33 ‘ In response, it was argued that
the absence of safeguards for non-criminal refugee claimants would fail to remedy
inconsistent decision-making by a regionally-based Board and, given the nature of
decisions being made, increase the chances of mistakes going unchecked under nar-
row grounds of appellate or judicial review.3
Not surprisingly, the Bill C-55 system led Rabbi Plaut to conclude that it “is not
a bill to determine refugee status primarily; it [is] a bill on how to deport people
primarily”? With the advent of Bill C-86, while abolition of the credible basis tri-
bunal freed resources that might have been allocated to an internal appeal struc-
ture,” ‘ proposals along those lines were again rejected in favour of increasingly re-
strictive procedures.
Governmental recognition of weaknesses in the refugee determination system
owing to the absence of such a mechanism has continued to surface. In January
1994, then Minister of Citizenship and Immigration Marchi, a strong supporter of
internal appeal while in opposition and during the first days of the Liberal govern-
ment,’
commissioned yet another study into the possibility of creating an appeal
“‘ Plaut Report, supra note 283 at 18-19. The need for maintaining a clear distinction between
immigration and refugee processes has been a source of ongoing concern (see e.g. Determination, su-
pra note 243 at 53ff.).
’33 See Plaut Report, ibid. at 107-19.
336 See: C-55 Proceedings, supra note 227 at 9:21-22 (Minister Weiner); C-86 Proceedings, supra
note 227 at 2:35 (Minister B. Valcourt).
33 See supra notes 227, 228.
… C-55 Proceedings, supra note 227 at 6:7.
… See: Detennination, supra note 243 at note 136, p. 107; C-86 Proceedings, supra note 227 at
5A: 19 (Canadian Council For Refugees).
3 See A. Thompson, “Minister Wants a More Hospitable Refugee Policy: Proposals Would Make
Appeals System Easier for Failed Claimants” The Toronto Star (28 November 1993) B4.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
process for refugee claimants within the I.R.B. The March 1994 Davis-Waldman
Report proposed a centralized, written appeal of all negative C.R.D.D. decisions as
the “best compromise to satisfy the competing concerns of cost-effectiveness, effi-
ciency and fairness”.-“‘ Despite expectations that the shift to single-member Board
panels would be tempered by the introduction of such a process, ” internal review
was once again “conspicuously absent”‘
from the Minister’s March 1995 an-
nouncement.
In fact, research into the actual operation of Bill C-55 confirmed predictions
that inconsistent decision-making by the regionally-based Board would be an ‘inevi-
table repercussion of failing to provide for internal review.'” In addition, Trial Di-
vision caselaw also appears to bear out the concern that, in practical terms, judicial
review has not addressed inconsistencies in decision-making at the Board level.”‘
The I.R.B. has made various attempts intemally to address problems of inconsis-
tency; these have themselves proved controversial and have given rise to judicial re-
view.’ Moreover, they do not represent or replace substantive review. The March
” Report to the Minister of Citizenship and Immigration, The Quality of Mercy: A study of the
processes available to persons who are determined not to be refugees and who seek humanitarian
and compassionate treatment by S. Davis & L. Waldman (Ottawa: March 1994) at 46 [hereinafter
Davis-WaIdman Report].
342 Up to days before the March 2 announcement, press reports suggested a system of review of
negative decisions would be introduced (see: L. Sarick, “Immigration Board Calculates a Way to Save
$3-Million” The [Toronto] Globe and Mail (24 February 1995) A2 [hereinafter “Calculates”];
“Revamp Refugee Boards”, supra note 297).
“Single Person Panels”, supra note 298.
According to the L.R.C.C. report:
One of the major problems observed … is great inconsistency in the way in which cases
essentially indistinguishable on their facts are decided [by the C.R.D.D.]. We also ob-
served wide variation from region to region in the rate of acceptance of refugee claim-
ants from particular countries. Regional disparities in refugee acceptance rates confirm,
not the supposed value of independence of individual decision-makers, but a failure of
fairness within the system. If the refugee determination process in Canada is to be
anything more than a lottery, it is imperative that measures be taken to promote greater
consistency (Determination, supra note 243 at 82).
