Article Volume 51:3

Competence Concerns in Charter Adjudication: Countering the Anti-Poverty Incompetence Argument

Table of Contents

Competence Concerns in Charter

Adjudication: Countering the Anti-Poverty

Incompetence Argument

David Wiseman*

Canadian courts are reluctant to impose anti-poverty
obligations upon governments under the Canadian Charter
of Rights and Freedoms. Concerns over the limits of the
institutional competence of courts have played an explicit
role in this reluctance. The anti-poverty incompetence
argument that has thus emerged is an instance of a broader
concern over competence that is evident across the
spectrum of types of Charter cases.

This article traces the emergence of a judicial
framework for recognizing and responding to competence
concerns in early Charter adjudication and describes the
main lines of its evolution in subsequent cases. At the same
time, and for the most part remaining within the confines of
issues and arguments contained in accumulated Charter
case law, the article critically evaluates the ongoing
application of the framework in anti-poverty Charter cases.
The central argument of the article is that the case law on
competence concerns cannot justify placing relatively
greater limits on the availability or rigour of Charter
protection for anti-poverty claims than for other types of
claims. Indeed, the argument is that the case law in fact
offers encouragement to courts to pursue responses that
manage the concerns or improve competence, thereby
allowing equally fulsome protection for anti-poverty
claims.

institutionnelle

des

tribunaux

Les tribunaux canadiens sont peu enclins imposer
des obligations de lutte antipauvret aux gouvernements
selon les termes de la Charte canadienne des droits et
liberts. Les proccupations quant aux limites la
comptence
ont
particulirement motiv ces hsitations. Largument
dincomptence en matire de revendications antipauvret
qui sest manifest reflte un souci plus large quant la
comptence, souci qui est vident dans tous les types de
causes intentes en vertu de la Charte.

Cet article trace lmergence dun cadre judiciaire qui
reconnatrait et qui rpondrait aux questions de comptence
manifestes dans les toutes premires dcisions prises en
vertu de la Charte, tout en exposant les grandes lignes de
lvolution de ce cadre dans les dcisions ultrieures.
Dautant plus, et sen tenant gnralement aux questions et
aux
jurisprudence
accumule de la Charte, larticle value dun oeil critique
lapplication continue de ce cadre dans les dcisions en
matire de lutte antipauvret appuyes sur la Charte. La
proposition centrale de larticle est que la jurisprudence
traitant les questions de comptence ne peut justifier
llaboration de limites relativement plus considrables la
disponibilit ou la rigueur de la protection de la Charte en
rponse aux revendications antipauvret. En effet, lauteur
propose que
ralit un
encouragement aux tribunaux proposer des rponses qui
grent
la comptence,
permettant ainsi une protection aussi ample aux
revendications antipauvret.

raisonnements soulevs dans

les questions ou amliorent

la

jurisprudence offre en

la

* Senior Advisor, National Judicial Institute (formerly Assistant Professor, Faculty of Law,
University of Windsor). I would like to thank Aaron Dhir and Dick Moon for helpful comments on an
earlier draft of this article and Melissa Mark for research assistance. I am also grateful for the very
useful comments of two anonymous reviewers. I wish to acknowledge the financial support, for
student research assistance, of the Law Foundation of Ontario and of the Community-University
Research Alliance (CURA) program of the Social Sciences and Humanities Research Council
(SSHRC). This article is part of the ongoing research of the CURA-SSHRC Social Rights
Accountability Project (SRAP).

David Wiseman 2006
To be cited as: (2006) 51 McGill L.J. 503
Mode de rfrence : (2006) 51 R.D. McGill 503

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

[Vol. 51

504

Introduction

505

507

510
510

511
511
513
514
517

519
521
523
524

524

528
532
532
534
540
545

I. Situating Competence as an Interpretive Consideration

II. Competence Concerns in Charter Decisions

A. Starting Out: The Irrelevance of Competence
B. Recognizing Competence Challenges: Framework

and Problems
1. The Foundational Framework for Competence Concerns
2. The Three Fundamental Challenges to Competence
3. The Four Responses to Competence Challenges
4. The Foundational Framework: Reprised and Theorized
5. The Foundational Framework and Anti-Poverty

Incompetence

6. Three Problems with the Foundational Framework
C. Subsequent Developments in Competence Concerns

1. Fiscal Impact, Competence, and Anti-Poverty Cases
a. Competence in Anti-Poverty Cases: Overview and

Assessment

Problems

b. Fiscal Impact Magnitude as Competence Measure:

2. Reconstructing Deference: A Normative Counterbalance

a. The Paramountcy of Vulnerable Groups Interests
b. Vulnerable Groups Interests as Counterbalance

3. Validating Creative Engagement or Competence-Building

Conclusion

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 505

Introduction
Over the first two decades of adjudication under the Canadian Charter of Rights
and Freedoms,1 Canadian courts have been reluctant to hold that the Charter requires
governments to prevent poverty from undermining the rights and freedoms it protects.
In other words, the courts have been unwilling to interpret and apply the Charter as
imposing anti-poverty obligations upon governments. One argument offered by both
judges and scholars in justifying this unwillingness is that courts lack the institutional
competence to adjudicate anti-poverty Charter claims and, therefore, the availability
or rigour of anti-poverty protection ought to be limited. In this article I take issue with
this anti-poverty incompetence argument.

The anti-poverty incompetence argument has played an express role in judicial
hesitancy over anti-poverty Charter obligations since an early 1990s lower court
decision rejecting a challenge to disqualification from social assistance due to the so-
called spouse-in-the-house rule.2 In the Supreme Court, it is evident as recently as
the decision in Gosselin v. Quebec (A.G.),3 which rejected a challenge to Quebecs use
of reduced rates of social assistance as an incentive for participation in
employability programs. In that case, Justice Arbour wrote a strong dissent in favour
of a Charter right to adequate social assistance. Nevertheless, she accepted that the
competence challenges associated with trying to define adequacy might render the
right unenforceable. Even more recently, the incompetence argument appears to have
informed the Courts decision in Chaoulli v. Quebec (A.G.),4 which jeopardized
Quebecs prohibition on private health insurance for medically necessary services
available through the public health care system.5 That case did not directly involve an
anti-poverty adequate health care obligation, yet most members of the Court indicated
their reluctance to recognize any such obligation.6 For some of them, this reluctance
was seemingly in part due to competence concerns.7

1 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B

to the Canada Act 1982 (U.K.), 1982, c. 11 [Charter].

2 See Conrad v. Halifax (County) (1993), 124 N.S.R. (2d) 251, 42 A.C.W.S. (3d) 1103 (N.S.S.C.)
[Conrad cited to N.S.R.]. In rejecting a claim that s. 7 imposed an obligation to provide adequate
social assistance and empowered a court to determine the criteria of provision, Gruchy J. stated, I do
not have the means, the information or the ability in the facts of the case before me to make any such
determination. It is also strongly arguable that such pronouncements are not for a court; they are
policy (ibid. at 271).

3 [2002] 4 S.C.R. 429, 221 D.L.R. (4th) 257, 2002 SCC 84 [Gosselin cited to S.C.R.].
4 [2005] 1 S.C.R. 791, 254 D.L.R. (4th) 577, [2005] SCC 35 [Chaoulli cited to S.C.R.].
5 I say jeopardized because the Supreme Court only achieved a majority on the issue of whether
the prohibition was consistent with the Quebec Charter of human rights and freedoms, and the precise
effect of inconsistency, and the remedial room available to the government, was not explained in the
decision (Charter of human rights and freedoms, R.S.Q. c. C-12, s. 45 [Quebec Charter]).

6 McLachlin C.J.C. and Major J. mentioned that the text of the Charter contained no express
freestanding right to adequate health care (Chaoulli, supra note 4 at para. 104). Binnie and LeBel JJ.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

506

The anti-poverty incompetence argument forms part of a broader landscape of
Charter case law addressing the competence concerns that have arisen across a
spectrum of Charter claimsfrom challenges to mandatory retirement8 to complaints
about restrictions on tobacco-product marketing,9 from claims about minority-
language education facilities10 to challenges to rape-shield laws.11 The general judicial
recognition of competence concerns dates back almost to the inception of Charter
adjudication.12

In this article I primarily assess and criticize the anti-poverty incompetence
argument from the perspective of the broader landscape of competence-related
Charter decisions. Thus, for the purposes of this article, I largely refrain from
questioning the judicial analysis of why and when the competence of courts is
challenged. I also leave undisturbed the judicial position that anti-poverty claims can
raise competence challenges. Instead, my argument is that the accumulated Charter
case law on competence concerns cannot justify placing relatively greater limits on
the availability or rigour of Charter protection offered to anti-poverty claims than to
other types of claims. Moreover, I argue, the accumulated case law ought to prompt
courts to pursue responses that manage the challenges or improve competence,
thereby allowing equally fulsome protection for anti-poverty claims.

The argument advanced in this article proceeds in the following stages. In Part I, I
situate the issue of the competence of courts as one of the three main types of
considerations that are relied upon in Charter interpretation and enforcement by the
judiciary. This includes illustrations of the appearance of each type of consideration
in anti-poverty Charter cases. In Part II, I explain and assess the anti-poverty

stated that it would be open to the Quebec government to install a U.S.-style health care system (ibid.
at para. 176).

7 In Chaoulli, ibid., all members of the Court recognized the relevance of competence concerns,
particularly in relation to evaluating the social science evidence presented and assessing the impact of
allowing a parallel private health care system. However, the majority defended its competence to
adjudicate the health care claim before it, arguing that the Court had been presented with sufficient
social science evidence, that no special expertise was required to evaluate it and that the conclusions
that could be drawn from the evidence were clear. For its part, the minority oscillated between taking
a similar position, but endorsing the trial judges view that the evidence clearly supported the opposite
conclusion to that drawn by the majority, and taking the position that the issues were contested and
complex and that there was a need for judicial restraint in accordance with the purportedly superior
competence of legislatures in such matters.

8 McKinney v. University of Guelph, [1990] 3 S.C.R. 229, 76 D.L.R. (4th) 545, 118 N.R. 1

[McKinney cited to S.C.R.].

MacDonald cited to S.C.R.].

9 RJR-MacDonald Inc. v. Canada (A.G.), [1995] 3 S.C.R. 199, 127 D.L.R. (4th) 1, 187 N.R. 1 [RJR-

10 See Mahe v. Alberta (A.G.), [1990] 1 S.C.R. 342, 68 D.L.R. (4th) 69, 105 N.R. 321 [Mahe cited to
S.C.R.] and Doucet-Boudreau v. Nova Scotia (Minister of Education), [2003] 3 S.C.R. 3, 232 D.L.R.
(4th) 577, 2003 SCC 62 [Doucet-Boudreau cited to S.C.R.].

11 R. v. OConnor, [1995] 4 S.C.R. 411, 130 D.L.R. (4th) 235, 191 N.R. 1 [OConnor cited to

S.C.R.].

12 See Part II.B, below, for more on this topic.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 507

incompetence argument in terms of accumulated Charter decisions dealing with the
issue of competence. In so doing, I build my counter-argument. I begin with the early
position that competence concerns were irrelevant to Charter adjudication and then
discuss the cases that overturned that position and established a foundational
framework for recognizing and responding to competence concerns. This includes a
brief confirmation that the framework is consistent with academic theories on court
competence. I next explain how the foundational framework allows for the anti-
poverty incompetence argument, using the basic parameters of a generic Charter
challenge to inadequate social assistance as context. I then identify various problems
with the foundational framework and the implications of those problems for the anti-
poverty incompetence argument. Keeping the problems in view, I go on to consider
subsequent applications of, and developments in, the foundational framework. This
includes an identification and assessment of the extent to which those developments
address the problems and support or counter the anti-poverty incompetence argument.
As such, the argument of this article is largely confined to the four corners of
accumulated Charter decisions and does not purport to be sufficient to overcome the
various competence-based objections
rights
adjudication that can be found in, or extrapolated from, the academic literature. But
since the debates in the academic literature and in case judgments inform one another,
addressing the arguments as they arise in the case law at least provides a resource for
addressing the academic arguments. Further, the treatment of competence concerns in
Charter adjudication has received very little academic attention to date, yet
competence concerns are clearly playing a role in a broad spectrum of Charter cases.
Therefore, some effort to describe and assess the role of competence concerns in
Charter decision making, even if only on its own terms, is overdue.

to anti-poverty constitutional

I. Situating Competence as an Interpretive Consideration
In all types of Charter cases, anti-poverty and otherwise, considerations of

competence are but one of the three main types of considerations relied upon by
Canadian judges in interpreting and applying the Charter. The two other main types
of considerations can be categorized as textual considerations and considerations of
legitimacy.

Textual considerations include such commonly relied upon factors as the text of
the specific Charter guarantee at issue; the surrounding text and the Charters textual
structure and underlying principles; the political, social, legal, moral, and ethical
traditions, values, and theories of Canadian society that inform textual meanings; and,
of course, prior decisions and comparative and international legal doctrineto the
extent that any of these factors are not based upon considerations of competence or
legitimacy. Such textual considerations clearly play a significant role in the reluctance
of courts to recognize anti-poverty obligations. In relation to the possibility of an
obligation to provide adequate social assistance, for instance, a crucial textual issue is
the meaning to be given to the use of the word deprived in the section 7 guarantee
that no person shall be deprived of the rights to life, liberty or security of the person

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

508

except in accordance with fundamental justice. In Gosselin, Justice Arbour took up
the question of whether the use of deprived meant that the section only imposed
negative obligations to refrain from interfering with the protected rights or whether it
was capable of imposing positive obligations to more fully realize those rights. She
preferred the latter interpretation,13 but her colleague, Justice Bastarache, relied upon
textual considerations to argue for a much narrower scope of any positive
obligations.14 In the earlier lower court decision in Masse v. Ontario (Ministry of
Community and Social Services),15 which concerned a challenge to reductions in
social assistance rates, it was held that an ameliorative program, such as social
assistance, even if inadequate, cannot simultaneously constitute a deprivation.16

In separating out considerations of legitimacy I mean to delineate concerns that
Charter review empowers an appointed judiciary to override the decisions of the
relatively more democratically accountable branches of government. Legitimacy
considerations often appear as brief, but telling, references to the democratic pedigree
of legislative decision making. This is how they surfaced, for instance, in the anti-
poverty context of Clark v. Peterborough Utilities Commission,17 which concerned a
claim that section 7 provided protection against withdrawal of essential utility
services. In holding that section 7 provided no such protection, Justice Howden
reasoned that the claim raised value and policy judgments better left to legislatures
in a democratic society.18 Similarly, in Gosselin, Justice Arbour acknowledged that,
when questions of resource allocation are at stake, legislatures might be better suited
to providing answers because they enjoy the express mandate of the taxpayers.19

In contrast, considerations of competence are manifest in concerns that judges
and courts face challenges in their capacity to, first, accurately or correctly assess the
normative and empirical questions that arise in the cases before them and, second,
effectively enforce their decisions. I have already mentioned Justice Arbours
concern, in Gosselin, that a court might lack the competence to evaluate whether a
given level of social assistance is adequate.20 Her concern was prompted by the
notorious divergence of opinion on that issue among social policy analysts and
advocates and within social science literature. The same issue raised the same

13 Gosselin, supra note 3 at para. 357.
14 Ibid. at paras. 218-23.
15 (1996), 134 D.L.R. (4th) 20, 89 O.A.C. 81, 40 Admin. L.R. (2d) 87 [Masse cited to D.L.R.].
16 Ibid. at 42, ODriscoll J. See also Chaoulli where McLachlin C.J.C., who formed part of the
majority, noted that the text of the Charter contained no express free-standing right to adequate health
care (supra note 4 at para. 104). For further textual issues and arguments relevant to the anti-poverty
protection potential of s. 7, see Peter W. Hogg, Constitutional Law of Canada, looseleaf vol. 2
(Scarborough: Carswell, 1992) at 44.2 and 44.7-44.9; Martha Jackman, The Protection of Welfare
Rights Under the Charter (1988) 20 Ottawa L. Rev. 257 at 322-23.

17 (1995), 24 O.R. (3d) 7, 56 A.C.W.S. (3d) 54 [PUC cited to O.R.].
18 Ibid. at 28.
19 Supra note 3 at para. 331.
20 Ibid. at para. 330.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 509

concerns for Justice OBrien in Masse.21 Competence concerns were also expressed
in the anti-poverty context of Collins v. Canada,22 which concerned a claim that the
exclusion of separated spouses from the federal spousal pension allowance violated
section 15 of the Charter. At trial, Justice Rothstein accepted that the exclusion
violated section 15, but he rejected the claim in the section 1 stage, principally on
grounds of competence. Specifically, Justice Rothstein emphasized the difficulty he
faced in analyzing the governments choices in such a complex area of public policy,
especially given the potentially significant fiscal impact of striking down the
exclusion.23

Competence concerns are therefore not the only type of consideration relevant to
determining the scope of protection offered by the Charters rights and freedoms in
general. Nor are they the only type of consideration that has been relied upon in
justifying reluctance to recognize anti-poverty obligations. Further, both analytically
and in judicial reasoning, it can be difficult to separate the various types of
considerations. Analytically, it can be argued, for instance, that it is illegitimate for a
court to attempt adjudication for which it is incompetent. In judicial reasoning,
competence and legitimacy considerations often coincide and are rarely explained.24
Moreover, it is entirely possible that in many cases these considerations are
submerged beneath the surface of debates over textual considerations. Nevertheless,
and as will become apparent in the more detailed discussion of Charter case law to
follow, competence concerns have emerged from judicial reasoning as a relevant and
ostensibly free-standing consideration in Charter adjudication in general. Moreover,
competence concerns have clearly played a role in the reluctance of courts to impose
anti-poverty Charter obligations. To be clear, though, the reluctance has not typically
taken the form of holding that anti-poverty claims are generally injusticiable. That is,
it has not typically been held that the Charter precludes the general possibility that
the rights and freedoms it guarantees cannot be violated by social welfare policy
decisions. Rather, it has been held that particular anti-poverty claims are specifically
injusticiable. That is, it has been held that particular social welfare policy decisions,
challenged in particular anti-poverty cases, do not engage the specific scope of
protection offered by particular sections of the Charter (especially sections 7 and 15).

The anti-poverty incompetence argument has arisen from the role that
competence concerns have played in anti-poverty Charter cases. I now turn to
explaining and assessing that argument in light of the judicial consideration of
competence concerns in Charter adjudication in general.

21 Supra note 15 at 54.
22 [2000] 2 F.C. 3, 178 F.T.R. 161 [Collins cited to F.C.], affd [2002] 3 F.C. 320, 212 F.T.R. 318,

2002 FCA 82, leave to S.C.C. refused, [2002] S.C.C.A. No. 198 (QL).

23 Ibid. at para. 135.
24 The discussion of cases in Part II.C.1.a, below, illustrates this. Further illustration is provided in

David Wiseman, The Charter and Poverty: Beyond Injusticiability (2001) 51 U.T.L.J. 425.

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

510

II. Competence Concerns in Charter Decisions

[Vol. 51

A. Starting Out: The Irrelevance of Competence
In its first two years of Charter decisions, and particularly with its decisions in R.

v. Big M Drug Mart Ltd.,25 Singh v. Canada (Minister of Employment and
Immigration),26 and R. v. Oakes,27 the Supreme Court of Canada established what was
described by Professors Andrew Petter and Patrick Monahan as a large and liberal
approach to Charter adjudication.28 In doing so, the Court appeared to regard
competence concerns as irrelevant. The central elements of the large and liberal
approach were an expansive reading of the various Charter rights coupled with a
narrow reading of the permissible limitations on those rights.29 Moreover, limitations
were to be justified in largely means-end empirical-instrumental terms.
In reviewing these early cases, and the framework for Charter adjudication they

established, Petter and Monahan identified both normative and practical problems.
Normatively, they critiqued what they saw as the crude liberalism of the Courts
approach whereby the state was conceived as the principal threat to absolute
individual liberty, rather than an essential means of creating, protecting, and
balancing relative individual liberties.30 The practical problems they identified have
since become the main competence concerns of Charter adjudication. Noting that the
Court was requiring itself to embark on a detailed analysis of the actual and potential
instrumental effects of the limitation at issue, and its alternatives, as well as to review
the balances struck between affected interests, they doubted whether the Court had
the institutional competence to do so successfully. More specifically, they observed
that in these early cases the Court often did not have sufficient empirical evidence
available to it and, more generally, they argued that the judiciary lacked familiarity
with and expertise in empirical matters. They suggested that the courts would have
difficulty in identifying the instrumental effects of alternative governmental measures,
as well as in identifying the effects of their own interventions.31

For its part, however, the Court dismissed such concerns over its legitimacy and
competence as were raised in the early cases. For instance, in Operation Dismantle
Inc. v. Canada,32 a claim was brought under section 7 of the Charter challenging the
Canadian governments decision to allow the testing of cruise missiles over its
territory. The respondents argued, in part, that the challenge was generally

25 [1985] 1 S.C.R. 295, 18 D.L.R. (4th) 321, 58 N.R. 81.
26 [1985] 1 S.C.R. 177, 17 D.L.R. (4th) 422, 58 N.R. 1 [Singh cited to S.C.R.].
27 [1986] 1 S.C.R. 103, 26 D.L.R. (4th) 200, 65 N.R. 87 [Oakes cited to S.C.R.].
28 Andrew J. Petter & Patrick J. Monahan, Developments in Constitutional Law: The 1986-87

Term (1988) 10 Sup. Ct. L. Rev. 61 at 63.

29 Ibid.
30 Ibid. at 66-70.
31 Ibid. at 84-89.
32 [1985] 1 S.C.R. 441, 18 D.L.R. (4th) 481, 59 N.R. 1 [Operation Dismantle cited to S.C.R.].

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 511

injusticiable because the courts lacked both the legitimacy and competence to review
foreign policy decisions. With respect to competence concerns, Justice Wilson
counter-argued that attention needed to be focused on whether the courts should or
must rather than on whether they can deal with [the issues before them].33 With these
words, Justice Wilson thus appeared to be giving little credence to competence
concerns.34 However, it was not long before the competence concerns voiced by
Petter and Monahan, and some litigants, became apparent to the Court.

B. Recognizing Competence Challenges: Framework and Problems
Over the course of the next four terms of Charter decisions, competence concerns
came to occupy a prominent role in Charter adjudication. By the end of this period,
competence concerns had been recognized and responded to in all stages of Charter
adjudication. In the process, the Supreme Court of Canada established a foundational
framework for the recognition and treatment of competence concerns that remains in
use to this day. In this section, I describe the foundational framework, identify its
problems, identify its detrimental implications for anti-poverty claims, and situate it
in relation to academic perspectives on court competence.

1. The Foundational Framework for Competence Concerns

The most significant decisions, in terms of establishing the foundational
framework, were R. v. Edwards Books and Art Ltd.,35 Reference Re Public Service
Employee Relations Act, Labour Relations Act and Police Officers Collective
Bargaining Act,36 and its companion cases (collectively known as the Labour
Trilogy),37 Irwin Toy v. Quebec (A.G.)38 and Mahe.39 In Edwards Books the Supreme
Court was faced with claims that restrictions on Sunday trading violated the Charter
guarantees of freedom of religion, liberty, and equality.40 The Court, by majority, held
that freedom of religion had been violated but that the violation could be saved under
section 1. In reaching this conclusion, the majority watered down the section 1
standard of justification, in part due to competence concerns. In the Labour Trilogy,

33 Ibid. at 467 [emphasis in original].
34 At the same time, though, all judges but Wilson J. agreed that some of the specific issues raised by
the claimantssuch as whether allowing the testing would increase the risk of nuclear warwere
incapable of proof and thus, in that more specific sense, injusticiable.

35 [1986] 2 S.C.R. 713, 35 D.L.R. (4th) 1, 71 N.R. 161 [Edwards Books cited to S.C.R.].
36 [1987] 1 S.C.R. 313, 38 D.L.R. (4th) 161, 74 N.R. 99 [Alberta Labour Reference cited to S.C.R.].
37 Public Service Alliance of Canada v. Canada, [1987] 1 S.C.R. 424, 38 D.L.R. (4th) 249, 75 N.R.
161 and Saskatchewan v. Retail, Wholesale & Department Store Union, [1987] 1 S.C.R. 460, 38
D.L.R. (4th) 277, 74 N.R. 321.

38 [1989] 1 S.C.R. 927, 58 D.L.R. (4th) 577, 94 N.R. 167 [Irwin Toy cited to S.C.R.].
39 Supra note 10.
40 It was also claimed that the Act was ultra vires the provinces legislative powers, on division of

powers grounds. This claim was unanimously rejected.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

512

competence concerns informed the majoritys holding that the Charter did not protect
rights to strike or collectively bargain and, therefore, those rights were injusticiable in
the violations review stage. In Irwin Toy the section 1 standard of justification was
again relaxed, in part due to competence concerns, enabling the government to
withstand a challenge, based on freedom of expression and the right to liberty, to
restrictions on advertising aimed at children. Finally, competence concerns were
recognized as relevant in the remedy-review stage in Mahe and prompted the
Supreme Court to exercise restraint in its directions to the government in relation to
its obligations with respect to minority language education.

The key prompt for the recognition of competence concerns in these cases was
the Courts realization that the crude liberal paradigm was inappropriate. In keeping
with the arguments made by Petter and Monahan, the Court came to recognize that
Charter adjudication often required a review of governmental action that involved the
balancing of rights and freedoms and of individual and group interests. Moreover, this
balancing implicated a variety of normative and empirical factors that had to be
reviewed in the absence of normative and empirical certainty.41 Consequently, it
became apparent that the crude liberal paradigm of expansive definitions of rights and
freedoms and rigourous means-ends standards for the justification of limitations was
inappropriate. The paradigm was premised on a different view of the nature of rights-
limiting governmental actiongovernment action as interference with, rather than
balancing of, rights and freedoms. Also, it presumed a degree of normative and
empirical certainty that might often be unavailable.

The approach of the crude liberal paradigm therefore had to be abandoned. The
Court was compelled to enter into a more deliberate, though by no means exhaustive,
consideration of what it could legitimately and competently do in terms of normative
and empirical analysis. The starting point for that consideration was the majority
judgment of Chief Justice Dickson in Irwin Toy in which he identified four factors
affecting the degree of legitimacy and competence of Charter adjudication, namely
whether the impugned governmental action could be categorized as individual-
antagonizing criminal justice or as group-mediating social policy; whether there was

41 Thus, in Edwards Books, La Forest J. noted that attempts to protect the rights of one group will
also inevitably impose burdens on the rights of other groups (supra note 35 at 795). This, in turn,
meant that legislatures had to be given reasonable room to manoeuvre to meet these conflicting
pressures (ibid.). Moreover, La Forest J. argued, it had to be recognized that the choices which
governments needed to make from among the alternative means available for achieving governmental
objectivesalternatives that represented different ways of reconciling the conflicting pressures
relied upon an in-depth knowledge of all the circumstances and, as such, were choices that a court
is not in a position to make (ibid. at 796). Similarly, in the Labour Trilogy, McIntyre J. noted that in
all three cases the section 1 inquiry involves the reconsideration by a court of the balance struck by
the Legislature in the development of labour policy but that such questions as which government
services are essential and whether the alternative of arbitration is adequate compensation for the loss
of a right to strike were not amenable to principled resolution (Alberta Labour Reference, supra
note 36 at 419). Indeed, McIntyre J. continued: There are no clearly correct answers to these
questions. They are of a nature peculiarly apposite to the functions of the Legislature (ibid.).

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 513

conflicting social science evidence; whether there were competing claims to scarce
fiscal resources; and whether the interests of vulnerable groups were at stake.
The majority judgment of Chief Justice Dickson in Irwin Toy offers little

explanation of how these factors raised legitimacy and competence concerns but,
when read together with the judgments in the other significant cases of this period, an
explanation emerges. Beginning with the issue of competence, the reasoning of these
cases indicates that the Court understood the structure of Charter adjudication
established in the earlier cases as raising a variety of normative and empirical issues
that had to be resolved in a principled fashion. In the section 1 review stage, for
instance, the requirement of rational connection raised the empirical issue of whether
the impugned limitation would actually further the given objective and might require
the Court to consider social science evidence on causes and effects. Courts
conducting Charter adjudication thus needed to be competent to undertake the
required normative and empirical analysis and, ultimately, needed to be competent to
authoritatively decide which arguments and evidence were correct or right. Over
the course of these decisions, however, the Court came to recognize that the
normative and empirical arguments and evidence were complex, contested, and
subject to uncertainty. In turn, it became concerned about whether it had the
institutional competence to determine which arguments and evidence were correct
or right and about how to respond when its competence was lacking. Ultimately,
the Court became concerned about three fundamental challenges to its competence
and developed four main responses, which I describe in more depth in the following
sections.