Board inconsistency has been acknowledged by government (see e.g. C-86 Proceedings, supra
note 227 at 2:35, for then Minister Valcourt’s justification of Bill C-86 amendments on the basis that
they would ensure greater consistency).
-‘4 See Determination, ibi at 82. Trial Division judges before whom the issue is raised are placed
in an at best awkward position (see: Tawflk, supra note 126; Cader v. Canada (ME..) (1 June 1993),
No. A-181-93, Reed J.; Shahzanian v. Canada (M.C.L), [1995] EC.J. No. 724 (QL), MacKay J.;
Shivatharshini v. Canada (M.C..), [1995] EC.J. No. 1213 (QL), McGillis J.; Ponnampalam v. Can-
ada (M.C..) (1995), 30 Imm. L.R. (2d) 178, Muldoon J. (application to stay removal order).
346 See Trial Division decisions discussing a “consistency project” initiated by Board officials in:
Gyimah v. Canada (M.E.L) (1994), 83 ET.R. 34, 25 Imm. L.R. (2d) 132, Simpson J.; Asare v. Can-
ada (M.C..) (1995), 94 FT.R. 283, Rothstein J.; Oduro v. Canada (M.E.L) (1993), 73 F.T.R. 191,
McKeown J. For Appeal Division decisions upholding the Board’s “reasons review policy”, see:
Weerasinge, supra note 8 at 337; Bovbel v. Canada (M.E.L), [1994] 2 RC. 563, 18 Admin. L.R. (2d)
1996]
M. HURLEY – REFUGEE DETERMINATION DECISIONS
1995 “Operational Framework” for the new board of inquiry refugee hearing model
might be interpreted by some as a small step toward review. Perhaps seeking to de-
flect adverse reaction to the government’s introduction of down-sized Board panels
and its failure to provide for an internal review structure, the document indicated,
under the heading “Other Initiatives”, that the I.R.B. would “supervise more closely
the [Board’s] exercise of its jurisdiction to reopen cases where there may have been
a breach of the principles of natural justice so as to ensure that a coherent and sys-
tematic approach is adopted”? 7 However, this statement of intention merely reflects
already existing authority to reconsider “natural justice” cases. Even on a generous
reading, the exercise of that authority does not qualify as internal review, as that
term is generally understood, and would benefit relatively few claimants.
The prospect of lone decision-makers does indeed heighten the need for a review
mechanism capable of correcting erroneous Board decisions and promoting institu-
tional coherence. Expert reports have recommended means to achieve those goals
while reducing the need for recourse to the Federal Court. The Davis-Waldman Report
and the Draft Final Report of the former L.R.C.C., for example, proposed mechanisms
having a number of common elements, which echoed aspects of options considered by
Rabbi Plaut a decade ago. In all three proposals, the effectiveness of internal review
was premised on the expertise of those recruited to conduct it? Most importantly, the
internal processes proposed would apply as of right. While all three reports have either
assumed or sanctioned limited access to subsequent judicial proceedings, under certain
circumstances, they have viewed internal substantive review as of right – because it
has the greatest potential to correct mistakes made at first instance –
as necessary to
offset this otherwise-unacceptable restriction.
The two most recent reports underscore other important distinctions between the
level of protection afforded by internal appeal or review and by judicial review by way
of leave. First, because a principal and explicit function of internal appeal or review
would be to correct mistakes, neither report suggested that first-instance decision-
makers should be entitled to deference for reasonable or any other errors.”‘ Both re-
ports also considered it essential that the reviewing body not be limited to the record of
the original proceeding –
as is the case in judicial review proceedings – but be able
to consider new evidence relevant to the claim. Furthermore, internal review or appeal
169, rev’g [1994] 1 F.C. 330. For decisions relating to the use of “boiler plate” reasons in Board de-
cisions, see: Fainshtein v. Canada (M.C.L), [1995] FC.J. No. 941 (QL), Simpson J.; Singh v. Canada
(M.C.I.) (27 January 1995), No. A-988-92, Noel J.