2. The Three Fundamental Challenges to Competence

The first fundamental challenge identified by the Court was that of gathering an
adequate quantity and quality of information relevant to normative and empirical
issues. In relation to normative issues, the Court recognized a need to consider not
only the normative perspectives of the immediate parties, but also the perspectives of
other individuals and groups potentially affected by its decisions. Similarly, the Court
recognized a need to ensure that its analysis of empirical issues was sufficiently
informed by an adequate range of empirical evidence. This information requirement
was especially necessary in respect of the empirically oriented section 1 issues of
rational connection and minimal impairment.42

The second fundamental challenge recognized by the Court was the difficulty in
competently evaluating the information presented to it, especially the empirical
information. This challenge is evident in Irwin Toy, where conflicting social science

42 In numerous early cases, the Court was willing to hold against governments that had failed to
understand their evidentiary burden in the s. 1 review stage, but the objective of the Court in doing so
was ultimately to ensure that it was privy to at least the empirical evidence that informed the
impugned governmental decisions.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

514

evidence was identified as a factor that jeopardizes the competence of courts. This
challenge may also be evident in that cases assertion that the expertise of judges in
criminal justice matters means the courts are relatively more competent in those
casesthe argument perhaps being that judges have sufficient expertise to resolve
conflicting social science on such matters.

The third fundamental challenge recognized by the Court was the difficulty in
reviewing the balances that governments had struck among the various potential
normative and empirical effects of the alternative courses of action open to them.
Such balancing inevitably involved trade-offs among the various interests affected
and the Court was concerned either that principled standards for reviewing those
trade-offs were not available or, if they were, that it lacked the expertise to apply
those standards. This challenge can be regarded as underlying the identification, in
Irwin Toy, of the presence of competing claims to scarce fiscal resources as a factor
undermining competence. Resolving such competing claims may require the
competent evaluation of empirical information on potential fiscal impacts. It may also
require the application of some standard for assessing the appropriateness of the
balances struck among competing fiscal demands. Likewise, when the majority in
Irwin Toy asserted that whether the interests of a vulnerable group were at stake was
an issue relevant to competence, they may have been concerned that courts lack the
normative expertise to properly weigh the interests of vulnerable groups. This is to
say that the majority may have been concerned that courts have tended to overlook
the interests of such groups. Finally, the decision in Mahe includes a recognition of
the challenges that courts face in reviewing the balances struck by governments in
designing and implementing social programs. Specifically, in opting for a declaratory
remedy, the Court in Mahe expressly chose to leave it to the government, at least in
the first instance, to identify and balance the relevant interests and considerations.43
Having identified these fundamental challenges to court competence, the question
then became how to respond to them.

3. The Four Responses to Competence Challenges

In the foundational cases on competence, the Court employed four responses to
the competence challenges that arose. First, in the Labour Trilogy, the means chosen
for responding to the competence concerns recognized was to hold the claim raising
those concerns injusticiable (in the specific sense). Injusticiability is a response that
operates in the violations review stage and has the consequence of completely
excluding not only the immediate claim but all other claims of the same type from the

43 This challenge also seems evident in the Labour Trilogy insofar as McIntyre J. emphasized that
the settlement of labour disputes typically, and appropriately, depended upon the balance of economic,
social, and political power between the parties and a willingness to compromise for a defined period
of years, the suggestion being that the resolution of labour disputes is a matter of pragmatics rather
than principles. See Alberta Labour Reference, supra note 36.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 515

realm of Charter protection. In this sense, injusticiability is a relatively severe
response.44 Interestingly, the majority in the Labour Trilogy opted for the
injusticiability response even though the competence concerns identified related
mostly to the requirements of the section 1 and remedy review stages of Charter
adjudication. The only explanation offered for choosing this response was Justice
McIntyres suggestion that withdrawal is the appropriate response to competence
concerns that arise from rights claims where there is some interpretive doubt about
whether the rights are protected by the Charter.
In Edwards Books and Irwin Toy the competence concerns arose in the section 1

review stage and were responded to in that stage by a second means, namely,
deference or, in other words, the reasonable basis standard of review. By virtue of
that standard, a government does not need to prove that a limiting measure is, for
instance, rationally connected or minimally impairing. Rather, it merely needs to
establish that it had a reasonable basis for believing that it was.45 As this response at
least allowed the rights upon which the claims were based to be included within the
scope of Charter protection, it is a relatively less severe response than injusticiability.
Further, since in these cases the legislation under challenge could be regarded as
protective of vulnerable groups (i.e., employees in small retail businesses and
children), opting for this response meant that those vulnerable groups received the
benefit of the competence concerns raised. But it is also worth noting that it was by
no means interpretively certain that corporations were entitled to lay claim to the
Charters guarantees of freedom of religion and expression. Thus, the choice of
deference, rather than injusticiability, appears potentially inconsistent with Justice
McIntyres position in the Labour Trilogy.46
Two further responses were evident in Mahe. In that case, the competence

concerns arose in both the violations review stage and the remedy review stage. At
each stage the concerns revolved around the issue of defining the precise
requirements of the section 23 guarantee. In the violations review stage this issue
arose because those requirements would define the scope of the guarantee and,
therefore, the circumstances amounting to a violation of the guarantee. In the remedy
review stage this issue arose because those requirements would define the goal for
remedial action. But the Court was concerned that it lacked the competence to define
the requirements of section 23 in detail, at least at such an early stage of section 23
adjudication. As has already been mentioned, the competence concerns arising in the

44 Alternatively, or as well, it might have been argued that the accumulation of competence
concerns, across the stages of Charter adjudication, was so great that it was best for the Court to
completely withdraw from adjudication of the claim.

45 This standard in fact originates from pre-Charter federalism decisions, particularly the Reference

Re Anti-Inflation Act, [1976] 2 S.C.R. 373 at 423, 68 D.L.R. (3d) 452, 9 N.R. 541, Laskin C.J.C.

46 The apparent inconsistency might still be reconciled on the basis that, overall, the competence
concerns arising in Irwin Toy and Edwards Books were not as great as those in the Labour Trilogy,
although no such argument was made in the judgments. See Alberta Labour Reference, supra note 36
at 316-17.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

516

remedy review stage were responded to in that stage by means of remedial restraint.
As compared to the responses of injusticiability and deference, this response is the
relatively least severe on the claimants as it not only permits inclusion of the claim
within the scope of Charter protection, but also allows for rigourous scrutiny in the
section 1 review stage. Nevertheless, the response of remedial restraint still involves a
degree of judicial withdrawal in the face of competence concerns.
In contrast, the response of the Mahe Court to the competence concerns arising in

the violations review stage indicates that there is a further response of seeking to
overcome these concerns by modifying adjudicatory approaches. Specifically, rather
than laying out a detailed set of requirements defining the precise obligations flowing
from section 23, the Court confined itself to describing the requirements in general
terms only. In doing so, the Court expressly left it to governments to undertake, in the
first instance, the task of deciding what particular policies and institutional
arrangements would fulfill the guarantee. While the Court recognized that this might
lead to further litigation to determine whether the guarantee had in fact been complied
with, it preferred that to the risk of imposing impractical solutions.47 Summing up
the philosophy behind this approach, the Court said that [s]ection 23 is a new type of
legal right in Canada and thus requires new responses from the courts.48 This
statement is significant for the suggestion that when legal rights evolve in ways that
pose challenges to traditional judicial approaches and to competence, the appropriate
response is not to withdraw from adjudicating the rights but to develop new
approaches that enable the rights to be adjudicated. This response, which I will call
competence-building, stands in stark contrast to the response of injusticiability
chosen in the Labour Trilogy, which also operated in the violations review stage. The
willingness of the Court in Mahe to consider competence-building appears to be
based on the clarity of the positive obligation entrenched in section 23that clarity
meant that the Court did not have the same room to capitulate to the competence
concerns, as it had in the Labour Trilogy (where the rights claimed were open to
interpretive doubt). Instead, it had no choice but to develop an approach for
competently adjudicating the right. If so, then it still remained possible for the Court
to choose to read down other rights, where there are greater interpretive doubts, on
competence grounds, and thus avoid having to apply new approaches in the
adjudication of those other rights. However, given the position taken in Mahe, any
such avoidance strategy would not be able to rely upon the argument that the Court
cannot develop its competence and modify its approach. The most that could be
argued would be that it should not. In addition, though, it should be noted that the
willingness of the Court to develop its competence in Mahe may also have been
related to a perception that there were special reasons not to withdraw from
adjudication of the section 23 guarantee. Specifically, section 23 might have special
status owing to the unique bilingual history of Canada and the possible centrality of

47 Mahe, supra note 10 at 376.
48 Ibid.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 517

the minority language provisions to the entire Charter enterprise.49 If so, this suggests
that competence concerns may need to be counterbalanced by normative concerns.50

In the foundational cases, then, the Court employed four responses to the
competence challenges it identified. In combination, the challenges and responses
constituted a foundational framework for addressing competence concerns. It should
be noted, however, that these were also the responses relied upon when legitimacy
challenges arose.51 Indeed, in numerous cases the use of one or another response is
justified by both competence and legitimacy concerns (and textualist concerns can
also be relevant). This complicates the task of assessing the justifiability of any
particular instance of reliance upon one of the responses. An assessment that one or
another response used in a particular case is not justified on grounds of incompetence
does not necessarily render the response unjustifiable when other grounds are taken
into account, especially since courts seldom adequately explain the weight of each
ground. Nevertheless, it remains possible to assess the extent to which competence
challenges justify any particular instance of reliance upon a response, which is the
most I purport to do in this article.
At this point it is convenient to reprise the basic elements of the foundational
framework and to briefly situate the framework in relation to academic theories of
court competence. Once that is done I will explain how it allows for the anti-poverty
incompetence argument.

4. The Foundational Framework: Reprised and Theorized

The foundational framework for the recognition and treatment of competence
concerns consisted, first, in an identification of three fundamental challenges to

49 Michael Mandel argues that Prime Minister Trudeaus primary motivation and objective in
pursuing the constitutional entrenchment of a Charter was to constitutionally entrench bilingualism.
See Michael Mandel, The Charter of Rights & the Legalization of Politics in Canada (Toronto:
Thomson, 1994).

50 The fourth response of competence-building is also evident in other ways in the cases under
discussion. For instance, the efforts of the Court, in these cases and others, to improve its access to
information by generally inviting and encouraging the presentation of a wide range of information and
arguments, both normative and empirical, can be understood as competence-building efforts. More
specifically, and taking its lead from constitutional division of powers decisions, the Court applied less
stringent admissibility standards to so-called legislative fact evidence that often includes social science
studies. And although at one point members of the SCC complained that the standards had become too
loose and tightened them somewhat (see Danson v. Ontario (A.G.), [1990] 2 S.C.R. 1086, 73 D.L.R.
(4th) 686, 112 N.R. 362), they have remained less stringent. In addition, the Court applied the rules on
interventions in a more welcoming fashion, although this too has ebbed and flowed. See Jillian Welch,
No Room at the Top: Interest Group Intervenors and Charter Litigation in the Supreme Court of
Canada (1985) 43 U.T. Fac. L. Rev. 204; Sharon Lavine, Advocating Values: Public Interest
Intervention in Charter Litigation (1992) 2 N.J.C.L. 27.

51 Obviously enough, in the context of legitimacy concerns, the fourth response would be labelled

legitimacy building.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

518

competence, namely, the challenge of gathering an adequate quantity and quality of
evidence and arguments, the challenge of evaluating the evidence and arguments, and
the challenge of reviewing the balancing and trading-off of relevant interests and
considerations. As part of this, it was argued that one or more of the following
situations would typically bring these challenges into play: the impugned
governmental action could be categorized as group-mediating social policy (rather
than as individual-antagonizing criminal justice); there was conflicting social science
evidence; there were competing claims to scarce fiscal resources; and the interests of
vulnerable groups were at stake. It was also recognized that these situations, or the
challenges, or both, could arise in any stage of Charter adjudication. Finally, the
foundational framework consisted in an identification of injusticiability, deference,
remedial restraint, and competence building as four types of responses to the
competence challenges.
As such, the foundational framework is consistent with the main line of argument
in academic debates over the issue of the institutional competence of courts. Briefly
put, in academic analyses it has been argued that the competence of courts can be
conditioned by four factors: first, the limiting effects of the forms of adjudication,
including, at its broadest, the rules of evidence and procedures of litigation, the
independence and expertise of judges, and the nature of legal rights and remedies;52
second, the ideological tilt of judicial values;53 third, the absolutely and relatively
small scale of the court system, as compared to other mechanisms for social decision
making;54 fourth, the dynamics of participation, which determine what issues will be
brought to court and by whom.55

By far the most prevalent line of argument in scholarly debates on competence
concerns, including in Charter scholarship, is the argument relating to the limiting
effects of the forms of adjudication. This argument has its foundations in the work of
Lon Fuller and was elaborated in some detail in a study of U.S. adjudication by
Donald Horowitz.56 A key plank of this argument is that the forms of adjudication

52 This argument was pioneered in Lon L. Fuller, The Forms and Limits of Adjudication (1978)
92 Harv. L. Rev. 353. See also Donald L. Horowitz, The Courts and Social Policy (Washington:
Brookings Institution, 1977).

53 A Canadian example of this line of argument is provided in Joel Bakan, Just Words:

Constitutional Rights and Social Wrongs (Toronto: University of Toronto Press, 1997).

54 This line of argument is an aspect of the comparative institutional analysis developed in Neil K.
Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics, and Policy (Chicago:
University of Chicago Press, 1994) [Komesar, Imperfect Alternatives].

55 This is the basic argument of Komesars comparative institutional analysis. See Komesar,
Imperfect Alternatives, ibid. and Neil Komesar, Laws Limits: The Rule of Law, and the Supply and
Demand of Rights (Cambridge, UK: Cambridge University Press, 2001).

56 See Fuller, supra note 52 and Horowitz, supra note 52. The leading Canadian treatments of this
argument are Christopher P. Manfredi, Judicial Power and the Charter: Canada and the Paradox of
Liberal Constitutionalism (Toronto: McClelland & Stewart, 1993), c. 6, 7; W. A. Bogart, Courts and
Country: The Limits of Litigation and the Social and Political Life of Canada (Toronto: Oxford
University Press, 1994); and Jackman, supra note 16. See also Craig Scott & Patrick Macklem,

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 519

limit the capacity of courts both to gather a sufficient quantity and quality of
normative and empirical arguments and evidence, and to accurately evaluate what is
gathered. In turn, by this argument, courts have a limited capacity to evaluate the
balances that governmental decision makers have struck in light of such arguments
and evidence. Clearly, this argument parallels the way that Canadian courts have
defined the challenges to their institutional competence. Further, the academic
treatments of this argument include consideration of a variety of potential responses
to the limits of the forms of adjudication, ranging from complete judicial withdrawal,
to cautious case-by-case intervention, and competence-building measures. Thus, the
entire foundational framework on court competence established by the Supreme
Court of Canada is broadly consistent with the main line of argument in academic
debates over that issue.57

5. The Foundational Framework and Anti-Poverty Incompetence

The foundational framework on court competence provided ample room for the
emergence of the anti-poverty incompetence argument. The instances, mentioned in
Part I, in which courts have raised competence concerns in limiting their scrutiny of
anti-poverty Charter claims illustrate this. Yet in each of those instances, and in the
further instances to be discussed in following sections,58 the explanation of the
competence concerns has been inadequate. The anti-poverty incompetence argument
thus remains only partially articulated in the case law. Consequently, to explain how
the foundational framework allowed for an anti-poverty incompetence argument, I
will use the example of a simplified generic anti-poverty challenge to inadequate
social assistance. This example is based on the facts and arguments in Masse and
shares similarities with the claim made in Gosselin.