‘ Immigration and Refugee Board, Refugee Status Determination Process: Specialized Board of
Inquiry Model (Ottawa: I.R.B., 1995) at 12.
… This stands to reason, since the creation of such a mechanism would, in and of itself, be of lim-
ited value in the absence of greater attention to development of the Board’s potential as an expert
agency through merit-based appointments and adequate training.
… This makes sense. A deference requirement would complicate the process, undermine the pur-
pose of review at the administrative level and arguably amount to a mini judicial review exercise in-
distinguishable for all practical purposes from the real thing.
MCGILL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
would involve the entire record rather than the partial record upon which leave appli-
cations determining access to judicial review are generally decided. “‘
The L.R.C.C. report characterized internal review as “a complement to, rather than
a replacement for judicial oversight”.” ‘ In fact, both reports saw internal review as a
means of avoiding or of simplifying subsequent court proceedings:”
Since there would now be a two stage process involving examination of
credibility issues and legal issues, it is anticipated that the role of the Federal
Court would be significantly diminished. The number of cases in which suc-
cessful applications for judicial review would be launched would be far fewer
than at present” 3
This consideration highlights a further fundamental distinction between internal and
judicial review. Under both proposals, the reviewing body’s broad authority to remedy
mistakes might be expected to achieve the much desired expedited finality of decision-
making in a significant number of refugee claims.3″ The remedial limitations of judicial
review,353 on the other hand, produce prolonged, costly and potentially repetitive”
shuttling of refugee claimants between the Board and the Trial Division of the Federal
Court because of errors, omissions or misidentified or unresolved issues at first in-
stance. In this light, the rejection of internal review seems doubly unfortunate as well
as short-sighted. Reduced operating costs for the I.R.B. and the Federal Court might be
expected to more than compensate for those associated with an internal review
mechanism.’
30 See Davis-Waldman Report, supra note 341, which foresaw that, as a result, the task of Federal
Court judges at the application for leave stage would be eased by ensured access to an adequate
record of prior proceedings (see ibid. at 48).
3.Detennination, supra note 243 at 124.
332 See ibid at 106, 107-108, 113, 126, 136-37.
3 Davis-Waldman Report, supra note 341 at 48.
.” Testifying before the Legislative Committee on Bill C-86, John Frecker reiterated the position of
the LR.C.C. report, stating: “Mhe planners of the system seem to have a pathological fear of intro-
ducing effective review of appeal rights, because they see that as an opportunity for people to use due
process to obstruct quick disposition of claims.” Frecker predicted that the failure to provide for inter-
nal review might “have the perverse effect of causing more delay” (C-86 Proceedings, supra note 227
at 11:109, 113, 11A:22-4).
” See the Federal Court Act 1992, supra note 94 at s. 18.1(3).
… For the applicants in Singh and Narang (Richard J.), supra note 193, Richard J.’s decision refer-
ring the matter back to the Board for redetermination was the second to grant judicial review follow-
ing two negative Board decisions reached on different grounds. Reed J.’s 1993 ruling Singh and
Narang (Reed J.), supra note 147. This sort of situation may be said to attest not only to absence of
expertise but also to the inadequacy of present Board processes to catch mistakes. The claimants in
Dhillon, supra note 210, and Cheng, supra note 213, were denied judicial review on their second
“round”, despite judicial recognition of evidence indicating the claimants faced difficulties upon re-
turn to their countries of origin.