Consider, then, a claim by a social assistance recipient for whom the present or
proposed level of assistance is too low to ensure security of the person, as guaranteed
by section 7. The claimant thus argues that her rights under section 7 have been
violated,59 that the violation cannot be saved under section 1, and that the appropriate

Constitutional Ropes of Sand or Justiciable Guarantees? Social Rights in a New South African
Constitution (1992) 141 U. Pa. L. Rev. 1; Paul Weiler, Two Models of Judicial Decision-Making
(1968) 46 Can. Bar Rev. 406; Barry L. Strayer, The Canadian Constitution and the Courts: The
Function and Scope of Judicial Review, 3rd ed. (Toronto: Butterworths, 1988); and Ralph Cavanagh
& Austin Sarat, Thinking about Courts: Toward and Beyond a Jurisprudence of Judicial
Competence (1980) 14 Law & Socy Rev. 371.

57 This is not to say that the academic argument on the limits of the forms of adjudication in fact
justifies the Courts foundational framework. Indeed, in my view, not only does the Courts
framework fail to take account of important nuances in the academic argument, but the academic
argument itself has a number of shortcomings. I do not advance that view here, but for preliminary
steps in that direction, see D. Wiseman, Taking Competence Seriously in S. Boyd et al., eds.,
Poverty: Rights, Social Citizenship and Governance (Vancouver: UBC Press, forthcoming 2006).

58 See Part II.C.1, below.
59 By the terms of s. 7, the claimant would need to show not only that she was deprived of security
of her person but also that the governments failure to address this insecurity did not accord with the

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

520

remedy is to order a higher level of assistance. Such a claim raises three main issues:
first, the issue of what constitutes security of the person, particularly in relation to
economically disadvantaged groups, and in what circumstances poverty can be
regarded as a violation of the right to security of the person; second, the issue of on
what basis a government can justify failing to provide an adequate level of social
assistance; third, the issue of what a government should be required to do to remedy
any unjustifiable failure.

In putting these issues before the court, this case could involve all of the factors
regarded as giving rise to competence challenges. At the general level of the
fundamental challenges to competence, a court would be required to gather and
evaluate a broad range of evidence and arguments on a variety of normative and
empirical issues, such as whether poverty ought to be recognized as threatening
security of the person and, if so, what constitutes poverty. A court would also be
required to review the governmental balancing of competing claims upon the scarce
fiscal resources available for government programs in general and anti-poverty
programs more particularly. At the level of the more specific situations bringing these
challenges into play, social assistance and anti-poverty policy is more in the realm of
group-mediating social policy than individual-antagonizing criminal justice. Further,
as indicated in Masse and Gosselin, there are conflicts of expert opinion within the
social policy and social science communities as to the appropriate concepts and
measures of poverty. There are also deep disagreements within these communities as
to whether, and to what extent, scarce fiscal resources ought to be devoted to
generating general economic opportunities (through such measures as tax cuts and
deficit reduction or wage subsidies and public works), rather than to direct poverty
alleviation. Clearly, the allocation of fiscal resources will be affected and the
competing interests of economically advantaged and disadvantaged groups in such
allocations will be present. Finally, by definition, the interests of the vulnerable group
of people living in poverty, and reliant upon social assistance, will be at stake.
According to the foundational framework then, the claim to adequate social
assistance raises the full range of competence concerns. Consequently, it allows an
argument that the courts will suffer incompetence in adjudicating the claim. Further, it
allows an argument that it would be best to render the claim specifically injusticiable
(i.e., to hold that poverty does not threaten security of the person) or to defer to
governmental justifications in the section 1 review stage. In other words, it allows the
anti-poverty incompetence argument.
The question then becomes whether, on closer inspection, accumulated Charter

case law on competence concerns ultimately justifies this anti-poverty incompetence
argument. In building my argument that it is not justified, I first identify some
problems with the foundational framework itself and explain their implications for the

principles of fundamental justice. Charter, supra note 1, s. 7. This latter step might be achieved by
arguing that it is arbitrary and unfair to establish a program to meet basic needs and then to refuse to
provide a level of assistance sufficient to do so.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 521

anti-poverty incompetence argument. I then move on to an assessment of subsequent
developments in competence case law.

6. Three Problems with the Foundational Framework

The foundational framework for recognizing and responding to competence
concerns was beset with three problems. The first was the difficulty in discerning the
reasons why one response to competence concerns was preferred to another in the
different cases. In this regard, I have already raised the potential difficulty in
reconciling the choice of injusticiability as the response in the Labour Trilogy case
with the choice of deference in Edwards Books and Irwin Toy. But of greater import is
the potential difficulty in explaining why efforts were not made in any of these cases
to explore the response of building competence that was utilized in Mahe. In relation
to the anti-poverty incompetence argument, this problem makes it harder to justify
responding to the competence concerns raised by anti-poverty claims with
injusticiability or deference, rather than remedial restraint or competence-building.

The second problem with the foundational framework was the reference point,
identified in Irwin Toy, for distinguishing between significant and insignificant
competence concernsthat is, the distinction between governmental decisions that
are individual-antagonizing criminal justice decisions and those that are group-
mediating social-policy decisions. The validity of this reference point seems to
depend upon the position that judges have greater familiarity with criminal justice
matters and thus greater expertise and competence in such matters. Also implied is the
position that criminal justice matters, since they are individual-antagonizing, do not
require a review of governmental balancing. The problems with these positions are
threefold. First, legislation that appears to fall into the category of social policy may
nevertheless utilize criminal sanctions and so the distinction might be difficult to
draw in many cases. An example, which subsequently came before the Court, is
legislation imposing criminal sanctions on corporate office-holders whose companies
violate restrictions on tobacco product advertising.60 Second, it is in any event
difficult to see how balancing is not relevant to criminal justice matters. In many such
matters, consideration may need to be given to a range of interests and considerations,
including the publics interest in order, liberty, and privacy; the interests of police in
the efficiency and effectiveness of procedural powers; the interests of the accused in
fundamental justice; the interests of victims in empowerment and justice; and
considerations of the costs of judicial proceedings and of the law enforcement process
in general. Third, merely because many judges end up hearing a large volume of
criminal cases does not mean that they have had any pre-bench experience in those
matters or that judicial experience will necessarily yield real normative expertise.
There is, after all, no end of criticism of judicial policy making in criminal justice
matters. Even if it were conceded that judges will become expert in evaluating the

60 RJR-MacDonald, supra note 9.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

522

normative considerations, it is difficult to see why the general challenge of gathering
and evaluating social science evidence, which is often highly relevant to criminal
justice matters, would not be as significant as in group-mediating social-policy
cases.61 Moreover, since the operation of the criminal justice and law enforcement
systems depends upon a significant devotion of fiscal resources, the competence
concerns raised by the presence of that factor would seem as applicable to criminal
justice matters as to social policy matters. With respect to the anti-poverty
incompetence argument, if these problems were recognized, and the distinction
abandoned, then the simplest avenue for categorizing anti-poverty claims as
competence challenging would be lost.62
The third problem with the foundational framework lies in the fact that in Irwin

Toy the Court had identified the factor of the interests of vulnerable groups as
affecting competence, and therefore as a factor militating in favour of deference, but
had failed to address the possibility that, in different circumstances, deference might
work against those interests. In both Edwards Books and Irwin Toy the impugned
legislation was regarded as protective of vulnerable groups and the Court relied upon
the relevance of the stake of a vulnerable group in being deferential at the section 1
review stage. Deference on this basis can be understood to be grounded in
competence concerns to the extent that the courts recognize that they may not have
the normative competence to give appropriate weight to the interests of vulnerable
groups. This can then be reinforced by the presence of the other factors affecting
competence, as well as by the broader normative position, put forward in Edwards
Books, that the Charter not be used as a vehicle for undermining legislation aimed at
assisting vulnerable groups. But what if the impugned legislation were being
challenged by a vulnerable group, and yet all the other factors affecting competence
were also present? The other factors would raise competence concerns and invite
deference, and yet to be deferential in these circumstances would work to the
disadvantage of the vulnerable group. Whether such a result is appropriate depends
upon the more precise reasons behind competence-based deference when the interests
of vulnerable groups are at stake. If the reason for deference is that the legislature is
better suited to assess the weight of the interests of vulnerable groups than are the
courts, then it is appropriate to defer to the government whenever those interests are
at stake, regardless of whether vulnerable groups view the governmental decision

61 Here it is worth noting that it was in relation to the decision in Oakes, supra note 27, which
addressed a reverse onus of proof for narcotics offences, that Petter and Monahan first raised their
concerns over competence. See Petter & Monahan, supra note 28.

62 With respect to the issues of gathering and evaluating social science evidence, the problem may
run much deeper than a simple failure to recognize the fuzziness of the distinction between individual-
antagonizing and group-mediating governmental decision making. In an illuminating study, Danielle
Pinnard argues that the SCCs treatment of concerns with gathering and evaluating factual evidence in
Charter cases lacks any coherent methodology. She also suggests that the apparently deliberate
mystery of the approach to factual issues is the most troubling form of judicial activism. See Danielle
Pinnard, Institutional Boundaries and Judicial ReviewSome Thoughts on How the Court is Going
About its Business: Desperately Seeking Coherence (2004) 25 Sup. Ct. L. Rev. (2d Series) 213.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 523

being impugned as protective or antagonistic. But if the reason for deference is to
give vulnerable groups, rather than the legislature, the benefit of the competence
challenges faced by the court, then deference is only appropriate when the impugned
governmental decision is protective.
At this point, though, it might be better to think of this third problem as the issue
of whether and to what extent competence concerns can be counterbalanced by
normative concerns (as may have occurred in Mahe). While considerations of
competence may invite deference to governments in the section 1 review stage, other
considerations of a more normative nature may pull against deference. Thus, where
an impugned governmental decision can be regarded as detrimental to the interests of
vulnerable groups, the normative consideration of ensuring that the Charter is a
vehicle for assisting vulnerable groups may counterbalance competence-based
arguments for deference. Indeed, as the decision in Mahe suggests, when competence
concerns and normative concerns pull in different directions, there may be a stronger
argument for exploring competence-building responses. In regard to the anti-poverty
incompetence argument, since anti-poverty claims involve the vulnerable group of the
poor, any entrenchment of that factor as either a ground for counterbalancing
competence concerns, or for justifying responses aimed at overcoming such concerns,
would weaken that argument.
As such, the various problems in the foundational framework demanded
attention. This was especially so from the perspective of anti-poverty Charter claims
because, depending upon how the more specific aspects of the general framework
were developed and applied, there could be serious detrimental implications for the
scope and rigour of anti-poverty Charter protection. In considering, then, the
recognition and treatment of competence concerns following the establishment of the
foundational framework, attention needs to be given not only to general
developments but also to the extent to which the potential problems with that
framework have been evident, to what extent they have been addressed, and to the
ongoing implications for the anti-poverty incompetence argument.

C. Subsequent Developments in Competence Concerns

There have been four general developments in the treatment of competence
concerns since the establishment of the foundational framework: (1) the abandonment
of the distinction between criminal justice and social policy matters as a reference
point for distinguishing matters where competence concerns are significant; (2) the
emergence of the factor of the magnitude of potential fiscal impact as a measure of
competence concerns and as a determinant of responses to those concerns; (3) the
reconstruction of the group of factors recognized as relevant to decisions as to
deference and, as part of that, the effort to ensure a normative component to such
decisions; (4) the efforts undertaken by some courts to modify the traditional forms of
adjudication in order to overcome competence concerns.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

524

The first of these developments occurred for reasons in keeping with those
outlined in the previous section as to the problems with attempting to use such a
distinction and does not merit further attention in what follows.63 Since the
maintenance of this distinction provides a plank of support for the anti-poverty
incompetence argument, its abandonment removes that plank.64

The other general developments require more detailed attention. The primary
context for the second general development has been anti-poverty cases. I will discuss
that development and its implications next, while at the same time outlining and
assessing the role of competence concerns in anti-poverty cases in general. I will then
move on to a discussion of the third and fourth developments and their implications.

1. Fiscal Impact, Competence, and Anti-Poverty Cases

a. Competence in Anti-Poverty Cases: Overview and Assessment
Over the course of the first two decades of Charter adjudication, Canadian courts
have been faced with numerous claims, usually launched under section 7 or section
15, or both, that have either directly asserted that the Charter imposes anti-poverty
obligations or have asserted obligations that are poverty related in the sense that they
address the accessibility of social services or the availability of social benefits. Some
early encouragement for such claims was offered in Irwin Toy when the Supreme
Court expressly left open the possibility that section 7 could offer anti-poverty
protection.65 Indeed, even after the decision in Gosselin, in which a claim that
inadequate social assistance violated sections 7 and 15 of the Charter was rejected by
a majority of the Court, this possibility remains open. Generally speaking, however,
anti-poverty and poverty-related claims have met with little success.

The role of competence concerns in anti-poverty cases, though, is not
straightforward. In the first place, in many instances in which anti-poverty claims

63 Initially, this distinction was reiterated and relied upon, for instance, in Reference Re ss. 193 and
195.1(1)(c) of the Criminal Code (Man.), [1990] 1 S.C.R. 1123, 109 N.R. 81, [1990] 4 W.W.R. 481
and McKinney, supra note 8. Its demise occurred over the course of decisions in the following cases:
R. v. Butler, [1992] 1 S.C.R. 452, 89 D.L.R. (4th) 449, 134 N.R. 81 [Butler cited to S.C.R.]; RJR-
MacDonald, supra note 9; OConnor, supra note 11; Thomson Newspapers Co. v Canada (A.G.),
[1998] 1 S.C.R. 877, (1997) D.L.R. (4th) 385, 38 O.R. (3d) 735 [Thomson Newspapers cited to
S.C.R.].

64 However, it must be noted that in Chaoulli, supra note 4, the majority and minority judges alike
referred to the legislation at issue as social policy, and seemed to link competence concerns to that
label, which may indicate a lingering refusal to recognize the incoherence of the distinction.

65 While rejecting an argument that s. 7 protected the economic liberty of a business corporation, the
SCC expressly refrained from foreclosing the possibility that the section protected economic rights
fundamental to human life or survival (Irwin Toy, supra note 38 at 1003).

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 525

were rejected, competence concerns were not mentioned.66 In contrast, in some of the
instances in which competence concerns were considered, the claims also failed,67 but
in other instances they succeeded.68 The question thus becomes whether the
recognition and treatment of competence concerns in anti-poverty cases is consistent
and coherent, both within the set of such cases and in relation to other types of cases.
It being beyond the scope of this article to undertake a detailed survey of these cases,

66 The following are examples of those claims that have been rejected by lower courts on the basis
of text-oriented interpretative factors and without reference to competence concerns: Brown v. B.C.
(Minister of Health) (1990), 66 D.L.R. (4th) 444, 66 B.C.L.R. (2d) 294, 48 C.R.R. 137 (rejecting a
claim based on ss. 7 and 15 for access to expensive and only partially publicly subsidized HIV/AIDS
medication); Ontario Nursing Home Association v. Ontario (1990), 72 D.L.R. (4th) 166, 74 O.R. (2d)
365, 21 A.C.W.S. (3d) 1278 (rejecting claims based on ss. 7 and 15 to protection against inadequate
care for extended care residents of nursing homes); Fernandes v. Manitoba (Director of Social
Service) (1992), 93 D.L.R. (4th) 402, 78 Man. R. (2d) 172, 7 Admin. L.R. (2d) 153 (rejecting claims
based on ss. 7 and 15 to an additional allowance for provision of full-time health care services in the
home, rather than as an in-patient); Bernard v. Dartmouth Housing Authority (1988), 53 D.L.R. (4th)
81, 88 N.S.R. (2d) 190, 50 R.P.R. 12 (rejecting claims based on ss. 7 and 15 to protection against a no-
grounds eviction from public housing). But see Sparks v. Dartmouth/Halifax County Regional
Housing Authority (1992), 112 N.S.R. (2d) 389, 33 A.C.W.S. (3d) 48 (for a different result on s. 15);
Conrad, supra note 2 (rejecting a claim based on ss. 7 and 15 to protection against withdrawal of
social assistance, without a hearing, on the ground of having resumed living with a spouse); R. v.
Clarke (1998), 23 C.R. (5th) 329, 40 W.C.B. (2d) 394 (rejecting a claim based on s. 7 to protection
against charges of mischief in relation to attempted occupation of vacant housing); R. v. Banks (2001),
205 D.L.R. (4th) 340, 55 O.R. (3d) 374, 45 C.R. (5th) 23 (rejecting claims based on ss. 7 and 15 to
protection against anti-panhandling laws).

67 The main claims to have failed are those in R. v. Prosper, [1994] 3 S.C.R. 236, 118 D.L.R. (4th)
154, 172 N.R. 161 [Prosper cited to S.C.R.]; PUC, supra note 17; Masse, supra note 15; Collins,
supra note 22; and Gosselin, supra note 3.

68 The main claims that have succeeded are those in Ttrault-Gadoury v. Canada (Employment and
Immigration Commission), [1991] 2 S.C.R. 22 [Ttrault-Gadoury]; Schachter v. Canada, [1992] 2
S.C.R. 679, 93 D.L.R. (4th) 1, 139 N.R. 1 [Schachter cited to S.C.R.]; R. v. Rehberg (1993), 111
D.L.R. (4th) 336, 127 N.S.R. (2d) 331, 19 C.R.R. (2d) 242 [Rehberg cited to D.L.R.]; Eldridge v.
British Columbia (A.G.), [1997] 3 S.C.R. 624, 151 D.L.R. (4th) 577, 218 N.R. 161 [Eldridge cited to
S.C.R.]; New Brunswick (Minister of Health and Community Services) v. G. (J.), [1999] 3 S.C.R. 46,
177 D.L.R. (4th) 124, 244 N.R. 276 [G.(J.) cited to S.C.R.]; Auton (Guardian ad litem of) v. British
Columbia (A.G.) (2002), 220 D.L.R. (4th) 411, [2003] W.W.R. 42, 173 B.C.A.C. 114 [Auton (BCCA)
cited to D.L.R.] (but this claim has now been rejected by the Supreme Court without addressing
competence concerns: [2004] 3 S.C.R. 657); Hodge v. Canada (Minister of Human Resources
Development), [2003] 1 F.C. 271, (2002), 214 D.L.R. (4th) 632, 2002 FCA 243 [Hodge cited to F.C.]
(but this claim has now been rejected by the Supreme Court without addressing competence concerns:
[2004] 3 S.C.R. 357); Falkiner v. Ontario (Ministry of Community and Social Services) (2002), 212
D.L.R. (4th) 633, 59 O.R. (3d) 481, 159 O.A.C. 135 [Falkiner]; Nova Scotia (Workers Compensation
Board) v. Martin, [2003] 2 S.C.R. 504, 231 D.L.R. (4th) 385, 310 N.R. 22 [Martin cited to S.C.R.].

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

526

in what follows I present the main conclusions that can be drawn from such a
survey.69

To begin, it can be said that when competence concerns have arisen in anti-
poverty cases, they have generally been recognized and treated in terms of, and
consistently with, the foundational framework established by the Supreme Court. In
this respect, and beyond general statements about the complexity of social policy
making in relation to poverty, one of the main concerns of courts has been their
competence to adjudicate upon the appropriate balance between the potential fiscal
impact of anti-poverty claims and competing claims to fiscal resources. Another main
competence concern has arisen in relation to choosing and designing remedial means
(especially when fiscal resources may be required).
At a more specific level, the second conclusion that can be drawn from a review
of competence concerns in anti-poverty cases is that there is some difficulty in
explaining differences in the recognition of the degree of competence challenges
arising in particular cases. This can be illustrated by contrasting the decisions in R. v.
Prosper70 and New Brunswick (Minister of Health and Community Services) v. G.
(J.).71 Both Prosper and G.(J.) concerned claims to government-funded legal services.
The claim in Prosper, based on the subsection 10(b) Charter guarantee of the right to
retain and instruct counsel, was to after-hours duty counsel services. The claim in
G.(J.), based on section 7, was for legal representation to appeal an extension of an
order for state removal of children from their family home. Although the claimant
achieved some success in Prosper, the specific claim to after hours duty-counsel
services was held injusticiable. This was in part due to concerns that the Court lacked
the competence to review the allocation of limited fiscal resources, to predict and
assess the potentially far-reaching consequences of recognizing a right to state-funded
duty counsel and to devise an appropriate remedy.72 In contrast, the claim in G.(J.)
was successful, with the competence concerns that were raised in Prosper being
brushed aside. Specifically, the majority in G.(J.) took the position that the scope of
the obligation was relatively limited, that the potential budgetary impact of upholding
the claim was therefore minimal, and that the detrimental effect of denying assistance
to parents outweighed the benefits of any budgetary savings.

The arguments underlying each of these grounds for distinguishing these cases
are not, however, quite as self-evident as the majority judgment seems to assume. The
actual cost of complying with the ruling in G.(J.) was only vaguely adverted to in the
reasoning, while in Prosper it was simply assumed to be significant. Moreover, at no
point was there any explanation of how, or in what way, competence challenges

69 A detailed survey is undertaken in D. Wiseman, Judging Poverty: The Charter, Poverty and
Institutional Competence (S.J.D. Thesis, Faculty of Law, University of Toronto, 2005) [forthcoming in
University Microfilms International].

70 Supra note 67.
71 Supra note 65.
72 See Prosper, supra note 67 at 288, LHeureux-Dub J., and at 267-68, Lamer C.J.C.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 527

increase as potential budgetary impact increases. Nor was there any explanation of
why competence concerns are non-existent or can be ignored when budgetary impact
is minimal. Further, the position as to the differential scope of the rulings depends
entirely upon how generally the ruling is stated. In Prosper, the majority took the
position that the claim to twenty-four-hour availability of duty counsel, if upheld for a
person being asked to take a breathalyzer test in Nova Scotia, would of necessity
apply to all arrests and detentions throughout the country, and thus would have far-
reaching implications. However, in G.(J.) the Court apparently neglected to consider
that the immediate claim, to state-funded legal counsel to ensure a fair hearing in a
state-custody order extension hearing in New Brunswick, would surely apply to all
such hearings across the country and could easily apply to a myriad of judicial and
administrative hearings affecting child-parent relationships for which legal aid is not
presently available. And even if the scope of the obligation in G.(J.) would still be
more confined than that in Prosper, the more it expands from the immediate case the
more significant competence concerns would become. Finally, perhaps it can be
conceded that the interests at stake of a parent unable to obtain legal representation to
appeal a state-custody extension order are greater than those at stake for a person
refusing to take a breathalyzer test when duty counsel are unavailable. But it is not
difficult to imagine circumstances in which the unavailability of duty counsel might
be of greater significance (e.g., if a person is being questioned in relation to a
homicide).

The third conclusion that can be drawn from a review of the treatment of
competence concerns in anti-poverty cases is that there is also some difficulty in
explaining the preference for different responses to competence concerns across the
range of anti-poverty and poverty-related cases. Again, the decision in Prosper is
illustrative. By holding that there was no right to summary legal advice under
subsection 10(b), the Court in Prosper not only denied that right to people being
asked to take breathalyzer tests but also to anybody else in need of summary legal
advice, no matter how grave the circumstances. However, it seems fairly easy to
imagine such graver circumstances arising. Therefore, it can be argued that the Court
would have been better to respond to the competence concerns raised in Prosper,
which really related to issues arising in the section 1 and remedy review stages,
through means of deference or remedial restraint, rather than injusticiability. For
instance, insofar as the potential budgetary impact of upholding the claimed
obligation raised competence concerns, it should be noted that similar concerns were
addressed through remedial restraint, in the form of a suspended declaration of
invalidity, in some earlier anti-poverty cases,73 and have been addressed similarly in
at least one subsequent anti-poverty case.74
Fourth, and finally, a review of the treatment of competence concerns in anti-

poverty cases reveals the possibility that the magnitude of the potential fiscal impact

73 See Eldridge, supra note 68, and Rehberg, supra note 68.
74 See Falkiner, supra note 68.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

528

of anti-poverty claims is becoming a determining factor both with respect to the
emphasis given to competence concerns and to the responses to these concerns. When
the potential fiscal impact is purported to be of significant magnitude,75 competence
concerns are emphasized and responses of more
limiting effect, such as
injusticiability and deference, are chosen. By contrast, when the potential fiscal
impact is purported to be minimal,76 competence concerns are rejected or ignored
and responses of least limiting effect, in particular remedial restraint, are chosen.
Such trends suggest a more general argument that courts are competent to assess the
reasonableness of balances involving the allocation of fiscal resources when the
quantity of resources at stake can be confidently estimated and is relatively small, but
that competence concerns become significant as potential budgetary impacts rise
above the minimal level. However, there are some problems with any such argument.
Since this argument has surfaced regularly enough to become a general development
in the framework for considering competence concerns, and because it potentially
poses a significant barrier to any adequate social assistance obligation, these
problems need to be considered in greater depth.

b. Fiscal Impact Magnitude as Competence Measure: Problems
An initial problem with using the magnitude of potential fiscal impact as a
measure of competence is that the metric for assessing whether potential fiscal
burdens are immaterial or material is vague and potentially inconsistent. In G.(J.) the
Court adverted both to the absolute dollar cost of providing the program necessary to
fulfill the potential obligation and to the relative proportion of the governments
entire budget this represented. Earlier, in Eldridge,77 in which an obligation to provide
interpreter services for deaf patients in the B.C. health care system was upheld, the
Court noted the absolute dollar cost (around $150,000) of providing the services but
also identified the relative proportion of the relevant departmental budget this

75 See e.g. Prosper, supra note 70, Masse, supra note 15, and Collins, supra note 22.
76 See e.g. Rehberg, supra note 68, Falkiner, supra note 68, and G.(J.), supra note 71. In Martin,
supra note 68, the government suggested that the limitation on benefits for chronic pain sufferers was
justified in the interests of maintaining the viability of the workers compensation scheme. This
implied that the potential fiscal impact was significant, but the Court responded that the government
had not presented sufficient evidence to establish this as a pressing and substantial objective. Further,
in the minimal impairment stage, the Court noted that fiscal impact is a grounds for deference, but
rejected a need for deference on the basis that there was a total impairment. The lower court decisions
in Auton (BCCA), supra note 68, and Hodge, supra note 68, may seem to counter the idea of a
correlation between magnitude of fiscal impact and success. In those cases the fiscal impacts were
potentially significant. But it should be noted that in the former it was argued that there were
counterbalancing long-term fiscal savings and in the latter the government was forced to bear the
burden of not having led any meaningful evidence on the impact.

77 Supra note 68.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 529

represented.78 Obviously enough, depending upon the circumstances, which measure
is used can yield quite a different point of materiality.
A further problem is that if G.(J.) and Eldridge establish the threshold of
materiality at around $150,000, above which the courts will either render a claim
injusticiable or defer to its limitation in the section 1 stage, then the scope of rights
protection, at least with respect to anti-poverty claims, will be very limited. This is a
problem in two ways. First, the idea that the fiscal impact of certain claims can justify
excluding those claims from the scope of protection offered by the Charters rights
and freedoms, through injusticiability, has generally been disavowed in Charter
decisions. The initial position of the courts was that costs could be taken into account
in determining the type of remedial order, but could not be relied upon to deny a
remedy nor to justify a non-minimal impairment nor to establish a pressing and
substantial objective for a limitation nor to truncate the scope of a right or freedom.79
After Irwin Toy, it was accepted that costs could be taken into account in choosing
between alternative limiting measures, but it was only with the decision in
Newfoundland (Treasury Board) v. Newfoundland and Labrador Association of
Public and Private Employees (N.A.P.E.)80 that the Supreme Court allowed fiscal
impact to be relied upon as a pressing and substantial objective for a limiting
measure. Even then, the Court took pains to emphasize that the fiscal impact had to
be of crisis proportions.81 That being the case, the decisions to hold the claims in
some anti-poverty cases injusticiable due to competence concerns associated with
potential fiscal impact appear inconsistent with the broader landscape of Charter
principles.

Second, accepting or rejecting claims according to the potential degree of fiscal
impact they involve is an approach that entails denying, rather than engaging,
competence concerns. Even if it can be argued that the greater the potential fiscal
impact the greater the challenges posed for the court, it does not necessarily follow
that there is a point where the fiscal impact is so small that competence is no longer
challenged. Thus, the unspecified amount at stake in G.(J.) may have been a small
proportion of the provincial budget of New Brunswick, just as the $150,000 at stake
in Eldridge may have been a small proportion of the health budget in British
Columbia, but there are surely many ways in which these amounts might be spent. If
the budget is tight, then that money will have to be taken from some other area. If the
need to assess fiscal resource allocation issues challenges competence because the
courts lack the normative and empirical expertise and standards to weigh and balance
competing needs, and if such assessments are required for reallocating even

78 It is worth noting that in Eldridge, ibid., as in the later case of Martin, supra note 68, the Court
also relied upon the fact that the impairment of the right at issue was total, rather than partial. This
argument was also relied upon in Ttrault-Gadoury, supra note 68, which preceded Eldridge.