… According to one report, in 1995 the Federal Court required a budget increase of 11.6 million
dollars for refugee cases, while the cost of “C-86” judicial appointments is approximately I million
dollars annually (see “Revamp Refugee Boards”, supra note 297). It was also reported that the inter-
19961
M. HURLEY – REFUGEE DETERMINATION DECISIONS
A final point involves the dilemma of judicial inconsistency previously identified
in Trial Division caselaw. Although a two-level administrative process would not di-
rectly eliminate all factors contributing to judicial inconsistency, arguably the potential
for harm arising from it would be largely diminished as a result of systematic screening
for factual and legal errors by a specialist reviewing authority prior to judicial proceed-
ings. Furthermore, a two-stage process conducted by qualified decision-makers would
result not only in a reduced workload for Trial Division judges, but also in the eleva-
tion of the notion of comparative expertise into a relatively more reliable factor when,
as admitted non-experts, they decide judicial review applications of refugee claimants.
Conclusion
The Canadian refugee determination system has evolved as a product of three
related factors: the 1985 Singh decision of the Supreme Court of Canada mandating
oral hearings for refugee claimants; significant, rapid increases in numbers of refu-
gee claimants; and the inability or unwillingness of government to respond to either
in a timely or appropriate fashion. Successive wholesale legislative changes culmi-
nating in the present system lead one to conclude that “[i]t is easier to accept the
principle of protection and deny it in practice than to deny the principle. ‘.. The re-
sult is that genuine refugees pay the price:
The refugee crisis was certainly not caused by Singh. … Singh dictated the form
of our response. There shall be hearings. But this form could in no way con-
tradict the substance: not more but rather less generosity in granting refugees
admission to the country. The right to a hearing turned out to be no more than a
consolation prize for our stinginess …”‘
For genuine refugees, neither front-loading the system with a flawed Board nor
“plumbing” it through the transfer of supervisory jurisdiction to single judges of the
Federal Court has worked to their benefit with consistency. Trial Division judges
have, understandably, been unable to avoid varied responses to the clash of theory
and practice produced by the junction of an ill-defined judicial role, a particular
legislative evolution and the Convention’s promise of protection to genuine refu-
gees. A growing body of caselaw attests to the inadequacy of judicial review as a
first-line review mechanism under the current system.
nal review structure anticipated in March 1995 would cost 2.5 million dollars, resulting in a signifi-
cant net savings to the streamlined Board (see “Calculates”, supra note 342).
35’D. Matas, “Faimess in Refugee Determination” (1989) 18 Man. L.J. 71 at 76.
3S9 M. Mandel, The Charter of Rights and the Legalization of Politics in Canada, rev. ed. (Toronto:
Thompson Educational, 1994) at 250. See also D.J. Mullan, “Natural Justice -The Challenges of
Nicholson, Deference Theory and the Charter” in N.R. Finkelstein & B.M. Rogers, eds., Recent De-
velopments in Adrninistrative Law (Toronto: Carswell, 1987) 1 at 48-49.
McGLL LAW JOURNAL/REVUE DE DROITDE MCGILL
[Vol. 41
If anything, inherent limitations of judicial review signal the ongoing need for a
substantive internal review mechanism under qualified decision-makers to redress
present gaps in procedural and substantive protection prior to compelling refugee
claimants to seek the courts’ “blunt solution”. While repeated government rejec-
tions of this formal reconsideration model are said to arise out of a “deep-seated
fear that the system will be open to serious abuse if liberal appeal rights are ex-
tended to refugee claimants”,3″ it might also be argued, from the perspective of the
genuine refugee, that the denial of such a mechanism represents a failure to provide
the promised measure of reasonable protection.
In 1985, Rabbi Plaut commented that Canada’s “adherence to the U.N. Con-
vention and the incorporation of its principles into Canadian law are flags we have
run up on our pole of moral purpose, and there they must continue to wave”.’. In-
ternal review continues to offer, as refugee advocates have argued all along, the
best option for ensuring that contradictions between Canada’s commitment and its
practice are resolved and the promise of non-refoulement fully honoured. Until the
necessary reform is realized, the flags on Canada’s pole of moral purpose fly at half
mast.
“, J.P. Frecker, “Reform of the Refugee Detennination Process” (Paper presented to the Conference
on Canadian Immigration Law and Policy II, 10-11 May 1991) at 29 [unpublished].
3″ Plaut Report, supra note 283 at 179.