79 See Singh, supra note 26, and Schachter, supra note 68.
80 [2004] 3 S.C.R. 381, 244 D.L.R. (4th) 294, 2004 SCC 66 [N.A.P.E. cited to S.C.R.].
81 See ibid. at paras. 59 to 76. It might be argued that the lack of rigour of the Courts assessment of

the purported crisis in that case undermines its own position.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

530

$150,000, then all that the courts are doing with the minimal impact distinction is
denying or ignoring the competence concerns. The cost of doing so, however, is the
severe limitation on the scope of rights protection.
A better approach, and one that is more consistent with decisions in non-poverty
cases (as will be explained below), would be for the courts to admit that competence
concerns will always be present when budgetary impacts are involved and to seek to
develop strategies for managing or overcoming those concerns. And this is especially
so because, contrary to what the courts may think, many non-poverty Charter claims
will have some budgetary impact and often a non-minimal impact. Many laws and
policies, if not all, are supported by or place demands upon an apparatus (whether
legislative, judicial, administrative, or some combination thereof) that requires fiscal
resources. While this is most evident with respect to social-benefits laws and policies,
such as social assistance and public health insurance, it is also true for laws and
policies that provide other types of services or that otherwise regulate social activity.
Whenever a court makes an order under the Charter that affects the substance or
process of a law or policy, and whether compliance requires negative or positive
governmental action, there will be a fiscal impact, at times only once-off, but at other
times ongoing. Two well-known examples of cases that do not address social benefits
but nevertheless involve significant fiscal impact are Singh and R. v. Askov.82 In
Singh, the Court imposed additional fair-hearing requirements in relation to
procedures for determining refugee status. It was reported that the initial estimate of
the additional cost of compliance, just for processing the backlog of refugee cases at
the time of the decision, was $50 million.83 In Askov the Court set out standards for
undue delay in criminal proceedings. The standard threatened to jeopardize so many
pending cases that, in the year following the decision, the Ontario government
reportedly expended an additional $39 million on increasing the capacity of the
criminal trial system in just one region of the Greater Toronto Area.84
Moreover, the prospect of significant fiscal impact extends beyond non-social-
benefits cases addressing procedural rights. Consider, for instance, the decision in
RJR-MacDonald to strike down some of the regulations affecting tobacco promotion
and packaging. Although the decision required only negative governmental action
and did not directly engage any social benefit program, the fiscal impact would
potentially have included the following: first, the now-wasted cost of the legislative
and administrative resources devoted to creating and implementing the regulations
prior to the decision; second, the future one-off cost of the legislative and
administrative resources required to withdraw or modify the regulations in
accordance with the decision; third, the future ongoing cost of the public health
resources and resources of other relevant social benefits programs that could have
been saved by any decline in tobacco consumption caused by the invalidated

82 [1990] 2 S.C.R. 1199, 74 D.L.R. (4th) 355, 113 N.R. 241 [Askov cited to S.C.R.].
83 The Globe and Mail (22 May 1985), cited in Mandel, supra note 49 at 243.
84 Law Times 1991:1 (24-30 June 1991), cited in Mandel, ibid. at 227.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 531

regulations; and, fourth, the future ongoing cost of re-establishing whatever public
programs were replaced by, or now needed to replace, the regulations. No mention
was made of any of these potential fiscal impacts in any judgments in the case, and
yet it is hard to believe they would amount to less than $150,000 per annum. Thus, if
fiscal impact is to be treated as an important issue in Charter cases, then it needs to be
recognized that it will be important in many cases beyond those anti-poverty cases
that directly seek an increase in the fiscal resources devoted to a social benefit
program. And if that is the case, then there is even more reason to abandon the use of
the minimal fiscal impact standard as a reference point for determining the degree
of Charter protection and to instead explore measures that would build court
competence in relation to fiscal issues.

Indeed, that is in fact what the courts are doing in some non-poverty cases, which
brings me to the third way in which using the magnitude of fiscal impact as a
determinant of injusticiability or deference in anti-poverty cases is a problem, namely,
because those responses are inconsistent with the responses used in non-poverty cases
under the Charter and other provisions of the Constitution. Most significantly, in
some non-poverty cases involving non-minimal fiscal impacts the courts have been
willing to undertake competence-building measures. This raises the question of why
such measures are not deployed in anti-poverty cases. But as the pursuit of
competence-building measures is one of the other main developments in the treatment
of competence concerns in Charter adjudication, I will defer further discussion of this
point until section 3, below.
Until these problems can be addressed, then, the idea, emerging from anti-poverty
cases, that the magnitude of the potential fiscal impact of a claim can be used as a
reference point for determining whether a court ought to withdraw from the claim on
competence grounds, must be regarded as unconvincing.85 As this idea emerged
subsequent to the establishment of the foundational framework, it is not an integral
part of the anti-poverty incompetence argument based upon that framework.
However, just as its emergence certainly has the potential to bolster that argument, its
problems foreclose that potential. Further, to the extent that the foundational
framework gives some significance to whether a claim necessitates review of
competing claims to fiscal resources, my argument that competence concerns relating
to fiscal issues need to be engaged, rather than ignored, is equally applicable. In turn,

85 One final observation can also be made as a consequence of this review, namely, that anti-poverty
and poverty-related cases appear less likely to succeed when they are cast in terms of violations of
Charter rights other than the right to equality. Of the claims that have succeeded, only the one in G.(J.)
was not cast in terms of s. 15. The claim in G.(J.) was cast in terms of s. 7 and, apart from that case, all
other s. 7 anti-poverty claimsthose in Masse, supra note 15, PUC, supra note 17, Rehberg, supra
note 68, Falkiner, supra note 68, and Gosselin, supra note 3along with the s. 10(b) claim in
Prosper, supra note 67, have either been rejected or left undecided. This raises the question of why the
courts tend to opt for the most severe response when anti-poverty claims that raise competence
concerns are cast in terms of rights and freedoms other than equality and whether that tendency is
justifiable. I will not take up that question here.

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

532

that argument goes some way to countering an aspect of the anti-poverty
incompetence argument.

[Vol. 51

2. Reconstructing Deference: A Normative Counterbalance

The third general development has been the reconstruction of the group of factors
recognized as relevant to decisions as to deference and, as part of that, an effort to re-
emphasize a normative component to such decisions. The primary source of this
development is the judgment of Justice Bastarache, on behalf of the majority, in
Thomson Newspapers.86 In what follows, I begin with an analysis of this judgment
and its potential to assist in countering the anti-poverty incompetence argument. I
then discuss and assess subsequent applications and modifications of the framework
of factors established by Justice Bastarache in Thomson Newspapers, with particular
emphasis on M v. H.87

a. The Paramountcy of Vulnerable Groups Interests

Thomson Newspapers concerned a challenge to provisions of the Canada
Elections Act88 that prohibited broadcasting, publication, and dissemination of
opinion survey results during the final three days of a federal election as violations of
freedom of expression (subsection 2(b)) and the right to vote (section 3) guaranteed in
the Charter. By the time this case was argued before the Court, the government had
conceded that the prohibitions violated subsection 2(b) and so the focus of argument
and judicial attention was on the application of section 1. By majority, the Supreme
Court held that the violation could not be saved under section 1. It was in framing the
majoritys section 1 analysis that Justice Bastarache reconsidered the factors relevant
to decisions as to whether to apply the deferential reasonable basis standard in
conducting the section 1 review. Justice Bastarache was uncomfortable either with the
language of deference, or with the idea that there were only two standards of section
1 review (rigourous or deferential), or both. Picking up the language of
contextualism, he argued that decisions about what types of arguments and what
degree of evidentiary certainty were required to satisfy the demands of section 1
needed to be made in the context of the case at hand.
At a general level, what the so-called contextual approach amounted to was a
commitment to avoid considering the questions posed by section 1 review in a way
that abstracts from the specific details and factual setting of the social problem and
governmental decisions at issue. According to Justice Bastarache, while paying close
attention to context is generally important for section 1 review, it is particularly
relevant in determining the type of proof which a court can demand of the legislator

86 Supra note 63.
87 [1999] 2 S.C.R. 3, 171 D.L.R. (4th) 577, 238 N.R. 179.
88 R.S.C. 2000, c. 9.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 533

to justify its measures under s. 1.89 The background to this statement is the position
taken by the Court in the earlier decision of RJR-MacDonald: arguments based in
reason, logic and common sense are as capable of satisfying the civil standard of
proof as are arguments based in social science. This position meant that, in the
absence of definitive social science evidence, the section 1 standard could still be
satisfied. Moreover, it was argued, satisfying the section 1 standard with nonsocial
science arguments did not necessarily entail deference. Rather, it might merely entail
an appreciation of context.
On closer inspection, according to Justice Bastarache, this willingness to accept
less-than-scientific types of proof could be attributed to the presence of one or more
of a number of contextual factors, namely, an attempt to protect a vulnerable group;90
a groups own subjective fears and apprehension of harm;91 an inability to
scientifically measure the particular harm in question, or the efficaciousness of the
remedy;92 and the nature of the activity infringed or, more accurately, the importance
of the activity infringed.93 In Justice Bastaraches view, these are all factors of which
the court must take account in assessing whether a limit has been demonstrably
justified according to the civil standard of proof.94

Two points are worth noting about this framework of factors. First, the factors
identified by Justice Bastarache have in common a correlation to the protection of the
interests of vulnerable groups. The effect of taking them into accountand in so
doing allowing less-than-scientific satisfaction of the section 1 standard of proofis
to make it easier for governments to justify limitations when those limitations are
aimed at the protection of the interests of vulnerable groups.95 Second, although each
of the factors can be grounded either in concerns over competence (that is, that the
courts lack the competence to appropriately weigh the interests of vulnerable groups)
or concerns over legitimacy (that is, that the Charter should not become a vehicle for
undermining measures protective of vulnerable groups), or both, their tight

89 Thomson Newspapers, supra note 63 at para. 88.
90 This factor was relied upon for deference in Irwin Toy, supra note 38.
91 This factor was relied upon for deference in Butler, supra note 64. In Butler the SCC was called
upon to determine whether the Criminal Code prohibition on possessing, selling and exposing to
public view obscene films violated the Charters guarantee of freedom of expression. The Court
unanimously adopted a deferential position in the s. 1 review stage, in part because of competence
challenges posed by conflicting and inconclusive social science evidence.

92 This factor was relied upon for deference in both Irwin Toy, supra note 38, and Butler, ibid.
93 For example, whether hate speech or commercial speech is of high or low valuefactors
considered in Irwin Toy, ibid., Butler, ibid., and other freedom of expression cases in the s. 1 review
stage.

94 Thomson Newspapers, supra note 63 at para. 90.
95 For example, the vulnerable group of children, manipulated by commercial advertising (in Irwin
Toy, supra note 38); the vulnerable group of women, demeaned by pornography (in Butler, supra note
64); the vulnerable groups of minorities, exposed to hate speech (in cases such as R. v. Keegstra,
[1990] 3 S.C.R. 697, 197 N.R. 26, 39 Alta. L.R. (3d) 305) and the vulnerable group of people addicted
to or attracted to smoking (as in RJR-MacDonald, supra note 9).

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

534

connection to the concern to protect vulnerable groups has the effect of limiting the
range of circumstances in which concerns over competence (or legitimacy) will
justify acceptance of less-than-scientific proof. Specifically, the effect is that
competence concerns will only justify a more loose standard of proof when that
standard would work to the advantage of vulnerable groups (via the governmental
actions that limit rights in order to protect vulnerable groups). As such, Justice
Bastaraches rearticulation of the issue of deference as the issue of context
represented a potentially significant modification of the role of competence concerns
in Charter adjudication. If this framework were adhered to, then competence
concerns would play second fiddle to the normative concern for the interests of
vulnerable groups.

This modification has the potential to make a significant contribution to
countering the anti-poverty incompetence argument as it might be applied to the
generic social assistance challenge. Specifically, since that challenge impugns a
governmental failure that operates to the detriment of a vulnerable groupthe poor
none of the contextual factors would be applicable and so a more loose standard of
justification would not be available to the government in the section 1 review stage.
But within a year, the approach elaborated by Justice Bastarache in Thomson

Newspapers, on behalf of a majority of the Court, was being implicitly contested by
other members of the Court. Moreover, Justice Bastarache himself was augmenting
his framework in such a way as to widen the role of competence concerns and the
circumstances in which a less stringent standard of proof would be allowed.

b. Vulnerable Groups Interests as Counterbalance

The seeming retreat from the paramountcy of the interests of vulnerable groups
occurred in the decision in M. v. H., which concerned a claim that the exclusion of
same-sex couples from the spousal support scheme of the Family Law Act96 violated
section 15 of the Charter. By majority, the Court held that section 15 had been
violated and that the violation could not be justified under section 1 of the Charter. In
rejecting the Attorney General of Ontarios submissions under section 1, both Justice
Iacobucci (who, with Justice Cory, wrote the leading majority judgment) and Justice
Bastarache (who wrote a separate judgment to the same result) considered the limits
of institutional competence in the context of addressing the issue of deference.
What is interesting about Justice Iacobuccis section 1 analysis is that his
explanation as to why and in what circumstances a court may need to be deferential
centres entirely, and expressly, on the issue of the limits of the institutional
competence of courts. Further, while Justice Iacobucci acknowledges the foundations
laid in Irwin Toy and specifically mentions the factors of competing claims and
conflicting social science evidence, he does not mention the factor of the stakes of

96 R.S.O. 1990, c. F-3.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 535

vulnerable groups. Although this aspect of Justice Iacobuccis judgment is not cast as
an objection to the framework set out by Justice Bastarache in Thomson Newspapers,
that is what it amounts to. And perhaps it was successful, because in his separate
judgment reaching the same result as Justice Iacobucci, Justice Bastarache augments
his framework by identifying three further factors that ought to be taken into account
in conducting the section 1 review. First, he identified the representativeness of the
process giving rise to the decision being challenged as a relevant factor, with the
argument being that the less representative the process the greater should be the
rigour of the section 1 analysis. The second factor he identified was the sensitivity of
the moral judgments embodied in the decision, with the argument being that the more
sensitive or, more accurately, the more controversial they were, the more forgiving
would be the section 1 analysis. Finally, Justice Bastarache identified the factor of the
polycentricity or complexity of the situation from which the claim arose as relevant,
with the argument being that the more complex the situation, the more circumspect
the court needed to be with any intervention.
As with the factors he identified in Thomson Newspapers, the three additional
factors Justice Bastarache identified in M. v. H. can be understood to be based either
in concerns over competence or concerns over legitimacy, or both, although he only
made a point of mentioning competence concerns in relation to the factor of
complexity. But, in contrast to the orientation of the factors he identified in Thomson
Newspapers, Justice Bastaraches new factors did not appear to be informed by any
special concern to protect the interests of vulnerable groups. With respect to the factor
of representativeness, it may well be the case that decisions detrimentally affecting
the interests of vulnerable groups, including decisions detrimental to social assistance
recipients, will be made in less representative processes (for instance, closed-door
Cabinet or Ministerial processes of amending social assistance regulations). But
similar processes might also be used for decisions detrimentally affecting advantaged
interests. With respect to the factor of moral sensitivity, since it seems reasonable to
argue that it is the moral status quo that creates vulnerability, and that efforts to
reform that status quo in favour of vulnerable groups will seem morally controversial,
the second factor might militate against the interests of those groups. Finally, to the
extent that vulnerable groups may often assert claims against governments in
polycentric circumstancesand here it must be noted that social assistance decisions
are certainly polycentricthe third factor has the potential to allow governments a
lower standard of justification in limiting those claims.
At the same time, though, Justice Bastaraches discussion in M. v. H. of the
factors earlier identified in Thomson Newspapers confirmed his continuing concern
that decisions as to the rigour of section 1 review ought to attend to the implications
for vulnerable groups. This was particularly evident in his discussion in M. v. H. of
what approach courts ought to take when reviewing the balances struck by
governments between competing interests. For Justice Bastarache, the issue of
balancing was to be approached via two of the factors he had identified in Thomson
Newspapers, namely, whether there was an attempt to protect the interests of a
vulnerable group and the nature of the interest affected. In keeping with the earlier

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

536

cases, Justice Bastarache argued that where the balancing of competing interests
involved the interests of vulnerable groups and resulted in a measure that was
protective of those groups, then the court should consider applying a less stringent
standard of review. But in a move beyond the earlier cases, Justice Bastarache also
argued that where such balancing appeared to result in a measure that was detrimental
to the interests of vulnerable groups, then the Court should consider applying a more
stringent standard of review. The implication of these arguments was that, for Justice
Bastarache, the challenges to competence associated with balancing ought to be
responded to by means that give the benefit of the lack of competence to vulnerable
groups. Thus, whether the need for balancing would allow the government a
deferential standard of proof would depend upon the extent to which the
governments actions were aligned with the interests of vulnerable groups.

So, on the one hand, as Justice Iacobuccis judgment led the majority, and given
the additional factors identified by Justice Bastarache, the reasoning in M. v. H.
constituted something of a retreat from the position in Thomson Newspapers that
competence concerns would play second fiddle to the interests of vulnerable groups.
On the other hand, the need to consider the implications of deference for the interests
of vulnerable groups continued to be emphasized.

In subsequent cases in which Justice Bastaraches framework has been applied, a
concern to assess on which side of the case, if any, the interests of vulnerable groups
lie, and to take that into account in determining whether to be deferential, has been
evident.97 Specifically, in a case concerning the ability of RCMP officers to
participate in labour relations regimes, a minority of the Court applied the framework
and rejected a need for deference. In so doing, it rejected the governments argument
that the exclusion was meant to protect the vulnerable public from a police strike and
instead held that it was the officers, as employees, who were the vulnerable group.98
In another case concerning a freedom of expressionbased challenge to Criminal
Code prosecution in relation to possession of child pornography, a minority applied
the framework and decided to be deferential. In so deciding, it noted that the
legislation was attempting to protect the vulnerable group of children.99 Finally, in a
case concerning the exclusion of agricultural workers from a labour relations regime,
the minority analysis in the RCMP case was followed and accordingly deference was

97 It is worth noting that in subsequent cases a re-emphasizing of a normative dimension to decisions
as to deference has been evident even though competence concerns have not arisen. Thus, in Sauv v.
Canada (Chief Electoral Officer), [2002] 3 S.C.R. 519, 294 N.R. 1, 2002 SCC 68, a majority rejected
deference to a limitation on the right to vote for prisoners because the right was of fundamental
normative significance. At the same time, the fact that the limitation would have a disproportionate
detrimental impact on Aboriginal peoples (because they are disproportionately disadvantaged and
imprisoned) was taken into account in the final stage of the s. 1 analysis. For a broader discussion of
the need to return a normative dimension to s. 1 analysis, see Timothy Macklem & John Terry,
Making the Justification Fit the Breach (2000) Sup. Ct. L. Rev. 11 S.C.L.R. (2d Series) 575.
98 Delisle v. Canada (Deputy A.G.), [1999] 2 S.C.R. 989, 176 D.L.R. (4th) 513, 244 N.R. 33.
99 R. v. Sharpe, [2001] 1 S.C.R. 45, 194 D.L.R. (4th) 1, 2001 SCC 2.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 537

not adopted on account of the fact that the governmental measure harmed the interests
of the vulnerable workers.100

Consequently, Justice Bastaraches initial efforts to give greater weight to the
interests of vulnerable groups in decisions as to deference appear to have wrought a
lasting adjustment to the foundational framework for the recognition and treatment of
competence concerns. Competence concerns may not have been relegated to a
secondary consideration behind the normative concern to protect the interests of
vulnerable groups, but certainly the potential for deference to work to the
disadvantage of those interests has been established as a normative factor that can
counterbalance competence concerns. In relation to the generic social assistance
challenge, which involves the vulnerable group of the poor, the subsequent
development of Justice Bastaraches Thomson Newspapers framework thus preserves
the availability of a means for countering the anti-poverty incompetence argument.
The subsequent developments may mean that the interests of the poor no longer
trump competence concerns, but they do at least counterbalance them. More
concretely, what this could mean is that until a government can provide reasonably
conclusive evidence on the so-called trickle-down effect of such measures as tax cuts
and deficit reduction, a court could rely upon the normative importance of the
interests of the poor to impose the burden of the conflicting evidence on the
government. Thus, when the government calls for deference to its belief that failing to
provide adequate social assistance is minimally impairing because of the resources
devoted to generating economic opportunities in general, a court could rely upon the
stake of the poor to refuse to defer.

In relation to the anti-poverty incompetence argument more generally, it is worth
noting that the establishment of the normative counterbalance might have more far-
reaching implications for the choice of response to competence concerns. In all of the
cases considered in this section, the normative counterbalance is relevant not at the
point where a court is determining its degree of competence but, rather, at the point
where a court is determining how to respond to competence challenges. And in all of
those cases the response being considered was deference. However, deference is only
one of four possible responses to competence challenges. In three of the five
situations in which Justice Bastaraches approach has been utilized, it was decided
that deference was not appropriate, but that did not mean that the competence
challenges were ignored. Rather, in each instance where deference was refused, and
the Charter violation was not saved under section 1, the response of remedial
restraint was recommended. As a result, the main effect of the establishment of the
normative counterbalance appears to be to prompt a consideration of less severe
responses to competence challenges. This effect, I would suggest, ought not to be
confined to decisions as to deference. In other words, the interests of vulnerable
groups should not only be available as a counterbalance to competence concerns in

100 Dunmore v. Ontario (A.G.), [2001] 3 S.C.R. 1016, 207 D.L.R. (4th) 193, 279 N.R. 201

[Dunmore].

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

538

decisions as to deference. They should also be available as a counterbalance in
decisions as to injusticiability, remedial restraint and competence-building. And if the
effect of this counterbalance is to prompt consideration of less severe responses, then
it should create a presumptive preference for utilizing the least severe response. That
is, the presumptive preference should not merely be for remedial restraint over
deference but for competence building over both (and over injusticiability as well).
Any such presumptive preference would provide a significant counter to the anti-
poverty incompetence arguments preference for injusticiability or deference.101

By the same token, it must be admitted that the framework of deference factors
has two problems. These problems will need to be addressed if the potential of the
normative counterbalance to undermine the anti-poverty incompetence argument is to
be coherently realized.

First, there is the problem of weighing and balancing the various factors should
the circumstances of a case be such that the various factors pull in different directions.
In each of the situations in which Justice Bastaraches framework was applied, the
various factors were all regarded as pulling in the same direction, that is, against
deference. But it is entirely possible that in certain circumstances the factors could
pull in different directions, which raises the potential problems associated with having
to weigh and balance the factors. This, in turn, might especially be a problem in the
circumstances of the generic social assistance challenge, where the interests of the
vulnerable group in receipt of social assistance may often coincide with polycentricity
in social welfare policy making, conflicting social science, demands on scarce
resources and competing interests. While the interests of vulnerable groups may
generally be available as a counterbalance to competence concerns, how much of a
counterbalance are they?102

101 The accumulated case law on competence contains little, if any, manifestation of such a
preference. Indeed, its only manifestation may be the extent to which the responses to competence
challenges in minority language education rights cases might be grounded in the normative
importance attributed to those rights. This failure to explore competence-building measures must be
seen as a shortcoming because, I would argue, it is precisely when the interests of vulnerable groups
are at stake that such exploration is most legitimate and desirable.

102 In fact, this very problem presented itself in Gosselin, supra note 3. In that case only Bastarache
J. addressed the issue of deference and he did so in the course of undertaking the s. 1 review stage in
relation to his finding (which was only that of a minority) that s. 15 had been violated. He opened his
s. 1 analysis by invoking his judgment in Thomson Newspapers, supra note 63, on the need for
contextual analysis and the need to consider the specific deference factors he identified therein, but his
ultimate discussion and application of those factors in Gosselin was perfunctory at best. Most
significantly, although this was clearly a situation in which Bastarache J. saw the factors pulling in
different directions, he did not explicitly address how the factors might be weighed and balanced or
even acknowledge this complication. By implication, it seems he reached the conclusion that the stake
of the vulnerable group is the most influential factor, since it is apparently capable of counteracting a
combination of factors pulling in the opposite direction. However, even this is by no means clear,
since the final level of deference is left somewhat vague. The vagueness of the level of deference was
then exacerbated by the lack of explicit reference to it in the detail of the s. 1 analysis.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 539

A second potential problem with Justice Bastaraches framework is that in the
cases in which the framework was applied it was possible to situate the vulnerable
interests on just one side of the caseeither as being protected by the impugned
measure or as challenging it. But what happens when vulnerable interests are at stake
on both sides of a challenge? Consider, for instance, the circumstances of R. v.
Mills,103 in which the Court had to review the governments response to its earlier
decision in OConnor104 that modified the common law evidentiary rules relating to
the production of complainant-related records to the accused in sexual assault cases.
Certainly, victims of sexual assault, whose interests were regarded as being protected
by the common law rules and the subsequent legislation, could be regarded as a
vulnerable group. But the criminally accused, even in sexual assault cases, can also
be regarded as a vulnerable group. Which vulnerable groups interest ought to take
priority in decisions as to deference? That said, perhaps this problem is less likely to
be relevant to a social assistance challenge, where the vulnerable group of people
living in poverty stands on one side and on the other stands the more diffuse interest
of the public in fair but prudent fiscal policy and budgetary allocations. Certainly
there are many budgetary allocations that could be regarded as protective of other
vulnerable groups, but assuming that not all government expenditure can be closely
associated with Charter guarantees (or that increasing taxes does not violate Charter
rights), the issue of balancing the interests of directly competing vulnerable groups is
unlikely to arise in a social assistance challenge.105

The general development of the reconstruction of the group of factors recognized
as relevant to decisions as to deference and, as part of that, the effort to re-emphasize
a normative component to such decisions, thus has some problems.106 Nevertheless,
since anti-poverty claims will typically seek to protect vulnerable groups, it is a
development that assists in countering the anti-poverty incompetence argument. In a
nutshell, the adjudication of a challenge to a failure to provide adequate social
assistance may raise valid competence concerns, but courts, in deciding how to
respond to those concerns, need to consider the counterbalancing factor that the
purpose of social assistance is to protect the interests of vulnerable groups. Indeed,
because of this, courts ought to presumptively prefer the response of exploring
competence-building measures or otherwise seek to respond in ways that seek to
provide as much Charter protection and scrutiny as possible. Consequently, to the
extent that the anti-poverty incompetence argument prefers the response of
injusticiablity, this development counters that injusticiability should only be

103 [1999] 3 S.C.R. 668, 180 D.L.R. (4th) 1, 248 N.R. 101 [Mills].
104 Supra note 11.
105 It should be noted that in N.A.P.E., the SCC seemingly accepted the argument that the
government of Newfoundland and Labrador was in the unenviable position of having to balance needs
of equal Charter importance (supra note 80 at paras. 93 and 95).

106 A third potential problem with Bastarache J.s framework is that in boosting the normative
component of decisions as to deference, Bastarache J. is assuming that courts have the competence for
normative analysis, but I will not address that problem here. See Dunmore, supra note 99.

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

540

considered, if at all, once all competence-building opportunities have been exhausted
and the less severe responses of remedial restraint and deference have been found
insufficient. In other words, injusticiability should be reserved for only the most
severe cases of incompetence.

[Vol. 51

3. Validating Creative Engagement or Competence-Building

The fourth general development in the recognition and treatment of competence
concerns in Charter adjudication has been the efforts undertaken by some courts to
modify aspects of the traditional forms of Charter adjudication. These efforts have
enabled courts to respond to competence concerns while maintaining an effective
degree of Charter protection and judicial supervision. The primary examples of this
development are the efforts undertaken by the Supreme Court of Canada in Reference
Re Remuneration of Judges of the Provincial Court of Prince Edward Island107 and
by a lower court (with subsequent Supreme Court approval)108 in Doucet-Boudreau v.
Nova Scotia (Minister of Education).109
In Judges Remuneration Reference, the Court considered a collection of claims,

made by numerous accused and a provincial court judges association, that reductions
in the salaries of provincial court judges by the governments of Prince Edward Island,
Alberta, and Manitoba compromised the independence of those judges in
contravention of subsection 11(d) of the Charter. The Court ultimately decided that
each province had contravened the requirements of section 11(d). The decision had a
number of interesting aspects. Garnering the most attention was the majoritys
argument as to the source of the guarantee of judicial independence. While the Court
decided the case on the basis of subsection 11(d), a majority was prepared to hold that
the Constitution contained a more general guarantee of judicial independence, arising
from unwritten constitutional standards recognized and affirmed in the preamble to
the Constitution Act, 1867.110 For present purposes, though, the most interesting
aspect of the decision was the Courts elaboration of what the guarantee of judicial
independence required of governments in relation to judicial salaries.
The requirements laid out by the Court in relation to judicial salaries can be

summarized in five propositions. First, governments are in general entitled to reduce,
freeze or increase provincial court judges salaries, either as part of broader measures
affecting the public service as a whole or as a narrower measure aimed at these judges
alone. But, second, before proceeding with proposals affecting these salaries,
governments must refer the proposals to independent, objective, and effective
commissions for review. And,
these

the recommendations of

third, while

107 [1997] 3 S.C.R. 3, 150 D.L.R. (4th) 577, 217 N.R. 1 [Judges Remuneration Reference cited to

108 Doucet-Boudreau, supra note 10.
109 (2000), 185 N.S.R. (2d) 246, 98 A.C.W.S. (3d) 1006.
110 (U.K.), 30 & 31 Vict., c. 3, reprinted in R.S.C. 1985, App. II, No. 5.

S.C.R.].

responded

to

if

the

recommendations and,

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 541

commissions on the proposals are not binding, the proposals must not proceed until
the governments have
the
recommendations are not followed, the responses are liable to judicial review
according to a standard of simple rationality. A failure of simple rationality would
render the response, and any action taken after it, unconstitutional (i.e., a violation of
subsection 11(d)). Fourth, in any event, these commissions must review the adequacy
of judges salaries (e.g., in relation to the cost of living) regularly (i.e., every three to
five years) to ensure they do not fall below minimum acceptable levels. Finally,
judges are prohibited from negotiating with the executive or legislature over
remuneration, although they may express concerns and make representations on the
issue.

Considerations of legitimacy and competence informed the Courts reasoning on
how much more specific it should be on the shape and powers of the commissions.
Rather than dictating the exact shape and powers, the majority confined itself to
setting out only the basic criteria on the grounds that the questions of detailed
institutional design are better left to the executive and the legislature.111 Nonetheless,
the basic criteria covered a variety of matters, including security of tenure of
members, the independence of the appointment process, and the criteria for review.
Also, the court added that it would be helpful112 if the provincial judiciary were
consulted by governments prior to creating these bodies. Further, in rejecting the
option of making the recommendations of the commissions binding, the majority
relied upon the argument that decisions about the allocation of public resources are
generally within the realm of the legislature, and through it, the executive,113 and that
[t]he expenditure of public funds … is an inherently political matter.114
At the same time, though, the majority defended its decision to expressly require
that judicial salaries not fall below a minimum acceptable level from concerns about
legitimacy and competence. In terms of legitimacy, the majority went to some lengths
to explain that the guarantee of minimum salary levels was not for the benefit of
judges but for the benefit of the public and, more specifically, its confidence in the
independence of the judiciary. In terms of competence, the majority declined to
attempt any identification of the minimum acceptable level in the case at hand,
preferring to do so on a case-by-case basis as the need to do so arose. It then noted
that this Court has in the past accepted its expertise to adjudicate upon rights with a
financial component, such as s. 23 of the Charter115 and went on to cite the decision
in Mahe.
With its decision in this case in relation to the process for reviewing measures
affecting judicial salaries, the Court at once acknowledged competence concerns and

111 Judges Remuneration Reference, supra note 107 at para. 167.
112 Ibid.
113 Ibid. at para. 176.
114 Ibid.
115 Ibid. at para. 195.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

542

responded to them in a way that provided a degree of procedural Charter protection
and allowed for a degree of judicial supervision. When compared to the responses of
injusticiability, deference, and remedial restraint, the response adopted in this case
evinces an effort to explore innovative means of attuning the degree of Charter
protection and judicial supervision to the limits of the institutional competence of
courts. Moreover, this response was used in the context of a claim with implications
for the allocation of scarce fiscal resources and involving the balancing of competing
interests. It can thus be argued that this or similar responses could be explored in the
context of the generic social assistance challenge, which has similar attributes. For
instance, if the validity of governmental action affecting judicial salaries depends
upon conformity to this process, why not demand that the validity of governmental
action affecting the adequacy of social assistance conform to a similar process? The
interest of society in the maintenance of judicial independence is by no means the
same as the interest of the poor to adequate social assistance. However, each may be
sufficiently important in its own way to justify similar responses to similar
competence concerns.
There is another aspect of the decision in Judges Remuneration Reference with

potential significance for anti-poverty cases, namely, the position that the Court was
competent to undertake an inquiry into the minimum level of salary necessary to
ensure judicial independence. This position was justified by reference to a general
level of competence to consider fiscal matters, rather than a narrow expertise in
matters of judicial salary. As a result, this makes available the argument that the Court
also has the competence to adjudicate the issue of the minimum level of social
assistance necessary to ensure security of the person. Yet it was precisely the
competence concerns raised in relation to defining a minimum adequate level of
social assistance in Gosselin that prompted Justice Arbour to argue that a right to
minimum social assistance might be recognized but might also remain largely
unenforceable.

The other main example of innovation is the remedial order made by a lower
court, subsequently approved by the Supreme Court, in Doucet-Boudreau. The case
arose from a claim that the Nova Scotia government had failed to comply with its
obligations under section 23 in respect of minority-language education. Specifically,
it had failed to build five French-language schools that were warranted by the
numbers of the minority community. The government essentially conceded the
obligation to build the schools and the main issue in the case ultimately became the
remedial order of the trial judge, Justice LeBlanc. Justice LeBlanc issued an order
that imposed an obligation on the government to use its best efforts116 to build the
five schools and, further, set deadlines for their completion that ranged from less than
a year from the date of judgment to almost two years. In addition, he imposed

116 Doucet-Boudreau v. Nova Scotia (Department of Education) (2000), 185 N.S.R. (2d) 246, (2000)

98 A.C.W.S. (3d) 1006, [2000] N.S.J. No. 191 (QL) at para. 195.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 543

reporting hearings,117 before himself, which required the government to lodge
affidavits on its progress in compliance. The claimants would also be permitted to
lodge their views. The government objected to the reporting hearings and appealed
their imposition. The Court of Appeal, by majority, upheld the governments
challenge on the grounds that Justice LeBlancs retention of jurisdiction violated the
common law doctrine of functus officio and was not otherwise authorized by any
relevant statute governing judicial powers or the Charter itself. On appeal by the
claimants to the Supreme Court, a majority overturned the Court of Appeals
decision.

In affirming the remedial approach taken by the trial judge, the Supreme Court
majority began by emphasizing the need for a purposive approach in Charter
adjudication. This included ensuring that the remedial sections of the Charter were
interpreted as empowering full, effective and meaningful118 remedies that were
responsive to the purpose of the rights and freedoms at issue. The majority regarded
section 23 as granting a set of rights that were unique in terms of their distinctively
Canadian content, the positive obligations they imposed and their numbers-dependent
collective nature. At the same time, though, the dependency on the size of the
community gave rise to a vulnerability to delay and inaction. Generally speaking,
then, in the case of violations of section 23, timely remedial compliance was crucial.

But the majority acknowledged that there was generally a good record of
governmental compliance with court orders in Canada. Further, they observed that the
courts needed to be sensitive to the limitations of their role and, in particular, ought
not to undertake remedial tasks that were better suited to other persons or bodies. In
other words, courts needed to respect the limits of their institutional competence.
However, since there was no easy formula for determining those limits, the best the
courts could do was to be guided by the nature of the right at issue and the context of
the case.
Against this background, the majority then argued that, as a matter of
straightforward textual interpretation, it could not be denied that the wording of
section 24 allowed courts to remedy governmental inaction, even if that involved a
failure to mobilize fiscal resources. Whether a court ought to do so in any particular
case depended upon whether it was, in the words of the section, appropriate and
just to do so. In turn, according to the Court, that determination depended upon a
consideration of five factors: whether the proposed remedy meaningfully vindicated
the rights and freedoms of the claimants; whether the remedial means were legitimate
within the framework of constitutional democracy; whether the remedy invoked the
powers and functions of courts; whether the remedy was fair to all parties; and
whether the remedial approach recognized the necessity that Charter adjudication
evolve to meet new challenges and circumstances. Applying these factors to the case
at hand, the majority held that the trial judges remedy was appropriate and just.

117 Doucet-Boudreau, supra note 10 at para. 8.
118 Ibid. at para. 25.

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

544

Of particular interest for present purposes is the reasoning in relation to the
second and third factors. The second factor essentially involved maintaining an
appropriate separation of powers and functions between the courts and the other
branches of government. Here the majority emphasized the fact that Justice LeBlancs
remedial order, like remedial orders in cases such as Eldridge and Mahe, left the
details of implementation to the government. The third factor was understood to be all
about the institutional competence of the courts. Here the majority argued that past
judicial practice could serve as a useful guide to the remedial limitations of courts.
The majority then pointed out that in many areas of lawincluding bankruptcy,
receivership, trusts, and family supportcourts were engaged in the use of active and
managerial powers. And while the exercise of these powers presented some
difficulties, courts had been able to overcome those difficulties. The powers exercised
by Justice LeBlanc, being entirely consistent with these past practices, were therefore
within the competence of the courts.
In Doucet-Boudreau, then, the majority essentially elaborated upon the position

in Mahe that the novelty of the rights protected by section 23 required new judicial
responses. But it did so in a way that established a framework applicable to all types
of Charter cases. Competence concerns were an important part of that framework,
but they were not regarded as overly alarming. And, interestingly, the majority was
willing to look beyond the borders of Charter adjudication to determine the limits of
the institutional competence of courts. As a result, it would seem reasonable to argue
that from this point forward there are no excuses for not exploring the potential for
innovations such as the one in Doucet-Boudreau to be used to build competence in
anti-poverty cases in general.119 Indeed, in a lower court decision on the affordability
of special needs health care services (Auton (BCCA)),120 further and similar remedial
innovations have already been explored. Specifically, due to competence concerns
relating to remedial design, the trial judge agreed to hold separate hearings on
liability and remedy. For its part, the Court of Appeal encouraged the parties to come
to their own agreement over processes for designing, implementing, and monitoring a
satisfactory treatment program. But the court maintained supervisory jurisdiction and
allowed the parties to return to the court to resolve any disagreements.
As such, these further aspects of the fourth general development in the
framework for recognizing and treating competence concerns in Charter adjudication
also assist in countering the anti-poverty incompetence argument. Specifically, if a
claim such as the generic social assistance challenge were ever upheld by a court,
then a government might argue that there remained a range of policy options for

119 For approving assessments of the decision in Doucet-Boudreau, supra note 10, which also
acknowledge the relevance of competence concerns, see Marilyn L. Pilkington, Enforcing the
Charter: The Supervisory Role of Superior Courts and the Responsibility of Legislatures for Remedial
Systems (2004) 25 Sup. Ct. L. Rev. (2d Series) 77 and Kent Roach, Principled Remedial Discretion
Under the Charter (2004) 25 Sup. Ct. L. Rev. (2d Series) 101.

120 Supra note 68. Although the SCC has since upheld an appeal from this decision, the innovation

described here was not at issue on appeal.

2006] D. WISEMAN COMPETENCE CONCERNS IN CHARTER ADJUDICATION 545

realizing security of the person for the poor and that the court lacked the competence
to choose and design the most appropriate remedy. In response, a court might exercise
remedial restraint by opting merely for a declaratory remedy and relinquishing
jurisdiction. However, following the lead of Doucet-Boudreau and Auton, a court
might seek to better equip itself for making a more detailed remedial order by
ordering a further hearing on the appropriate remedy. Alternatively, it might
encourage the parties to negotiate a process for determining the most appropriate
remedy and maintain jurisdiction to address any conflicts that might arise.

Conclusion
Any initial hope that the entrenchment of the Charter would provide an
additional means for prompting Canadian governments to take stronger action against
poverty is far from being realized. In part this is because of the role that competence
concerns have played in anti-poverty Charter adjudication. Generally speaking, the
tendency has been for competence concerns to limit the extent of anti-poverty
protection offered by the Charter. Indeed, through injusticiability holdings, it is often
the case that no protection at all is offered. The accumulated case law in which
competence concerns have worked to the detriment of anti-poverty Charter cases
appears to be underwritten by an anti-poverty incompetence argument. This
argument, in turn, is based upon the foundational framework for the recognition and
treatment of competence concerns established by the Supreme Court. That
foundational framework holds that court competence is challenged by the need to
gather and evaluate a sufficient quantity and quality of evidence and arguments on
both normative and empirical issues, and by the need to assess the balances struck in
governmental decisions. Since all three of these needs typically arise in anti-poverty
cases, so too do competence concerns. Moreover, anti-poverty cases typically possess
the more specific attributes regarded as raising competence concerns. That is, anti-
poverty cases involve the adjudication of group-mediating social policy and of
conflicting social science evidence. Further, they involve the need to balance
competing interests (especially in relation to scarce fiscal resources) and the interests
of vulnerable groups. Consequently, according to the anti-poverty incompetence
argument, the degree of Charter protection offered to anti-poverty claims ought to be
limited, either totally (through injusticability) or significantly (through deference).

The anti-poverty incompetence argument is, however, afflicted with some of the
problems that beset the foundational framework itself. Further, it is out of step with
countervailing aspects of subsequent general developments in the application of that
framework. In other words, the very same body of case law that allows, and gives
expression to, the anti-poverty incompetence argument also provides the means for
countering it (at least as it is expressed in accumulated Charter cases). The counter-
argument has three main components. First, the Supreme Court itself has rightly
questioned some of its own earlier conceptions of the attributes that signal
incompetence, particularly the social policy/criminal justice distinction. Second, in
other types of Charter cases, where competence challenges similar to those that can

[Vol. 51

MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL

546

arise in anti-poverty cases have arisen, the courts have been willing to respond to
those challenges in ways that provide more fulsome Charter protection. In particular,
courts have been willing to undertake competence-building measures that could be
applied in anti-poverty cases. Third, in a number of cases involving the interests of
vulnerable groupsinterests that are at stake in anti-poverty casesthe Supreme
Court has recognized that competence concerns can be counterbalanced by normative
concerns. This counterbalance has been relied upon to respond to competence
challenges in ways that provide as much Charter protection as possible. Similar
reliance could be placed upon it in anti-poverty cases.
As such, my counter-argument accepts, for present purposes at least, that the
forms of adjudication can cause competence challenges to arise in anti-poverty cases.
However, the anti-poverty incompetence argument both overstates the challenges and
overreacts to those challenges, especially as compared to the recognition and
treatment of competence challenges arising in other types of cases. In fact, on closer
analysis, it is possible to argue that what is most coherent in accumulated competence
decisions pulls in the opposite direction to the anti-poverty incompetence argument.
In other words, it is possible to argue, at least within the confines of accumulated case
law, that courts ought to work much harder to respond to competence concerns in
ways that limit the force of anti-poverty Charter obligations as little as possible.
Better yet, courts ought to follow their own lead in exploring the competence-
building means by which they can engage and overcome the competence concerns
that may arise in anti-poverty Charter cases.